From 3fb45ff1762867cbeb53e31cd4492dda5cf8dbb7 Mon Sep 17 00:00:00 2001 From: Blargian Date: Tue, 14 May 2024 19:17:56 +0200 Subject: [PATCH 001/260] Add setting and implementation --- src/Core/Settings.h | 1 + src/Formats/FormatFactory.cpp | 1 + src/Formats/FormatSettings.h | 1 + src/Formats/SchemaInferenceUtils.cpp | 44 ++++++++++++++++++- .../0_stateless/03150_infer_type_variant.sql | 4 ++ 5 files changed, 50 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03150_infer_type_variant.sql diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 4a0de354a03..0c0614550e5 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -1066,6 +1066,7 @@ class IColumn; M(Bool, input_format_json_defaults_for_missing_elements_in_named_tuple, true, "Insert default value in named tuple element if it's missing in json object", 0) \ M(Bool, input_format_json_throw_on_bad_escape_sequence, true, "Throw an exception if JSON string contains bad escape sequence in JSON input formats. If disabled, bad escape sequences will remain as is in the data", 0) \ M(Bool, input_format_json_ignore_unnecessary_fields, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields", 0) \ + M(Bool, input_format_json_infer_variant_from_multitype_array, false, "Try to infer variant type rather than tuple when column/array has multiple", 0) \ M(Bool, input_format_try_infer_integers, true, "Try to infer integers instead of floats while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_dates, true, "Try to infer dates from string fields while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_datetimes, true, "Try to infer datetimes from string fields while schema inference in text formats", 0) \ diff --git a/src/Formats/FormatFactory.cpp b/src/Formats/FormatFactory.cpp index 43ccee173f0..e027c693094 100644 --- a/src/Formats/FormatFactory.cpp +++ b/src/Formats/FormatFactory.cpp @@ -137,6 +137,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.json.read_arrays_as_strings = settings.input_format_json_read_arrays_as_strings; format_settings.json.try_infer_numbers_from_strings = settings.input_format_json_try_infer_numbers_from_strings; format_settings.json.infer_incomplete_types_as_strings = settings.input_format_json_infer_incomplete_types_as_strings; + format_settings.json.infer_variant_from_multitype_array = settings.input_format_json_infer_variant_from_multitype_array; format_settings.json.validate_types_from_metadata = settings.input_format_json_validate_types_from_metadata; format_settings.json.validate_utf8 = settings.output_format_json_validate_utf8; format_settings.json_object_each_row.column_for_object_name = settings.format_json_object_each_row_column_for_object_name; diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index d5fedf99adb..6cac41bd63e 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -223,6 +223,7 @@ struct FormatSettings bool compact_allow_variable_number_of_columns = false; bool try_infer_objects_as_tuples = false; bool infer_incomplete_types_as_strings = true; + bool infer_variant_from_multitype_array = false; bool throw_on_bad_escape_sequence = true; bool ignore_unnecessary_fields = true; } json{}; diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 02c0aa6dd77..fc1a26f9b2f 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -307,6 +308,22 @@ namespace type_indexes.erase(TypeIndex::UInt64); } + /// if setting input_format_json_infer_variant_from_multitype_array is true + /// and nested types are not equal then we convert to type variant. + void transformVariant(DataTypes & data_types, TypeIndexesSet & type_indexes) + { + auto variant_type = std::make_shared(data_types); + /// replace separate types with a single variant type + data_types.clear(); + type_indexes.clear(); + data_types.push_back(variant_type); + type_indexes.insert(TypeIndex::Variant); + + // push it back again + data_types.push_back(variant_type); + type_indexes.insert(TypeIndex::Variant); + } + /// If we have only Date and DateTime types, convert Date to DateTime, /// otherwise, convert all Date and DateTime to String. void transformDatesAndDateTimes(DataTypes & data_types, TypeIndexesSet & type_indexes) @@ -649,6 +666,12 @@ namespace /// Check settings specific for JSON formats. + if (settings.json.infer_variant_from_multitype_array) + { + transformVariant(data_types, type_indexes); + return; + } + /// Convert numbers inferred from strings back to strings if needed. if (settings.json.try_infer_numbers_from_strings || settings.json.read_numbers_as_strings) transformJSONNumbersBackToString(data_types, settings, type_indexes, json_info); @@ -677,6 +700,12 @@ namespace if constexpr (!is_json) return; + if (settings.json.infer_variant_from_multitype_array) + { + transformVariant(data_types, type_indexes); + return; + } + /// Convert JSON tuples with same nested types to arrays. transformTuplesWithEqualNestedTypesToArrays(data_types, type_indexes); @@ -822,7 +851,6 @@ namespace if (checkIfTypesAreEqual(nested_types_copy)) return std::make_shared(nested_types_copy.back()); - return std::make_shared(nested_types); } else @@ -1482,6 +1510,20 @@ DataTypePtr makeNullableRecursively(DataTypePtr type) return nested_type ? std::make_shared(nested_type) : nullptr; } + if (which.isVariant()) + { + const auto * variant_type = assert_cast(type.get()); + DataTypes nested_types; + for (const auto & nested_type: variant_type->getVariants()) + { + /// unlike tuple or array, here we do not want to make any of the variants nullable + /// so we do not call makeNullableRecursively + nested_types.push_back(nested_type); + } + + return std::make_shared(nested_types); + } + if (which.isTuple()) { const auto * tuple_type = assert_cast(type.get()); diff --git a/tests/queries/0_stateless/03150_infer_type_variant.sql b/tests/queries/0_stateless/03150_infer_type_variant.sql new file mode 100644 index 00000000000..ac544d04f6e --- /dev/null +++ b/tests/queries/0_stateless/03150_infer_type_variant.sql @@ -0,0 +1,4 @@ +SET input_format_json_infer_variant_from_multitype_array=1; +SELECT arr, toTypeName(arr) FROM format('JSONEachRow', '{"arr" : [1, "Hello", {"a" : 32}]}'); +SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : 42}, {"x" : "Hello"}'); +SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : [1, 2, 3]}, {"x" : {"a" : 42}}'); From 467366af990215e11b4b0309b90b3e6d9ebca5fd Mon Sep 17 00:00:00 2001 From: Blargian Date: Tue, 14 May 2024 21:55:56 +0200 Subject: [PATCH 002/260] Fix unimplemented serialization error and update reference file --- src/Formats/SchemaInferenceUtils.cpp | 13 ++++++++--- .../03150_infer_type_variant.reference | 22 +++++++++++++++++++ 2 files changed, 32 insertions(+), 3 deletions(-) create mode 100644 tests/queries/0_stateless/03150_infer_type_variant.reference diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index fc1a26f9b2f..d0d29892dec 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -319,7 +319,7 @@ namespace data_types.push_back(variant_type); type_indexes.insert(TypeIndex::Variant); - // push it back again + // make the second type variant as well data_types.push_back(variant_type); type_indexes.insert(TypeIndex::Variant); } @@ -669,7 +669,6 @@ namespace if (settings.json.infer_variant_from_multitype_array) { transformVariant(data_types, type_indexes); - return; } /// Convert numbers inferred from strings back to strings if needed. @@ -703,7 +702,6 @@ namespace if (settings.json.infer_variant_from_multitype_array) { transformVariant(data_types, type_indexes); - return; } /// Convert JSON tuples with same nested types to arrays. @@ -1440,6 +1438,15 @@ void transformFinalInferredJSONTypeIfNeededImpl(DataTypePtr & data_type, const F return; } + + if (const auto * variant_type = typeid_cast(data_type.get())) + { + auto nested_types = variant_type->getVariants(); + for (auto & nested_type : nested_types) + transformFinalInferredJSONTypeIfNeededImpl(nested_type, settings, json_info, remain_nothing_types); + data_type = std::make_shared(nested_types); + return; + } } void transformFinalInferredJSONTypeIfNeeded(DataTypePtr & data_type, const FormatSettings & settings, JSONInferenceInfo * json_info) diff --git a/tests/queries/0_stateless/03150_infer_type_variant.reference b/tests/queries/0_stateless/03150_infer_type_variant.reference new file mode 100644 index 00000000000..ffb4209eadb --- /dev/null +++ b/tests/queries/0_stateless/03150_infer_type_variant.reference @@ -0,0 +1,22 @@ + ┏━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ arr ┃ toTypeName(arr) ┃ + ┡━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ [1,'Hello',(32)] │ Array(Variant(Int64, String, Tuple(…│ + │ │… a Int64))) │ + └──────────────────┴─────────────────────────────────────┘ + ┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ x ┃ toTypeName(x) ┃ + ┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ 42 │ Variant(Int64, String) │ + ├───────┼────────────────────────┤ +2. │ Hello │ Variant(Int64, String) │ + └───────┴────────────────────────┘ + ┏━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ x ┃ toTypeName(x) ┃ + ┡━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ [1,2,3] │ Variant(Array(Int64), Tuple(…│ + │ │… a Int64)) │ + ├─────────┼──────────────────────────────┤ +2. │ (42) │ Variant(Array(Int64), Tuple(…│ + │ │… a Int64)) │ + └─────────┴──────────────────────────────┘ From 4066c6bc548979703f45ba264437f5966c403d6a Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Wed, 15 May 2024 02:13:53 +0000 Subject: [PATCH 003/260] Update setting name --- src/Core/Settings.h | 2 +- src/Formats/FormatFactory.cpp | 2 +- src/Formats/FormatSettings.h | 2 +- src/Formats/SchemaInferenceUtils.cpp | 6 +++--- tests/queries/0_stateless/03150_infer_type_variant.sql | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 0c0614550e5..ffc337b674f 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -1066,7 +1066,7 @@ class IColumn; M(Bool, input_format_json_defaults_for_missing_elements_in_named_tuple, true, "Insert default value in named tuple element if it's missing in json object", 0) \ M(Bool, input_format_json_throw_on_bad_escape_sequence, true, "Throw an exception if JSON string contains bad escape sequence in JSON input formats. If disabled, bad escape sequences will remain as is in the data", 0) \ M(Bool, input_format_json_ignore_unnecessary_fields, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields", 0) \ - M(Bool, input_format_json_infer_variant_from_multitype_array, false, "Try to infer variant type rather than tuple when column/array has multiple", 0) \ + M(Bool, input_format_json_infer_variant_from_multi_type_array, false, "Try to infer variant type rather than tuple when column/array has multiple", 0) \ M(Bool, input_format_try_infer_integers, true, "Try to infer integers instead of floats while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_dates, true, "Try to infer dates from string fields while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_datetimes, true, "Try to infer datetimes from string fields while schema inference in text formats", 0) \ diff --git a/src/Formats/FormatFactory.cpp b/src/Formats/FormatFactory.cpp index e027c693094..792ac08a5df 100644 --- a/src/Formats/FormatFactory.cpp +++ b/src/Formats/FormatFactory.cpp @@ -137,7 +137,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.json.read_arrays_as_strings = settings.input_format_json_read_arrays_as_strings; format_settings.json.try_infer_numbers_from_strings = settings.input_format_json_try_infer_numbers_from_strings; format_settings.json.infer_incomplete_types_as_strings = settings.input_format_json_infer_incomplete_types_as_strings; - format_settings.json.infer_variant_from_multitype_array = settings.input_format_json_infer_variant_from_multitype_array; + format_settings.json.infer_variant_from_multi_type_array = settings.input_format_json_infer_variant_from_multi_type_array; format_settings.json.validate_types_from_metadata = settings.input_format_json_validate_types_from_metadata; format_settings.json.validate_utf8 = settings.output_format_json_validate_utf8; format_settings.json_object_each_row.column_for_object_name = settings.format_json_object_each_row_column_for_object_name; diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index 6cac41bd63e..d2c75872326 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -223,7 +223,7 @@ struct FormatSettings bool compact_allow_variable_number_of_columns = false; bool try_infer_objects_as_tuples = false; bool infer_incomplete_types_as_strings = true; - bool infer_variant_from_multitype_array = false; + bool infer_variant_from_multi_type_array = false; bool throw_on_bad_escape_sequence = true; bool ignore_unnecessary_fields = true; } json{}; diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index d0d29892dec..f693916c584 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -308,7 +308,7 @@ namespace type_indexes.erase(TypeIndex::UInt64); } - /// if setting input_format_json_infer_variant_from_multitype_array is true + /// if setting input_format_json_infer_variant_from_multi_type_array is true /// and nested types are not equal then we convert to type variant. void transformVariant(DataTypes & data_types, TypeIndexesSet & type_indexes) { @@ -666,7 +666,7 @@ namespace /// Check settings specific for JSON formats. - if (settings.json.infer_variant_from_multitype_array) + if (settings.json.infer_variant_from_multi_type_array) { transformVariant(data_types, type_indexes); } @@ -699,7 +699,7 @@ namespace if constexpr (!is_json) return; - if (settings.json.infer_variant_from_multitype_array) + if (settings.json.infer_variant_from_multi_type_array) { transformVariant(data_types, type_indexes); } diff --git a/tests/queries/0_stateless/03150_infer_type_variant.sql b/tests/queries/0_stateless/03150_infer_type_variant.sql index ac544d04f6e..2ea849248f7 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.sql +++ b/tests/queries/0_stateless/03150_infer_type_variant.sql @@ -1,4 +1,4 @@ -SET input_format_json_infer_variant_from_multitype_array=1; +SET input_format_json_infer_variant_from_multi_type_array=1; SELECT arr, toTypeName(arr) FROM format('JSONEachRow', '{"arr" : [1, "Hello", {"a" : 32}]}'); SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : 42}, {"x" : "Hello"}'); SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : [1, 2, 3]}, {"x" : {"a" : 42}}'); From 2762cf86d2ba3f5c1ac86040b6ef484feb40837b Mon Sep 17 00:00:00 2001 From: Blargian Date: Wed, 15 May 2024 14:19:17 +0200 Subject: [PATCH 004/260] fix test file --- tests/queries/0_stateless/03150_infer_type_variant.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03150_infer_type_variant.sql b/tests/queries/0_stateless/03150_infer_type_variant.sql index 2ea849248f7..ac544d04f6e 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.sql +++ b/tests/queries/0_stateless/03150_infer_type_variant.sql @@ -1,4 +1,4 @@ -SET input_format_json_infer_variant_from_multi_type_array=1; +SET input_format_json_infer_variant_from_multitype_array=1; SELECT arr, toTypeName(arr) FROM format('JSONEachRow', '{"arr" : [1, "Hello", {"a" : 32}]}'); SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : 42}, {"x" : "Hello"}'); SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : [1, 2, 3]}, {"x" : {"a" : 42}}'); From 4800aa6a6cdf5a8431fcc3d6fd96672590da0fff Mon Sep 17 00:00:00 2001 From: Blargian Date: Wed, 15 May 2024 14:21:52 +0200 Subject: [PATCH 005/260] rename setting in test file --- tests/queries/0_stateless/03150_infer_type_variant.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03150_infer_type_variant.sql b/tests/queries/0_stateless/03150_infer_type_variant.sql index ac544d04f6e..2ea849248f7 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.sql +++ b/tests/queries/0_stateless/03150_infer_type_variant.sql @@ -1,4 +1,4 @@ -SET input_format_json_infer_variant_from_multitype_array=1; +SET input_format_json_infer_variant_from_multi_type_array=1; SELECT arr, toTypeName(arr) FROM format('JSONEachRow', '{"arr" : [1, "Hello", {"a" : 32}]}'); SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : 42}, {"x" : "Hello"}'); SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : [1, 2, 3]}, {"x" : {"a" : 42}}'); From fbf34519a5e72bb03c57ee6bc0feea1adddcb309 Mon Sep 17 00:00:00 2001 From: Blargian Date: Wed, 15 May 2024 16:03:23 +0200 Subject: [PATCH 006/260] Add setting to SettingsChangesHistory.h --- src/Core/SettingsChangesHistory.h | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h index ece48620618..0665d1d6ca6 100644 --- a/src/Core/SettingsChangesHistory.h +++ b/src/Core/SettingsChangesHistory.h @@ -93,6 +93,7 @@ static std::map sett {"prefer_external_sort_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Prefer maximum block bytes for external sort, reduce the memory usage during merging."}, {"input_format_force_null_for_omitted_fields", false, false, "Disable type-defaults for omitted fields when needed"}, {"output_format_pretty_preserve_border_for_multiline_string", 1, 1, "Applies better rendering for multiline strings."}, + {"input_format_json_infer_variant_from_multi_type_array", 0, 0, "Allows inference of variant type if columns/arrays have multiple types."}, }}, {"24.4", {{"input_format_json_throw_on_bad_escape_sequence", true, true, "Allow to save JSON strings with bad escape sequences"}, {"max_parsing_threads", 0, 0, "Add a separate setting to control number of threads in parallel parsing from files"}, From 777e0b313a48975fa51d645fb09a4f6ebfac1d1e Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Wed, 15 May 2024 18:51:29 +0200 Subject: [PATCH 007/260] Update 03150_infer_type_variant.sql Fix failing test --- tests/queries/0_stateless/03150_infer_type_variant.sql | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/03150_infer_type_variant.sql b/tests/queries/0_stateless/03150_infer_type_variant.sql index 2ea849248f7..3253ddfe179 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.sql +++ b/tests/queries/0_stateless/03150_infer_type_variant.sql @@ -1,4 +1,4 @@ SET input_format_json_infer_variant_from_multi_type_array=1; -SELECT arr, toTypeName(arr) FROM format('JSONEachRow', '{"arr" : [1, "Hello", {"a" : 32}]}'); -SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : 42}, {"x" : "Hello"}'); -SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : [1, 2, 3]}, {"x" : {"a" : 42}}'); +SELECT arr, toTypeName(arr) FROM format('JSONEachRow', '{"arr" : [1, "Hello", {"a" : 32}]}') FORMAT Pretty; +SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : 42}, {"x" : "Hello"}') FORMAT Pretty; +SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : [1, 2, 3]}, {"x" : {"a" : 42}}') FORMAT Pretty; From dd8d5c46c4678f570b9357dbdd912bcd6f4a267e Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Wed, 15 May 2024 18:55:34 +0200 Subject: [PATCH 008/260] Fix style --- src/Formats/SchemaInferenceUtils.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index f693916c584..2cbb680af97 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -319,7 +319,7 @@ namespace data_types.push_back(variant_type); type_indexes.insert(TypeIndex::Variant); - // make the second type variant as well + /// make the second type variant as well data_types.push_back(variant_type); type_indexes.insert(TypeIndex::Variant); } From 04800f596c4471d10e15c40a533c539c6b549b06 Mon Sep 17 00:00:00 2001 From: Blargian Date: Fri, 24 May 2024 21:20:20 +0200 Subject: [PATCH 009/260] Incorporate review changes --- src/Core/Settings.h | 2 +- src/Core/SettingsChangesHistory.h | 2 +- src/Formats/FormatFactory.cpp | 2 +- src/Formats/FormatSettings.h | 2 +- src/Formats/SchemaInferenceUtils.cpp | 60 +++++++++++++------ .../03150_infer_type_variant.reference | 29 +++++---- .../0_stateless/03150_infer_type_variant.sql | 3 +- 7 files changed, 67 insertions(+), 33 deletions(-) diff --git a/src/Core/Settings.h b/src/Core/Settings.h index ffc337b674f..be7564794e9 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -1066,7 +1066,7 @@ class IColumn; M(Bool, input_format_json_defaults_for_missing_elements_in_named_tuple, true, "Insert default value in named tuple element if it's missing in json object", 0) \ M(Bool, input_format_json_throw_on_bad_escape_sequence, true, "Throw an exception if JSON string contains bad escape sequence in JSON input formats. If disabled, bad escape sequences will remain as is in the data", 0) \ M(Bool, input_format_json_ignore_unnecessary_fields, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields", 0) \ - M(Bool, input_format_json_infer_variant_from_multi_type_array, false, "Try to infer variant type rather than tuple when column/array has multiple", 0) \ + M(Bool, input_format_try_infer_variants, false, "Try to infer the Variant type in text formats when there is more than one possible type for column/array elements", 0) \ M(Bool, input_format_try_infer_integers, true, "Try to infer integers instead of floats while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_dates, true, "Try to infer dates from string fields while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_datetimes, true, "Try to infer datetimes from string fields while schema inference in text formats", 0) \ diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h index 0665d1d6ca6..75fcb538b2b 100644 --- a/src/Core/SettingsChangesHistory.h +++ b/src/Core/SettingsChangesHistory.h @@ -93,7 +93,7 @@ static std::map sett {"prefer_external_sort_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Prefer maximum block bytes for external sort, reduce the memory usage during merging."}, {"input_format_force_null_for_omitted_fields", false, false, "Disable type-defaults for omitted fields when needed"}, {"output_format_pretty_preserve_border_for_multiline_string", 1, 1, "Applies better rendering for multiline strings."}, - {"input_format_json_infer_variant_from_multi_type_array", 0, 0, "Allows inference of variant type if columns/arrays have multiple types."}, + {"input_format_try_infer_variants", 0, 0, "Try to infer Variant type in text formats when there is more than one possible type for column/array elements"}, }}, {"24.4", {{"input_format_json_throw_on_bad_escape_sequence", true, true, "Allow to save JSON strings with bad escape sequences"}, {"max_parsing_threads", 0, 0, "Add a separate setting to control number of threads in parallel parsing from files"}, diff --git a/src/Formats/FormatFactory.cpp b/src/Formats/FormatFactory.cpp index 792ac08a5df..2854802453e 100644 --- a/src/Formats/FormatFactory.cpp +++ b/src/Formats/FormatFactory.cpp @@ -137,7 +137,6 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.json.read_arrays_as_strings = settings.input_format_json_read_arrays_as_strings; format_settings.json.try_infer_numbers_from_strings = settings.input_format_json_try_infer_numbers_from_strings; format_settings.json.infer_incomplete_types_as_strings = settings.input_format_json_infer_incomplete_types_as_strings; - format_settings.json.infer_variant_from_multi_type_array = settings.input_format_json_infer_variant_from_multi_type_array; format_settings.json.validate_types_from_metadata = settings.input_format_json_validate_types_from_metadata; format_settings.json.validate_utf8 = settings.output_format_json_validate_utf8; format_settings.json_object_each_row.column_for_object_name = settings.format_json_object_each_row_column_for_object_name; @@ -266,6 +265,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.max_parser_depth = context->getSettingsRef().max_parser_depth; format_settings.client_protocol_version = context->getClientProtocolVersion(); format_settings.date_time_overflow_behavior = settings.date_time_overflow_behavior; + format_settings.try_infer_variant = settings.input_format_try_infer_variants; /// Validate avro_schema_registry_url with RemoteHostFilter when non-empty and in Server context if (format_settings.schema.is_server) diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index 927a7e691d8..a2a9e75bd44 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -36,6 +36,7 @@ struct FormatSettings bool decimal_trailing_zeros = false; bool defaults_for_omitted_fields = true; bool is_writing_to_terminal = false; + bool try_infer_variant = false; bool seekable_read = true; UInt64 max_rows_to_read_for_schema_inference = 25000; @@ -223,7 +224,6 @@ struct FormatSettings bool compact_allow_variable_number_of_columns = false; bool try_infer_objects_as_tuples = false; bool infer_incomplete_types_as_strings = true; - bool infer_variant_from_multi_type_array = false; bool throw_on_bad_escape_sequence = true; bool ignore_unnecessary_fields = true; } json{}; diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 2cbb680af97..298127cad68 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -239,6 +239,16 @@ namespace return true; } + bool checkIfTypesContainVariant(const DataTypes & types) + { + for (size_t i = 0; i < types.size(); ++i) + { + if (isVariant(types[i])) + return true; + } + return false; + } + void updateTypeIndexes(DataTypes & data_types, TypeIndexesSet & type_indexes) { type_indexes.clear(); @@ -308,20 +318,31 @@ namespace type_indexes.erase(TypeIndex::UInt64); } - /// if setting input_format_json_infer_variant_from_multi_type_array is true + /// if setting try_infer_variant is true /// and nested types are not equal then we convert to type variant. void transformVariant(DataTypes & data_types, TypeIndexesSet & type_indexes) { + auto typesAreEqual = checkIfTypesAreEqual(data_types); + auto typesContainVariant = checkIfTypesContainVariant(data_types); + if (typesAreEqual || typesContainVariant) + return; + + DataTypes new_data_types; + TypeIndexesSet new_type_indexes; + auto variant_type = std::make_shared(data_types); - /// replace separate types with a single variant type + size_t i = 0; + while (i != data_types.size()) + { + new_data_types.push_back(variant_type); + new_type_indexes.insert(TypeIndex::Variant); + i++; + } + data_types.clear(); type_indexes.clear(); - data_types.push_back(variant_type); - type_indexes.insert(TypeIndex::Variant); - - /// make the second type variant as well - data_types.push_back(variant_type); - type_indexes.insert(TypeIndex::Variant); + data_types = new_data_types; + type_indexes = new_type_indexes; } /// If we have only Date and DateTime types, convert Date to DateTime, @@ -661,16 +682,14 @@ namespace if (settings.try_infer_dates || settings.try_infer_datetimes) transformDatesAndDateTimes(data_types, type_indexes); + if (settings.try_infer_variant) + transformVariant(data_types, type_indexes); + if constexpr (!is_json) return; /// Check settings specific for JSON formats. - if (settings.json.infer_variant_from_multi_type_array) - { - transformVariant(data_types, type_indexes); - } - /// Convert numbers inferred from strings back to strings if needed. if (settings.json.try_infer_numbers_from_strings || settings.json.read_numbers_as_strings) transformJSONNumbersBackToString(data_types, settings, type_indexes, json_info); @@ -685,6 +704,10 @@ namespace if (settings.json.try_infer_objects_as_tuples) mergeJSONPaths(data_types, type_indexes, settings, json_info); + + if (settings.try_infer_variant) + transformVariant(data_types, type_indexes); + }; auto transform_complex_types = [&](DataTypes & data_types, TypeIndexesSet & type_indexes) @@ -696,14 +719,12 @@ namespace /// If there is at least one non Nothing type, change all Nothing types to it. transformNothingComplexTypes(data_types, type_indexes); + if (settings.try_infer_variant) + transformVariant(data_types, type_indexes); + if constexpr (!is_json) return; - if (settings.json.infer_variant_from_multi_type_array) - { - transformVariant(data_types, type_indexes); - } - /// Convert JSON tuples with same nested types to arrays. transformTuplesWithEqualNestedTypesToArrays(data_types, type_indexes); @@ -715,6 +736,9 @@ namespace if (json_info && json_info->allow_merging_named_tuples) mergeNamedTuples(data_types, type_indexes, settings, json_info); + + if (settings.try_infer_variant) + transformVariant(data_types, type_indexes); }; transformTypesRecursively(types, transform_simple_types, transform_complex_types); diff --git a/tests/queries/0_stateless/03150_infer_type_variant.reference b/tests/queries/0_stateless/03150_infer_type_variant.reference index ffb4209eadb..f3f53057845 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.reference +++ b/tests/queries/0_stateless/03150_infer_type_variant.reference @@ -2,7 +2,7 @@ ┃ arr ┃ toTypeName(arr) ┃ ┡━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ 1. │ [1,'Hello',(32)] │ Array(Variant(Int64, String, Tuple(…│ - │ │… a Int64))) │ + │ │… a Nullable(Int64)))) │ └──────────────────┴─────────────────────────────────────┘ ┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┓ ┃ x ┃ toTypeName(x) ┃ @@ -11,12 +11,21 @@ ├───────┼────────────────────────┤ 2. │ Hello │ Variant(Int64, String) │ └───────┴────────────────────────┘ - ┏━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ - ┃ x ┃ toTypeName(x) ┃ - ┡━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ -1. │ [1,2,3] │ Variant(Array(Int64), Tuple(…│ - │ │… a Int64)) │ - ├─────────┼──────────────────────────────┤ -2. │ (42) │ Variant(Array(Int64), Tuple(…│ - │ │… a Int64)) │ - └─────────┴──────────────────────────────┘ + ┏━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ x ┃ toTypeName(x) ┃ + ┡━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ [1,2,3] │ Variant(Array(Nullable(Int64)), Tuple(…│ + │ │… a Nullable(Int64))) │ + ├─────────┼────────────────────────────────────────┤ +2. │ (42) │ Variant(Array(Nullable(Int64)), Tuple(…│ + │ │… a Nullable(Int64))) │ + └─────────┴────────────────────────────────────────┘ + ┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┓ + ┃ c1 ┃ toTypeName(c1) ┃ c2 ┃ toTypeName(c2) ┃ + ┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━┩ +1. │ 1 │ Nullable(Int64) │ Hello World! │ Nullable(String) │ + ├────┼─────────────────┼──────────────┼──────────────────┤ +2. │ 2 │ Nullable(Int64) │ [1,2,3] │ Nullable(String) │ + ├────┼─────────────────┼──────────────┼──────────────────┤ +3. │ 3 │ Nullable(Int64) │ 2020-01-01 │ Nullable(String) │ + └────┴─────────────────┴──────────────┴──────────────────┘ diff --git a/tests/queries/0_stateless/03150_infer_type_variant.sql b/tests/queries/0_stateless/03150_infer_type_variant.sql index 3253ddfe179..45126ccd471 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.sql +++ b/tests/queries/0_stateless/03150_infer_type_variant.sql @@ -1,4 +1,5 @@ -SET input_format_json_infer_variant_from_multi_type_array=1; +SET input_format_try_infer_variants=1; SELECT arr, toTypeName(arr) FROM format('JSONEachRow', '{"arr" : [1, "Hello", {"a" : 32}]}') FORMAT Pretty; SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : 42}, {"x" : "Hello"}') FORMAT Pretty; SELECT x, toTypeName(x) FROM format('JSONEachRow', '{"x" : [1, 2, 3]}, {"x" : {"a" : 42}}') FORMAT Pretty; +SELECT c1, toTypeName(c1), c2, toTypeName(c2) FROM format('CSV', '1,Hello World!\n2,"[1,2,3]"\n3,"2020-01-01"\n') FORMAT Pretty; \ No newline at end of file From 655262d1a1b21d85f4fbe284e0835065bcca379b Mon Sep 17 00:00:00 2001 From: Blargian Date: Wed, 5 Jun 2024 14:34:17 +0200 Subject: [PATCH 010/260] Fix issue with nullables --- src/Formats/SchemaInferenceUtils.cpp | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 298127cad68..43120cb7b22 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -1547,11 +1547,13 @@ DataTypePtr makeNullableRecursively(DataTypePtr type) DataTypes nested_types; for (const auto & nested_type: variant_type->getVariants()) { - /// unlike tuple or array, here we do not want to make any of the variants nullable - /// so we do not call makeNullableRecursively - nested_types.push_back(nested_type); + auto is_low_cardinality = nested_type->lowCardinality(); + auto has_sub_types = nested_type->haveSubtypes(); + if (!is_low_cardinality && has_sub_types) + nested_types.push_back(makeNullableRecursively(nested_type)); + else + nested_types.push_back(nested_type); } - return std::make_shared(nested_types); } From 418fc7f4438abd25eae4928f36ff0c3fef2395f8 Mon Sep 17 00:00:00 2001 From: Blargian Date: Thu, 6 Jun 2024 10:02:19 +0200 Subject: [PATCH 011/260] Fix incorrect inference for other formats --- src/Formats/SchemaInferenceUtils.cpp | 29 +++++++++-- .../03150_infer_type_variant.reference | 48 +++++++++---------- 2 files changed, 49 insertions(+), 28 deletions(-) diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 43120cb7b22..a8b5d4343f5 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -318,19 +318,40 @@ namespace type_indexes.erase(TypeIndex::UInt64); } - /// if setting try_infer_variant is true - /// and nested types are not equal then we convert to type variant. + /// if setting 'try_infer_variant' is true then we convert to type variant. void transformVariant(DataTypes & data_types, TypeIndexesSet & type_indexes) { auto typesAreEqual = checkIfTypesAreEqual(data_types); auto typesContainVariant = checkIfTypesContainVariant(data_types); - if (typesAreEqual || typesContainVariant) + if (typesAreEqual) return; DataTypes new_data_types; TypeIndexesSet new_type_indexes; + std::shared_ptr variant_type; + + /// extract the nested types of variant and make a new variant with the nested types and the other type. + /// eg. Type 1: variant, Type 2: Date -> variant. + if (typesContainVariant) + { + DataTypes extracted_types; + for (size_t i=0; i(data_types[i].get())) + extracted_types = variant->getVariants(); + } + else + extracted_types.push_back(data_types[i]); + } + variant_type = std::make_shared(extracted_types); + } + else + { + variant_type = std::make_shared(data_types); + } - auto variant_type = std::make_shared(data_types); size_t i = 0; while (i != data_types.size()) { diff --git a/tests/queries/0_stateless/03150_infer_type_variant.reference b/tests/queries/0_stateless/03150_infer_type_variant.reference index f3f53057845..a5f56cb3618 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.reference +++ b/tests/queries/0_stateless/03150_infer_type_variant.reference @@ -1,9 +1,9 @@ - ┏━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ - ┃ arr ┃ toTypeName(arr) ┃ - ┡━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ -1. │ [1,'Hello',(32)] │ Array(Variant(Int64, String, Tuple(…│ - │ │… a Nullable(Int64)))) │ - └──────────────────┴─────────────────────────────────────┘ + ┏━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ arr ┃ toTypeName(arr) ┃ + ┡━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ [1,'Hello',(32)] │ Array(Variant(Int64, String, Tuple( + a Nullable(Int64)))) │ + └──────────────────┴─────────────────────────────────────────────────────────────┘ ┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┓ ┃ x ┃ toTypeName(x) ┃ ┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━┩ @@ -11,21 +11,21 @@ ├───────┼────────────────────────┤ 2. │ Hello │ Variant(Int64, String) │ └───────┴────────────────────────┘ - ┏━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ - ┃ x ┃ toTypeName(x) ┃ - ┡━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ -1. │ [1,2,3] │ Variant(Array(Nullable(Int64)), Tuple(…│ - │ │… a Nullable(Int64))) │ - ├─────────┼────────────────────────────────────────┤ -2. │ (42) │ Variant(Array(Nullable(Int64)), Tuple(…│ - │ │… a Nullable(Int64))) │ - └─────────┴────────────────────────────────────────┘ - ┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┓ - ┃ c1 ┃ toTypeName(c1) ┃ c2 ┃ toTypeName(c2) ┃ - ┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━┩ -1. │ 1 │ Nullable(Int64) │ Hello World! │ Nullable(String) │ - ├────┼─────────────────┼──────────────┼──────────────────┤ -2. │ 2 │ Nullable(Int64) │ [1,2,3] │ Nullable(String) │ - ├────┼─────────────────┼──────────────┼──────────────────┤ -3. │ 3 │ Nullable(Int64) │ 2020-01-01 │ Nullable(String) │ - └────┴─────────────────┴──────────────┴──────────────────┘ + ┏━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ x ┃ toTypeName(x) ┃ + ┡━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ [1,2,3] │ Variant(Array(Nullable(Int64)), Tuple( + a Nullable(Int64))) │ + ├─────────┼───────────────────────────────────────────────────────────────┤ +2. │ (42) │ Variant(Array(Nullable(Int64)), Tuple( + a Nullable(Int64))) │ + └─────────┴───────────────────────────────────────────────────────────────┘ + ┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ c1 ┃ toTypeName(c1) ┃ c2 ┃ toTypeName(c2) ┃ + ┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ 1 │ Nullable(Int64) │ Hello World! │ Variant(Array(Nullable(Int64)), String) │ + ├────┼─────────────────┼──────────────┼─────────────────────────────────────────┤ +2. │ 2 │ Nullable(Int64) │ [1,2,3] │ Variant(Array(Nullable(Int64)), String) │ + ├────┼─────────────────┼──────────────┼─────────────────────────────────────────┤ +3. │ 3 │ Nullable(Int64) │ 2020-01-01 │ Variant(Array(Nullable(Int64)), String) │ + └────┴─────────────────┴──────────────┴─────────────────────────────────────────┘ From 9cf11a210f07110676b373b864ea098583d87ff6 Mon Sep 17 00:00:00 2001 From: Blargian Date: Tue, 11 Jun 2024 11:11:06 +0200 Subject: [PATCH 012/260] Review changes --- src/Core/SettingsChangesHistory.h | 3 +- src/Formats/SchemaInferenceUtils.cpp | 73 ++++++------------- .../03150_infer_type_variant.reference | 24 +++--- 3 files changed, 35 insertions(+), 65 deletions(-) diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h index 3f743ef42bf..661ecc607ba 100644 --- a/src/Core/SettingsChangesHistory.h +++ b/src/Core/SettingsChangesHistory.h @@ -92,6 +92,7 @@ static std::map sett {"hdfs_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in HDFS table engine"}, {"azure_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in AzureBlobStorage table engine"}, {"s3_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in S3 table engine"}, + {"input_format_try_infer_variants", 0, 0, "Try to infer Variant type in text formats when there is more than one possible type for column/array elements"}, }}, {"24.5", {{"allow_deprecated_functions", true, false, "Allow usage of deprecated functions"}, {"allow_experimental_join_condition", false, false, "Support join with inequal conditions which involve columns from both left and right table. e.g. t1.y < t2.y."}, @@ -103,8 +104,6 @@ static std::map sett {"prefer_external_sort_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Prefer maximum block bytes for external sort, reduce the memory usage during merging."}, {"input_format_parquet_use_native_reader", false, false, "When reading Parquet files, to use native reader instead of arrow reader."}, {"input_format_force_null_for_omitted_fields", false, false, "Disable type-defaults for omitted fields when needed"}, - {"output_format_pretty_preserve_border_for_multiline_string", 1, 1, "Applies better rendering for multiline strings."}, - {"input_format_try_infer_variants", 0, 0, "Try to infer Variant type in text formats when there is more than one possible type for column/array elements"}, {"cast_string_to_dynamic_use_inference", false, false, "Add setting to allow converting String to Dynamic through parsing"}, {"allow_experimental_dynamic_type", false, false, "Add new experimental Dynamic type"}, {"azure_max_blocks_in_multipart_upload", 50000, 50000, "Maximum number of blocks in multipart upload for Azure."}, diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index a8b5d4343f5..b7c71a95b29 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -239,16 +239,6 @@ namespace return true; } - bool checkIfTypesContainVariant(const DataTypes & types) - { - for (size_t i = 0; i < types.size(); ++i) - { - if (isVariant(types[i])) - return true; - } - return false; - } - void updateTypeIndexes(DataTypes & data_types, TypeIndexesSet & type_indexes) { type_indexes.clear(); @@ -321,49 +311,28 @@ namespace /// if setting 'try_infer_variant' is true then we convert to type variant. void transformVariant(DataTypes & data_types, TypeIndexesSet & type_indexes) { - auto typesAreEqual = checkIfTypesAreEqual(data_types); - auto typesContainVariant = checkIfTypesContainVariant(data_types); - if (typesAreEqual) + if (checkIfTypesAreEqual(data_types)) return; - DataTypes new_data_types; - TypeIndexesSet new_type_indexes; - std::shared_ptr variant_type; - - /// extract the nested types of variant and make a new variant with the nested types and the other type. - /// eg. Type 1: variant, Type 2: Date -> variant. - if (typesContainVariant) + DataTypes variant_types; + for (const auto & type : data_types) { - DataTypes extracted_types; - for (size_t i=0; i(type.get())) { - if (isVariant(data_types[i])) - { - if (const auto * variant = typeid_cast(data_types[i].get())) - extracted_types = variant->getVariants(); - } - else - extracted_types.push_back(data_types[i]); + const auto & current_variants = variant_type->getVariants(); + variant_types.insert(variant_types.end(), current_variants.begin(), current_variants.end()); + } + else + { + variant_types.push_back(type); } - variant_type = std::make_shared(extracted_types); - } - else - { - variant_type = std::make_shared(data_types); } - size_t i = 0; - while (i != data_types.size()) - { - new_data_types.push_back(variant_type); - new_type_indexes.insert(TypeIndex::Variant); - i++; - } + auto variant_type = std::make_shared(variant_types); - data_types.clear(); - type_indexes.clear(); - data_types = new_data_types; - type_indexes = new_type_indexes; + for (auto & type : data_types) + type = variant_type; + type_indexes = {TypeIndex::Variant}; } /// If we have only Date and DateTime types, convert Date to DateTime, @@ -703,11 +672,12 @@ namespace if (settings.try_infer_dates || settings.try_infer_datetimes) transformDatesAndDateTimes(data_types, type_indexes); - if (settings.try_infer_variant) - transformVariant(data_types, type_indexes); - if constexpr (!is_json) + { + if (settings.try_infer_variant) + transformVariant(data_types, type_indexes); return; + } /// Check settings specific for JSON formats. @@ -740,11 +710,12 @@ namespace /// If there is at least one non Nothing type, change all Nothing types to it. transformNothingComplexTypes(data_types, type_indexes); - if (settings.try_infer_variant) - transformVariant(data_types, type_indexes); - if constexpr (!is_json) + { + if (settings.try_infer_variant) + transformVariant(data_types, type_indexes); return; + } /// Convert JSON tuples with same nested types to arrays. transformTuplesWithEqualNestedTypesToArrays(data_types, type_indexes); diff --git a/tests/queries/0_stateless/03150_infer_type_variant.reference b/tests/queries/0_stateless/03150_infer_type_variant.reference index a5f56cb3618..a43fa1e1227 100644 --- a/tests/queries/0_stateless/03150_infer_type_variant.reference +++ b/tests/queries/0_stateless/03150_infer_type_variant.reference @@ -1,16 +1,16 @@ - ┏━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ - ┃ arr ┃ toTypeName(arr) ┃ - ┡━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ -1. │ [1,'Hello',(32)] │ Array(Variant(Int64, String, Tuple( + ┏━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ + ┃ arr ┃ toTypeName(arr) ┃ + ┡━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ +1. │ ['1','Hello',(32)] │ Array(Variant(String, Tuple( a Nullable(Int64)))) │ - └──────────────────┴─────────────────────────────────────────────────────────────┘ - ┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┓ - ┃ x ┃ toTypeName(x) ┃ - ┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━┩ -1. │ 42 │ Variant(Int64, String) │ - ├───────┼────────────────────────┤ -2. │ Hello │ Variant(Int64, String) │ - └───────┴────────────────────────┘ + └────────────────────┴──────────────────────────────────────────────────────┘ + ┏━━━━━━━┳━━━━━━━━━━━━━━━━━━┓ + ┃ x ┃ toTypeName(x) ┃ + ┡━━━━━━━╇━━━━━━━━━━━━━━━━━━┩ +1. │ 42 │ Nullable(String) │ + ├───────┼──────────────────┤ +2. │ Hello │ Nullable(String) │ + └───────┴──────────────────┘ ┏━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ ┃ x ┃ toTypeName(x) ┃ ┡━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩ From 548c90901020317669a10d191a0b6f8a7d0a0511 Mon Sep 17 00:00:00 2001 From: Blargian Date: Tue, 11 Jun 2024 12:14:36 +0200 Subject: [PATCH 013/260] Add documentation --- docs/en/operations/settings/settings-formats.md | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/docs/en/operations/settings/settings-formats.md b/docs/en/operations/settings/settings-formats.md index 6aae1ea62e5..8bbb469547b 100644 --- a/docs/en/operations/settings/settings-formats.md +++ b/docs/en/operations/settings/settings-formats.md @@ -194,6 +194,17 @@ If enabled, ClickHouse will try to infer type `DateTime64` from string fields in Enabled by default. +## input_format_try_infer_variants {#input_format_try_infer_variants} + +If enabled, ClickHouse will try to infer type [`Variant`](../../sql-reference/data-types/variant.md) in schema inference for text formats when there is more than one possible type for column/array elements. + +Possible values: + +- 0 — Disabled. +- 1 — Enabled. + +Default value: `0`. + ## date_time_input_format {#date_time_input_format} Allows choosing a parser of the text representation of date and time. From bad5e27bbffa9c1f6727a0416edcb135dadcc1fe Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Tue, 11 Jun 2024 13:32:34 +0200 Subject: [PATCH 014/260] Update src/Formats/SchemaInferenceUtils.cpp Co-authored-by: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> --- src/Formats/SchemaInferenceUtils.cpp | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 0ac8b32f8aa..240830013c6 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -1539,9 +1539,7 @@ DataTypePtr makeNullableRecursively(DataTypePtr type) DataTypes nested_types; for (const auto & nested_type: variant_type->getVariants()) { - auto is_low_cardinality = nested_type->lowCardinality(); - auto has_sub_types = nested_type->haveSubtypes(); - if (!is_low_cardinality && has_sub_types) + if (!nested_type->lowCardinality() && nested_type->haveSubtypes()) nested_types.push_back(makeNullableRecursively(nested_type)); else nested_types.push_back(nested_type); From 384aa9feb90bbf95c5bc0e5498af4aca769c2531 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Mon, 24 Jun 2024 13:51:20 +0200 Subject: [PATCH 015/260] Move setting to 24.7 changes --- src/Core/SettingsChangesHistory.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h index 1ab7dc69f60..deaeba2a7de 100644 --- a/src/Core/SettingsChangesHistory.h +++ b/src/Core/SettingsChangesHistory.h @@ -86,6 +86,8 @@ namespace SettingsChangesHistory /// It's used to implement `compatibility` setting (see https://github.com/ClickHouse/ClickHouse/issues/35972) static const std::map settings_changes_history = { + {"24.7", {{"input_format_try_infer_variants", 0, 0, "Try to infer Variant type in text formats when there is more than one possible type for column/array elements"}, + }}, {"24.6", {{"materialize_skip_indexes_on_insert", true, true, "Added new setting to allow to disable materialization of skip indexes on insert"}, {"materialize_statistics_on_insert", true, true, "Added new setting to allow to disable materialization of statistics on insert"}, {"input_format_parquet_use_native_reader", false, false, "When reading Parquet files, to use native reader instead of arrow reader."}, @@ -97,7 +99,6 @@ static const std::map Date: Wed, 3 Jul 2024 09:33:57 +0200 Subject: [PATCH 016/260] Update SettingsChangesHistory.cpp --- src/Core/SettingsChangesHistory.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 7719fe1e837..828031f4c23 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -60,6 +60,7 @@ static std::initializer_list Date: Tue, 23 Jul 2024 18:04:38 +0800 Subject: [PATCH 017/260] stash --- src/Functions/FunctionOverlay.cpp | 481 ++++++++++++++++++++++++++++++ 1 file changed, 481 insertions(+) create mode 100644 src/Functions/FunctionOverlay.cpp diff --git a/src/Functions/FunctionOverlay.cpp b/src/Functions/FunctionOverlay.cpp new file mode 100644 index 00000000000..6160335ad79 --- /dev/null +++ b/src/Functions/FunctionOverlay.cpp @@ -0,0 +1,481 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ +extern const int ILLEGAL_COLUMN; +extern const int ILLEGAL_TYPE_OF_ARGUMENT; +extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; +extern const int ZERO_ARRAY_OR_TUPLE_INDEX; +} + +namespace +{ + +/// If 'is_utf8' - measure offset and length in code points instead of bytes. +/// Syntax: overlay(input, replace, offset[, length]) +template +class FunctionOverlay : public IFunction +{ +public: + static constexpr auto name = is_utf8 ? "OverlayUTF8" : "Overlay"; + + static FunctionPtr create(ContextPtr) { return std::make_shared(); } + String getName() const override { return name; } + bool isVariadic() const override { return true; } + size_t getNumberOfArguments() const override { return 0; } + bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } + bool useDefaultImplementationForConstants() const override { return true; } + + DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + { + const size_t number_of_arguments = arguments.size(); + if (number_of_arguments < 3 || number_of_arguments > 4) + throw Exception( + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Number of arguments for function {} doesn't match: " + "passed {}, should be 3 or 4", + getName(), + number_of_arguments); + + /// first argument is string + if (!isString(arguments[0])) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "Illegal type {} of first argument of function {}, expected String", + arguments[0]->getName(), + getName()); + + /// second argument is string + if (!isString(arguments[1])) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "Illegal type {} of second argument of function {}, expected String", + arguments[1]->getName(), + getName()); + + if (!isNativeNumber(arguments[2])) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "Illegal type {} of third argument of function {}, expected (U)Int8|16|32|64", + arguments[2]->getName(), + getName()); + + if (number_of_arguments == 4 && !isNativeNumber(arguments[3])) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "Illegal type {} of second argument of function {}, expected (U)Int8|16|32|64", + arguments[3]->getName(), + getName()); + + return std::make_shared(); + } + + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override + { + const size_t number_of_arguments = arguments.size(); + + ColumnPtr column_string = arguments[0].column; + ColumnPtr column_offset = arguments[1].column; + ColumnPtr column_length; + if (number_of_arguments == 3) + column_length = arguments[2].column; + + const ColumnConst * column_offset_const = checkAndGetColumn(column_offset.get()); + const ColumnConst * column_length_const = nullptr; + if (number_of_arguments == 3) + column_length_const = checkAndGetColumn(column_length.get()); + + Int64 offset = 0; + Int64 length = 0; + if (column_offset_const) + offset = column_offset_const->getInt(0); + if (column_length_const) + length = column_length_const->getInt(0); + + auto res_col = ColumnString::create(); + auto & res_data = res_col->getChars(); + auto & res_offsets = res_col->getOffsets(); + } + +private: +template + void constantConstant( + size_t rows, + const StringRef & input, + const StringRef & replace, + const ColumnPtr & column_offset, + const ColumnPtr & column_length, + Int64 const_offset, + Int64 const_length, + ColumnString::Chars & res_data, + ColumnString::Offsets & res_offsets) + { + if (!three_args && length_is_const && const_length < 0) + { + constantConstant(input, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); + return; + } + + Int64 offset = 0; // start from 1, maybe negative + size_t valid_offset = 0; // start from 0, not negative + if constexpr (offset_is_const) + { + offset = const_offset; + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + size_t replace_size = replace.size; + Int64 length = 0; // maybe negative + size_t valid_length = 0; // not negative + if constexpr (!three_args && length_is_const) + { + assert(const_length >= 0); + valid_length = const_length; + } + else if constexpr (three_args) + { + valid_length = replace_size; + } + + size_t res_offset = 0; + size_t input_size = input.size; + for (size_t i = 0; i < rows; ++i) + { + if constexpr (!offset_is_const) + { + offset = column_offset->getInt(i); + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + if constexpr (!three_args && !length_is_const) + { + length = column_length->getInt(i); + valid_length = length >= 0 ? length : replace_size; + } + + size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data, prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data + prefix_size + valid_length, suffix_size); + res_offset += suffix_size; + } + + /// add zero terminator + res_data[res_offset] = 0; + ++res_offset; + + res_offsets[i] = res_offset; + } + } + + template + void vectorConstant( + const ColumnString::Chars & input_data, + const ColumnString::Offsets & input_offsets, + const StringRef & replace, + const ColumnPtr & column_offset, + const ColumnPtr & column_length, + Int64 const_offset, + Int64 const_length, + ColumnString::Chars & res_data, + ColumnString::Offsets & res_offsets) + { + if (!three_args && length_is_const && const_length < 0) + { + vectorConstant(input_data, input_offsets, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); + return; + } + + Int64 offset = 0; // start from 1, maybe negative + size_t valid_offset = 0; // start from 0, not negative + if constexpr (offset_is_const) + { + offset = const_offset; + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + size_t replace_size = replace.size; + Int64 length = 0; // maybe negative + size_t valid_length = 0; // not negative + if constexpr (!three_args && length_is_const) + { + assert(const_length >= 0); + valid_length = const_length; + } + else if constexpr (three_args) + { + valid_length = replace_size; + } + + size_t rows = input_offsets.size(); + size_t res_offset = 0; + for (size_t i = 0; i < rows; ++i) + { + size_t input_offset = input_offsets[i - 1]; + size_t input_size = input_offsets[i] - input_offsets[i - 1] - 1; + + if constexpr (!offset_is_const) + { + offset = column_offset->getInt(i); + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + if constexpr (!three_args && !length_is_const) + { + length = column_length->getInt(i); + valid_length = length >= 0 ? length : replace_size; + } + + size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &input_data[input_offset], prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15( + &res_data[res_offset], &input_data[input_offset + prefix_size + valid_length], suffix_size); + res_offset += suffix_size; + } + + /// add zero terminator + res_data[res_offset] = 0; + ++res_offset; + + res_offsets[i] = res_offset; + } + } + + template + void constantVector( + const StringRef & input, + const ColumnString::Chars & replace_data, + const ColumnString::Offsets & replace_offsets, + const ColumnPtr & column_offset, + const ColumnPtr & column_length, + Int64 const_offset, + Int64 const_length, + ColumnString::Chars & res_data, + ColumnString::Offsets & res_offsets) + { + if (!three_args && length_is_const && const_length < 0) + { + constantVector(input, replace_data, replace_offsets, column_offset, column_length, const_offset, -1, res_data, res_offsets); + return; + } + + Int64 offset = 0; // start from 1, maybe negative + size_t valid_offset = 0; // start from 0, not negative + if constexpr (offset_is_const) + { + offset = const_offset; + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + Int64 length = 0; // maybe negative + size_t valid_length = 0; // not negative + if constexpr (!three_args && length_is_const) + { + assert(const_length >= 0); + valid_length = const_length; + } + + size_t rows = replace_offsets.size(); + size_t input_size = input.size; + size_t res_offset = 0; + for (size_t i = 0; i < rows; ++i) + { + size_t replace_offset = replace_offsets[i - 1]; + size_t replace_size = replace_offsets[i] - replace_offsets[i - 1] - 1; + + if constexpr (!offset_is_const) + { + offset = column_offset->getInt(i); + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + if constexpr (three_args) + { + // length = replace_size; + valid_length = replace_size; + } + else if constexpr (!length_is_const) + { + length = column_length->getInt(i); + valid_length = length >= 0 ? length : replace_size; + } + + size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data, prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data + prefix_size + valid_length, suffix_size); + res_offset += suffix_size; + } + + /// add zero terminator + res_data[res_offset] = 0; + ++res_offset; + + res_offsets[i] = res_offset; + } + } + + template + void vectorVector( + const ColumnString::Chars & input_data, + const ColumnString::Offsets & input_offsets, + const ColumnString::Chars & replace_data, + const ColumnString::Offsets & replace_offsets, + const ColumnPtr & column_offset, + const ColumnPtr & column_length, + Int64 const_offset, + Int64 const_length, + ColumnString::Chars & res_data, + ColumnString::Offsets & res_offsets) + { + if (!three_args && length_is_const && const_length < 0) + { + vectorVector( + input_data, + input_offsets, + replace_data, + replace_offsets, + column_offset, + column_length, + const_offset, + -1, + res_data, + res_offsets); + return; + } + + + Int64 offset = 0; // start from 1, maybe negative + size_t valid_offset = 0; // start from 0, not negative + if constexpr (offset_is_const) + { + offset = const_offset; + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + Int64 length = 0; // maybe negative + size_t valid_length = 0; // not negative + if constexpr (!three_args && length_is_const) + { + assert(const_length >= 0); + valid_length = const_length; + } + + size_t rows = input_offsets.size(); + size_t res_offset = 0; + for (size_t i = 0; i < rows; ++i) + { + size_t input_offset = input_offsets[i - 1]; + size_t input_size = input_offsets[i] - input_offsets[i - 1] - 1; + size_t replace_offset = replace_offsets[i - 1]; + size_t replace_size = replace_offsets[i] - replace_offsets[i - 1] - 1; + + if constexpr (!offset_is_const) + { + offset = column_offset->getInt(i); + valid_offset = offset > 0 ? (offset - 1) : (-offset); + } + + if constexpr (three_args) + { + // length = replace_size; + valid_length = replace_size; + } + else if constexpr (!length_is_const) + { + length = column_length->getInt(i); + valid_length = length >= 0 ? length : replace_size; + } + + size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &input_data[input_offset], prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15( + &res_data[res_offset], &input_data[input_offset + prefix_size + valid_length], suffix_size); + res_offset += suffix_size; + } + + /// add zero terminator + res_data[res_offset] = 0; + ++res_offset; + + res_offsets[i] = res_offset; + } + } +}; + +} + +REGISTER_FUNCTION(Overlay) +{ + factory.registerFunction>({}, FunctionFactory::CaseInsensitive); + factory.registerFunction>({}, FunctionFactory::CaseSensitive); +} + +} From 81688e0efdf75a6a3923d6b95f09579d37e93e2a Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Tue, 23 Jul 2024 20:29:35 +0800 Subject: [PATCH 018/260] almost finish --- src/Functions/FunctionOverlay.cpp | 154 ++++++++++++++++++++++++++---- 1 file changed, 134 insertions(+), 20 deletions(-) diff --git a/src/Functions/FunctionOverlay.cpp b/src/Functions/FunctionOverlay.cpp index 6160335ad79..65af4d811f5 100644 --- a/src/Functions/FunctionOverlay.cpp +++ b/src/Functions/FunctionOverlay.cpp @@ -15,10 +15,8 @@ namespace DB namespace ErrorCodes { -extern const int ILLEGAL_COLUMN; extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; -extern const int ZERO_ARRAY_OR_TUPLE_INDEX; } namespace @@ -86,32 +84,145 @@ public: ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const size_t number_of_arguments = arguments.size(); + bool three_args = number_of_arguments == 3; - ColumnPtr column_string = arguments[0].column; - ColumnPtr column_offset = arguments[1].column; + ColumnPtr column_offset = arguments[2].column; ColumnPtr column_length; - if (number_of_arguments == 3) - column_length = arguments[2].column; + if (!three_args) + column_length = arguments[3].column; const ColumnConst * column_offset_const = checkAndGetColumn(column_offset.get()); const ColumnConst * column_length_const = nullptr; - if (number_of_arguments == 3) + if (!three_args) column_length_const = checkAndGetColumn(column_length.get()); - Int64 offset = 0; - Int64 length = 0; + bool offset_is_const = false; + bool length_is_const = false; + Int64 offset = -1; + Int64 length = -1; if (column_offset_const) + { offset = column_offset_const->getInt(0); + offset_is_const = true; + } + if (column_length_const) + { length = column_length_const->getInt(0); + length_is_const = true; + } + auto res_col = ColumnString::create(); auto & res_data = res_col->getChars(); auto & res_offsets = res_col->getOffsets(); + res_offsets.resize_exact(input_rows_count); + + ColumnPtr column_input = arguments[0].column; + ColumnPtr column_replace = arguments[1].column; + + const auto * column_input_const = checkAndGetColumn(column_input.get()); + const auto * column_input_string = checkAndGetColumn(column_input.get()); + if (column_input_const) + { + StringRef input = column_input_const->getDataAt(0); + res_data.reserve(input.size * input_rows_count); + } + else + { + res_data.reserve(column_input_string->getChars().size()); + } + + const auto * column_replace_const = checkAndGetColumn(column_replace.get()); + const auto * column_replace_string = checkAndGetColumn(column_replace.get()); + bool input_is_const = column_input_const != nullptr; + bool replace_is_const = column_replace_const != nullptr; + +#define OVERLAY_EXECUTE_CASE(THREE_ARGS, OFFSET_IS_CONST, LENGTH_IS_CONST) \ + if (input_is_const && replace_is_const) \ + constantConstant( \ + input_rows_count, \ + column_input_const->getDataAt(0), \ + column_replace_const->getDataAt(0), \ + column_offset, \ + column_length, \ + offset, \ + length, \ + res_data, \ + res_offsets); \ + else if (input_is_const) \ + constantVector( \ + column_input_const->getDataAt(0), \ + column_replace_string->getChars(), \ + column_replace_string->getOffsets(), \ + column_offset, \ + column_length, \ + offset, \ + length, \ + res_data, \ + res_offsets); \ + else if (replace_is_const) \ + vectorConstant( \ + column_input_string->getChars(), \ + column_input_string->getOffsets(), \ + column_replace_const->getDataAt(0), \ + column_offset, \ + column_length, \ + offset, \ + length, \ + res_data, \ + res_offsets); \ + else \ + vectorVector( \ + column_input_string->getChars(), \ + column_input_string->getOffsets(), \ + column_replace_string->getChars(), \ + column_replace_string->getOffsets(), \ + column_offset, \ + column_length, \ + offset, \ + length, \ + res_data, \ + res_offsets); + + if (three_args) + { + if (offset_is_const) + { + OVERLAY_EXECUTE_CASE(true, true, false) + } + else + { + OVERLAY_EXECUTE_CASE(true, false, false) + } + } + else + { + if (offset_is_const && length_is_const) + { + OVERLAY_EXECUTE_CASE(false, true, true) + } + else if (offset_is_const && !length_is_const) + { + OVERLAY_EXECUTE_CASE(false, true, false) + } + else if (!offset_is_const && length_is_const) + { + OVERLAY_EXECUTE_CASE(false, false, true) + } + else + { + OVERLAY_EXECUTE_CASE(false, false, false) + } + } +#undef OVERLAY_EXECUTE_CASE + + return res_col; } + private: -template + template void constantConstant( size_t rows, const StringRef & input, @@ -121,11 +232,12 @@ template ( + rows, input, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); return; } @@ -194,7 +306,7 @@ template + template void vectorConstant( const ColumnString::Chars & input_data, const ColumnString::Offsets & input_offsets, @@ -204,11 +316,12 @@ template ( + input_data, input_offsets, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); return; } @@ -281,7 +394,7 @@ template + template void constantVector( const StringRef & input, const ColumnString::Chars & replace_data, @@ -291,11 +404,12 @@ template ( + input, replace_data, replace_offsets, column_offset, column_length, const_offset, -1, res_data, res_offsets); return; } @@ -379,11 +493,11 @@ template ( + vectorVector( input_data, input_offsets, replace_data, From 9785f85ca32e3af5760a9bc24e998e7d050fc073 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Tue, 23 Jul 2024 21:08:02 +0800 Subject: [PATCH 019/260] fix style --- src/Functions/FunctionOverlay.cpp | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/src/Functions/FunctionOverlay.cpp b/src/Functions/FunctionOverlay.cpp index 65af4d811f5..7d0e2e86de2 100644 --- a/src/Functions/FunctionOverlay.cpp +++ b/src/Functions/FunctionOverlay.cpp @@ -1,14 +1,10 @@ #include -#include #include -#include -#include #include #include #include #include -#include - +#include namespace DB { From f4138ee6c67bbdb82269a9087b1b054f33cb35a8 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 24 Jul 2024 10:20:27 +0800 Subject: [PATCH 020/260] fix bugs about corner cases --- src/Functions/FunctionOverlay.cpp | 89 +++++++++++++++++-------------- 1 file changed, 48 insertions(+), 41 deletions(-) diff --git a/src/Functions/FunctionOverlay.cpp b/src/Functions/FunctionOverlay.cpp index 7d0e2e86de2..d3ee7e1df6d 100644 --- a/src/Functions/FunctionOverlay.cpp +++ b/src/Functions/FunctionOverlay.cpp @@ -218,6 +218,26 @@ public: private: + /// input offset is 1-based, maybe negative + /// output result is 0-based valid offset, within [0, input_size] + static size_t getValidOffset(Int64 offset, size_t input_size) + { + if (offset > 0) + { + if (static_cast(offset) > input_size + 1) [[unlikely]] + return input_size; + else + return offset - 1; + } + else + { + if (input_size < -static_cast(offset)) [[unlikely]] + return 0; + else + return input_size + offset; + } + } + template void constantConstant( size_t rows, @@ -237,13 +257,10 @@ private: return; } - Int64 offset = 0; // start from 1, maybe negative + size_t input_size = input.size; size_t valid_offset = 0; // start from 0, not negative if constexpr (offset_is_const) - { - offset = const_offset; - valid_offset = offset > 0 ? (offset - 1) : (-offset); - } + valid_offset = getValidOffset(const_offset, input_size); size_t replace_size = replace.size; Int64 length = 0; // maybe negative @@ -258,14 +275,14 @@ private: valid_length = replace_size; } + Int64 offset = 0; // start from 1, maybe negative size_t res_offset = 0; - size_t input_size = input.size; for (size_t i = 0; i < rows; ++i) { if constexpr (!offset_is_const) { offset = column_offset->getInt(i); - valid_offset = offset > 0 ? (offset - 1) : (-offset); + valid_offset = getValidOffset(offset, input_size); } if constexpr (!three_args && !length_is_const) @@ -274,7 +291,7 @@ private: valid_length = length >= 0 ? length : replace_size; } - size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator res_data.resize(new_res_size); @@ -321,14 +338,6 @@ private: return; } - Int64 offset = 0; // start from 1, maybe negative - size_t valid_offset = 0; // start from 0, not negative - if constexpr (offset_is_const) - { - offset = const_offset; - valid_offset = offset > 0 ? (offset - 1) : (-offset); - } - size_t replace_size = replace.size; Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative @@ -343,16 +352,22 @@ private: } size_t rows = input_offsets.size(); + Int64 offset = 0; // start from 1, maybe negative + size_t valid_offset = 0; // start from 0, not negative size_t res_offset = 0; for (size_t i = 0; i < rows; ++i) { size_t input_offset = input_offsets[i - 1]; size_t input_size = input_offsets[i] - input_offsets[i - 1] - 1; - if constexpr (!offset_is_const) + if constexpr (offset_is_const) + { + valid_offset = getValidOffset(const_offset, input_size); + } + else { offset = column_offset->getInt(i); - valid_offset = offset > 0 ? (offset - 1) : (-offset); + valid_offset = getValidOffset(offset, input_size); } if constexpr (!three_args && !length_is_const) @@ -361,7 +376,7 @@ private: valid_length = length >= 0 ? length : replace_size; } - size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator res_data.resize(new_res_size); @@ -409,13 +424,10 @@ private: return; } - Int64 offset = 0; // start from 1, maybe negative + size_t input_size = input.size; size_t valid_offset = 0; // start from 0, not negative if constexpr (offset_is_const) - { - offset = const_offset; - valid_offset = offset > 0 ? (offset - 1) : (-offset); - } + valid_offset = getValidOffset(const_offset, input_size); Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative @@ -426,7 +438,7 @@ private: } size_t rows = replace_offsets.size(); - size_t input_size = input.size; + Int64 offset = 0; // start from 1, maybe negative size_t res_offset = 0; for (size_t i = 0; i < rows; ++i) { @@ -436,12 +448,11 @@ private: if constexpr (!offset_is_const) { offset = column_offset->getInt(i); - valid_offset = offset > 0 ? (offset - 1) : (-offset); + valid_offset = getValidOffset(offset, input_size); } if constexpr (three_args) { - // length = replace_size; valid_length = replace_size; } else if constexpr (!length_is_const) @@ -450,7 +461,7 @@ private: valid_length = length >= 0 ? length : replace_size; } - size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator res_data.resize(new_res_size); @@ -507,15 +518,6 @@ private: return; } - - Int64 offset = 0; // start from 1, maybe negative - size_t valid_offset = 0; // start from 0, not negative - if constexpr (offset_is_const) - { - offset = const_offset; - valid_offset = offset > 0 ? (offset - 1) : (-offset); - } - Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative if constexpr (!three_args && length_is_const) @@ -525,6 +527,8 @@ private: } size_t rows = input_offsets.size(); + Int64 offset = 0; // start from 1, maybe negative + size_t valid_offset = 0; // start from 0, not negative size_t res_offset = 0; for (size_t i = 0; i < rows; ++i) { @@ -533,15 +537,18 @@ private: size_t replace_offset = replace_offsets[i - 1]; size_t replace_size = replace_offsets[i] - replace_offsets[i - 1] - 1; - if constexpr (!offset_is_const) + if constexpr (offset_is_const) + { + valid_offset = getValidOffset(const_offset, input_size); + } + else { offset = column_offset->getInt(i); - valid_offset = offset > 0 ? (offset - 1) : (-offset); + valid_offset = getValidOffset(offset, input_size); } if constexpr (three_args) { - // length = replace_size; valid_length = replace_size; } else if constexpr (!length_is_const) @@ -550,7 +557,7 @@ private: valid_length = length >= 0 ? length : replace_size; } - size_t prefix_size = valid_offset > input_size ? input_size : valid_offset; + size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator res_data.resize(new_res_size); From fd3f0cf92b7800b171c5723541a329748a0dad1b Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 24 Jul 2024 14:17:58 +0800 Subject: [PATCH 021/260] support overlayUTF8 --- src/Functions/FunctionOverlay.cpp | 281 +++++++++++++----- .../0_stateless/03205_overlay.reference | 168 +++++++++++ tests/queries/0_stateless/03205_overlay.sql | 60 ++++ .../0_stateless/03206_overlay_utf8.reference | 168 +++++++++++ .../0_stateless/03206_overlay_utf8.sql | 60 ++++ 5 files changed, 665 insertions(+), 72 deletions(-) create mode 100644 tests/queries/0_stateless/03205_overlay.reference create mode 100644 tests/queries/0_stateless/03205_overlay.sql create mode 100644 tests/queries/0_stateless/03206_overlay_utf8.reference create mode 100644 tests/queries/0_stateless/03206_overlay_utf8.sql diff --git a/src/Functions/FunctionOverlay.cpp b/src/Functions/FunctionOverlay.cpp index d3ee7e1df6d..61d2df88ab1 100644 --- a/src/Functions/FunctionOverlay.cpp +++ b/src/Functions/FunctionOverlay.cpp @@ -3,8 +3,10 @@ #include #include #include +#include #include #include +#include namespace DB { @@ -15,6 +17,8 @@ extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; } +using namespace GatherUtils; + namespace { @@ -24,7 +28,7 @@ template class FunctionOverlay : public IFunction { public: - static constexpr auto name = is_utf8 ? "OverlayUTF8" : "Overlay"; + static constexpr auto name = is_utf8 ? "overlayUTF8" : "overlay"; static FunctionPtr create(ContextPtr) { return std::make_shared(); } String getName() const override { return name; } @@ -238,6 +242,15 @@ private: } } + /// get character count of a slice [data, data+bytes) + static size_t getSliceSize(const UInt8 * data, size_t bytes) + { + if constexpr (is_utf8) + return UTF8::countCodePoints(data, bytes); + else + return bytes; + } + template void constantConstant( size_t rows, @@ -257,13 +270,12 @@ private: return; } - size_t input_size = input.size; + size_t input_size = getSliceSize(reinterpret_cast(input.data), input.size); size_t valid_offset = 0; // start from 0, not negative if constexpr (offset_is_const) valid_offset = getValidOffset(const_offset, input_size); - size_t replace_size = replace.size; - Int64 length = 0; // maybe negative + size_t replace_size = getSliceSize(reinterpret_cast(replace.data), replace.size); size_t valid_length = 0; // not negative if constexpr (!three_args && length_is_const) { @@ -276,6 +288,9 @@ private: } Int64 offset = 0; // start from 1, maybe negative + Int64 length = 0; // maybe negative + const UInt8 * input_begin = reinterpret_cast(input.data); + const UInt8 * input_end = reinterpret_cast(input.data + input.size); size_t res_offset = 0; for (size_t i = 0; i < rows; ++i) { @@ -293,28 +308,57 @@ private: size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; - size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator - res_data.resize(new_res_size); - /// copy prefix before replaced region - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data, prefix_size); - res_offset += prefix_size; - - /// copy replace - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace_size); - res_offset += replace_size; - - /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. - if (suffix_size) + if constexpr (!is_utf8) { - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data + prefix_size + valid_length, suffix_size); - res_offset += suffix_size; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data, prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data + prefix_size + valid_length, suffix_size); + res_offset += suffix_size; + } + } + else + { + const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + size_t prefix_bytes = prefix_end > input_end ? input.size : prefix_end - input_begin; + + const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + size_t suffix_bytes = input_end - suffix_begin; + + size_t new_res_size = res_data.size() + prefix_bytes + replace.size + suffix_bytes + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input_begin, prefix_bytes); + res_offset += prefix_bytes; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace.size); + res_offset += replace.size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_bytes is zero. + if (suffix_bytes) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], suffix_begin, suffix_bytes); + res_offset += suffix_bytes; + } } /// add zero terminator res_data[res_offset] = 0; ++res_offset; - res_offsets[i] = res_offset; } } @@ -338,7 +382,7 @@ private: return; } - size_t replace_size = replace.size; + size_t replace_size = getSliceSize(reinterpret_cast(replace.data), replace.size); Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative if constexpr (!three_args && length_is_const) @@ -358,7 +402,8 @@ private: for (size_t i = 0; i < rows; ++i) { size_t input_offset = input_offsets[i - 1]; - size_t input_size = input_offsets[i] - input_offsets[i - 1] - 1; + size_t input_bytes = input_offsets[i] - input_offsets[i - 1] - 1; + size_t input_size = getSliceSize(&input_data[input_offset], input_bytes); if constexpr (offset_is_const) { @@ -378,29 +423,59 @@ private: size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; - size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator - res_data.resize(new_res_size); - /// copy prefix before replaced region - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &input_data[input_offset], prefix_size); - res_offset += prefix_size; - - /// copy replace - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace_size); - res_offset += replace_size; - - /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. - if (suffix_size) + if constexpr (!is_utf8) { - memcpySmallAllowReadWriteOverflow15( - &res_data[res_offset], &input_data[input_offset + prefix_size + valid_length], suffix_size); - res_offset += suffix_size; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &input_data[input_offset], prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15( + &res_data[res_offset], &input_data[input_offset + prefix_size + valid_length], suffix_size); + res_offset += suffix_size; + } + } + else + { + const auto * input_begin = &input_data[input_offset]; + const auto * input_end = &input_data[input_offset + input_bytes]; + const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + size_t prefix_bytes = prefix_end > input_end ? input_bytes : prefix_end - input_begin; + const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + size_t suffix_bytes = input_end - suffix_begin; + + size_t new_res_size = res_data.size() + prefix_bytes + replace.size + suffix_bytes + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &input_data[input_offset], prefix_bytes); + res_offset += prefix_bytes; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], replace.data, replace.size); + res_offset += replace.size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_bytes is zero. + if (suffix_bytes) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], suffix_begin, suffix_bytes); + res_offset += suffix_bytes; + } } /// add zero terminator res_data[res_offset] = 0; ++res_offset; - res_offsets[i] = res_offset; } } @@ -424,7 +499,7 @@ private: return; } - size_t input_size = input.size; + size_t input_size = getSliceSize(reinterpret_cast(input.data), input.size); size_t valid_offset = 0; // start from 0, not negative if constexpr (offset_is_const) valid_offset = getValidOffset(const_offset, input_size); @@ -438,12 +513,15 @@ private: } size_t rows = replace_offsets.size(); + const auto * input_begin = reinterpret_cast(input.data); + const auto * input_end = reinterpret_cast(input.data + input.size); Int64 offset = 0; // start from 1, maybe negative size_t res_offset = 0; for (size_t i = 0; i < rows; ++i) { size_t replace_offset = replace_offsets[i - 1]; - size_t replace_size = replace_offsets[i] - replace_offsets[i - 1] - 1; + size_t replace_bytes = replace_offsets[i] - replace_offsets[i - 1] - 1; + size_t replace_size = getSliceSize(&replace_data[replace_offset], replace_bytes); if constexpr (!offset_is_const) { @@ -463,28 +541,55 @@ private: size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; - size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator - res_data.resize(new_res_size); - /// copy prefix before replaced region - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data, prefix_size); - res_offset += prefix_size; - - /// copy replace - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_size); - res_offset += replace_size; - - /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. - if (suffix_size) + if constexpr (!is_utf8) { - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data + prefix_size + valid_length, suffix_size); - res_offset += suffix_size; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data, prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input.data + prefix_size + valid_length, suffix_size); + res_offset += suffix_size; + } + } + else + { + const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + size_t prefix_bytes = prefix_end > input_end ? input.size : prefix_end - input_begin; + const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + size_t suffix_bytes = input_end - suffix_begin; + size_t new_res_size = res_data.size() + prefix_bytes + replace_bytes + suffix_bytes + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input_begin, prefix_bytes); + res_offset += prefix_bytes; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_bytes); + res_offset += replace_bytes; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_bytes is zero + if (suffix_bytes) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], suffix_begin, suffix_bytes); + res_offset += suffix_bytes; + } } /// add zero terminator res_data[res_offset] = 0; ++res_offset; - res_offsets[i] = res_offset; } } @@ -533,9 +638,12 @@ private: for (size_t i = 0; i < rows; ++i) { size_t input_offset = input_offsets[i - 1]; - size_t input_size = input_offsets[i] - input_offsets[i - 1] - 1; + size_t input_bytes = input_offsets[i] - input_offsets[i - 1] - 1; + size_t input_size = getSliceSize(&input_data[input_offset], input_bytes); + size_t replace_offset = replace_offsets[i - 1]; - size_t replace_size = replace_offsets[i] - replace_offsets[i - 1] - 1; + size_t replace_bytes = replace_offsets[i] - replace_offsets[i - 1] - 1; + size_t replace_size = getSliceSize(&replace_data[replace_offset], replace_bytes); if constexpr (offset_is_const) { @@ -559,29 +667,58 @@ private: size_t prefix_size = valid_offset; size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; - size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator - res_data.resize(new_res_size); - /// copy prefix before replaced region - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &input_data[input_offset], prefix_size); - res_offset += prefix_size; - - /// copy replace - memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_size); - res_offset += replace_size; - - /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. - if (suffix_size) + if constexpr (!is_utf8) { - memcpySmallAllowReadWriteOverflow15( - &res_data[res_offset], &input_data[input_offset + prefix_size + valid_length], suffix_size); - res_offset += suffix_size; + size_t new_res_size = res_data.size() + prefix_size + replace_size + suffix_size + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &input_data[input_offset], prefix_size); + res_offset += prefix_size; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_size); + res_offset += replace_size; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_size is zero. + if (suffix_size) + { + memcpySmallAllowReadWriteOverflow15( + &res_data[res_offset], &input_data[input_offset + prefix_size + valid_length], suffix_size); + res_offset += suffix_size; + } + } + else + { + const auto * input_begin = &input_data[input_offset]; + const auto * input_end = &input_data[input_offset + input_bytes]; + const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + size_t prefix_bytes = prefix_end > input_end ? input_bytes : prefix_end - input_begin; + const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + size_t suffix_bytes = input_end - suffix_begin; + size_t new_res_size = res_data.size() + prefix_bytes + replace_bytes + suffix_bytes + 1; /// +1 for zero terminator + res_data.resize(new_res_size); + + /// copy prefix before replaced region + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], input_begin, prefix_bytes); + res_offset += prefix_bytes; + + /// copy replace + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], &replace_data[replace_offset], replace_bytes); + res_offset += replace_bytes; + + /// copy suffix after replaced region. It is not necessary to copy if suffix_bytes is zero. + if (suffix_bytes) + { + memcpySmallAllowReadWriteOverflow15(&res_data[res_offset], suffix_begin, suffix_bytes); + res_offset += suffix_bytes; + } } /// add zero terminator res_data[res_offset] = 0; ++res_offset; - res_offsets[i] = res_offset; } } diff --git a/tests/queries/0_stateless/03205_overlay.reference b/tests/queries/0_stateless/03205_overlay.reference new file mode 100644 index 00000000000..9e79db2e131 --- /dev/null +++ b/tests/queries/0_stateless/03205_overlay.reference @@ -0,0 +1,168 @@ +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark_SQL +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark CORE +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Spark ANSI SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL +Structured SQL diff --git a/tests/queries/0_stateless/03205_overlay.sql b/tests/queries/0_stateless/03205_overlay.sql new file mode 100644 index 00000000000..b131312c934 --- /dev/null +++ b/tests/queries/0_stateless/03205_overlay.sql @@ -0,0 +1,60 @@ +SELECT overlay('Spark SQL', 'ANSI ', 7, 0) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, 0) from numbers(3); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlay('Spark SQL', 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), 0) from numbers(3); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, materialize(0)) from numbers(3); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), materialize(0)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), 0) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, materialize(0)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), materialize(0)) from numbers(3); +SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); + +SELECT overlay('Spark SQL', '_', 6) from numbers(3); +SELECT overlay(materialize('Spark SQL'), '_', 6) from numbers(3); +SELECT overlay('Spark SQL', materialize('_'), 6) from numbers(3); +SELECT overlay('Spark SQL', '_', materialize(6)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('_'), 6) from numbers(3); +SELECT overlay(materialize('Spark SQL'), '_', materialize(6)) from numbers(3); +SELECT overlay('Spark SQL', materialize('_'), materialize(6)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('_'), materialize(6)) from numbers(3); + +SELECT overlay('Spark SQL', 'CORE', 7) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'CORE', 7) from numbers(3); +SELECT overlay('Spark SQL', materialize('CORE'), 7) from numbers(3); +SELECT overlay('Spark SQL', 'CORE', materialize(7)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('CORE'), 7) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'CORE', materialize(7)) from numbers(3); +SELECT overlay('Spark SQL', materialize('CORE'), materialize(7)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('CORE'), materialize(7)) from numbers(3); + +SELECT overlay('Spark SQL', 'ANSI ', 7, 0) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, 0) from numbers(3); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlay('Spark SQL', 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), 0) from numbers(3); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, materialize(0)) from numbers(3); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), materialize(0)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); + +SELECT overlay('Spark SQL', 'tructured', 2, 4) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'tructured', 2, 4) from numbers(3); +SELECT overlay('Spark SQL', materialize('tructured'), 2, 4) from numbers(3); +SELECT overlay('Spark SQL', 'tructured', materialize(2), 4) from numbers(3); +SELECT overlay('Spark SQL', 'tructured', 2, materialize(4)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('tructured'), 2, 4) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'tructured', materialize(2), 4) from numbers(3); +SELECT overlay(materialize('Spark SQL'), 'tructured', 2, materialize(4)) from numbers(3); +SELECT overlay('Spark SQL', materialize('tructured'), materialize(2), 4) from numbers(3); +SELECT overlay('Spark SQL', materialize('tructured'), 2, materialize(4)) from numbers(3); +SELECT overlay('Spark SQL', 'tructured', materialize(2), materialize(4)) from numbers(3); +SELECT overlay(materialize('Spark SQL'), materialize('tructured'), materialize(2), materialize(4)) from numbers(3); diff --git a/tests/queries/0_stateless/03206_overlay_utf8.reference b/tests/queries/0_stateless/03206_overlay_utf8.reference new file mode 100644 index 00000000000..19878c97184 --- /dev/null +++ b/tests/queries/0_stateless/03206_overlay_utf8.reference @@ -0,0 +1,168 @@ +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark_SQL和CH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark CORECH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Spark ANSI SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH +Structured SQL和CH diff --git a/tests/queries/0_stateless/03206_overlay_utf8.sql b/tests/queries/0_stateless/03206_overlay_utf8.sql new file mode 100644 index 00000000000..00b756c8b5b --- /dev/null +++ b/tests/queries/0_stateless/03206_overlay_utf8.sql @@ -0,0 +1,60 @@ +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, 0) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, materialize(0)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), materialize(0)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), 0) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, materialize(0)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), materialize(0)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); + +SELECT overlayUTF8('Spark SQL和CH', '_', 6) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), '_', 6) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('_'), 6) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', '_', materialize(6)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('_'), 6) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), '_', materialize(6)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('_'), materialize(6)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('_'), materialize(6)) from numbers(3); + +SELECT overlayUTF8('Spark SQL和CH', 'CORE', 7) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'CORE', 7) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('CORE'), 7) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'CORE', materialize(7)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('CORE'), 7) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'CORE', materialize(7)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('CORE'), materialize(7)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('CORE'), materialize(7)) from numbers(3); + +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, 0) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, 0) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), 0) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, materialize(0)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), 0) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, materialize(0)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), materialize(0)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); + +SELECT overlayUTF8('Spark SQL和CH', 'tructured', 2, 4) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'tructured', 2, 4) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('tructured'), 2, 4) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'tructured', materialize(2), 4) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'tructured', 2, materialize(4)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('tructured'), 2, 4) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'tructured', materialize(2), 4) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), 'tructured', 2, materialize(4)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('tructured'), materialize(2), 4) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', materialize('tructured'), 2, materialize(4)) from numbers(3); +SELECT overlayUTF8('Spark SQL和CH', 'tructured', materialize(2), materialize(4)) from numbers(3); +SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('tructured'), materialize(2), materialize(4)) from numbers(3); From c09c22b17575396e38fb45cb385dcc8a49f9a183 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 24 Jul 2024 14:45:47 +0800 Subject: [PATCH 022/260] finish doc --- .../functions/string-replace-functions.md | 72 +++++++++++++++++++ ...new_functions_must_be_documented.reference | 2 + 2 files changed, 74 insertions(+) diff --git a/docs/en/sql-reference/functions/string-replace-functions.md b/docs/en/sql-reference/functions/string-replace-functions.md index 8793ebdd1a3..4e1f89fd974 100644 --- a/docs/en/sql-reference/functions/string-replace-functions.md +++ b/docs/en/sql-reference/functions/string-replace-functions.md @@ -223,3 +223,75 @@ SELECT translateUTF8('Münchener Straße', 'üß', 'us') AS res; │ Munchener Strase │ └──────────────────┘ ``` + +## overlay + +Replace the string `s` with the string `replace` starting from the 1-based `position` for `length` bytes. If `length` is omitted or negative, then it defaults to the length of `replace`. + +**Syntax** + +```sql +overlay(s, replace, position[, length]) +``` + +**Parameters** + +- `s`: A string type [String](../data-types/string.md). +- `replace`: A string type [String](../data-types/string.md). +- `position`: An integer type [Int](../data-types/int.md). +- `length`: Optional. An integer type [Int](../data-types/int.md). + +**Returned value** + +- A [String](../data-types/string.md) data type value. If `position` is negative the position is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. + +**Example** + +```sql +SELECT overlay('Spark SQL', 'CORE', 7) AS res; +``` + +Result: + +```text + ┌─res────────┐ + │ Spark CORE │ + └────────────┘ +``` + +## overlayUTF8 + +Replace the string `s` with the string `replace` starting from the 1-based `position` for `length` UTF-8 characters. If `length` is omitted or negative, then it defaults to the length of `replace`. + +Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. + +**Syntax** + +```sql +overlayUTF8(s, replace, position[, length]) +``` + +**Parameters** + +- `s`: A string type [String](../data-types/string.md). +- `replace`: A string type [String](../data-types/string.md). +- `position`: An integer type [Int](../data-types/int.md). +- `length`: Optional. An integer type [Int](../data-types/int.md). + +**Returned value** + +- A [String](../data-types/string.md) data type value. If `position` is negative the position is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. + +**Example** + +```sql +SELECT overlayUTF8('ClickHouse是一款OLAP数据库', '开源', 12, 2) AS res; +``` + +Result: + +```text +┌─res────────────────────────┐ +│ ClickHouse是开源OLAP数据库 │ +└────────────────────────────┘ +``` diff --git a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference index a152066a460..ba9d3fb7a83 100644 --- a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference +++ b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference @@ -512,6 +512,8 @@ nullIf nullIn nullInIgnoreSet or +overlay +overlayUTF8 parseDateTime parseDateTime32BestEffort parseDateTime32BestEffortOrNull From c837541a7783f14780a7d2535dd6fa2cbf5effd5 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 25 Jul 2024 10:11:53 +0800 Subject: [PATCH 023/260] fix style --- utils/check-style/aspell-ignore/en/aspell-dict.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index 943caf918d6..fa26cc0ff1f 100644 --- a/utils/check-style/aspell-ignore/en/aspell-dict.txt +++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt @@ -2166,6 +2166,7 @@ outfile overcommit overcommitted overfitting +overlayUTF overparallelization packetpool packetsize From 3db505a1327fc5bf96c93f2a510436402be13f3b Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Sat, 27 Jul 2024 12:53:38 +0200 Subject: [PATCH 024/260] Update SettingsChangesHistory.cpp --- src/Core/SettingsChangesHistory.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 1dda9e72084..dc3bf984cc6 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -67,6 +67,7 @@ static std::initializer_list Date: Mon, 29 Jul 2024 13:03:21 +0200 Subject: [PATCH 025/260] Move setting to 24.8 version --- src/Core/SettingsChangesHistory.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index dc3bf984cc6..41319ac7645 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -57,6 +57,8 @@ String ClickHouseVersion::toString() const /// Note: please check if the key already exists to prevent duplicate entries. static std::initializer_list> settings_changes_history_initializer = { + {"24.8", {{"input_format_try_infer_variants", false, false, "Try to infer Variant type in text formats when there is more than one possible type for column/array elements"}, + }}, {"24.7", {{"output_format_parquet_write_page_index", false, true, "Add a possibility to write page index into parquet files."}, {"output_format_binary_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in RowBinaryWithNamesAndTypes output format"}, {"input_format_binary_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in RowBinaryWithNamesAndTypes input format"}, @@ -66,7 +68,6 @@ static std::initializer_list Date: Mon, 5 Aug 2024 21:10:31 +0000 Subject: [PATCH 026/260] Allow to specify min and max for random settings in the test --- docs/en/development/tests.md | 22 +++++ tests/clickhouse-test | 92 ++++++++++++++++--- ...mic_read_subcolumns_compact_merge_tree.sql | 1 + ...ynamic_read_subcolumns_wide_merge_tree.sql | 1 + ...merges_1_horizontal_compact_merge_tree.sql | 4 + ..._merges_1_horizontal_compact_wide_tree.sql | 2 + ...c_merges_1_vertical_compact_merge_tree.sql | 2 + ...amic_merges_1_vertical_wide_merge_tree.sql | 2 + ...merges_2_horizontal_compact_merge_tree.sql | 1 + ...ic_merges_2_horizontal_wide_merge_tree.sql | 1 + ...c_merges_2_vertical_compact_merge_tree.sql | 1 + ...amic_merges_2_vertical_wide_merge_tree.sql | 1 + ...sted_dynamic_merges_compact_horizontal.sql | 1 + ...nested_dynamic_merges_compact_vertical.sql | 1 + ..._nested_dynamic_merges_wide_horizontal.sql | 1 + ...38_nested_dynamic_merges_wide_vertical.sql | 1 + 16 files changed, 122 insertions(+), 12 deletions(-) diff --git a/docs/en/development/tests.md b/docs/en/development/tests.md index 269995a1a96..f0afa983fec 100644 --- a/docs/en/development/tests.md +++ b/docs/en/development/tests.md @@ -91,6 +91,28 @@ SELECT 1 In addition to the above settings, you can use `USE_*` flags from `system.build_options` to define usage of particular ClickHouse features. For example, if your test uses a MySQL table, you should add a tag `use-mysql`. +### Specifying limits for random settings + +A test can specify minimum and maximum allowed values for settings that can be randomized during test run. + +For `.sh` tests limits are written as a comment on the line next to tags or on the second line if no tags are specified: + +```bash +#!/usr/bin/env bash +# Tags: no-fasttest +# Random settings limits: max_block_size=(1000, 10000), index_granularity=(100, None) +``` + +For `.sql` tests tags are placed as a SQL comment in the line next to tags or in the first line: + +```sql +-- Tags: no-fasttest +-- Random settings limits: max_block_size=(1000, 10000), index_granularity=(100, None) +SELECT 1 +``` + +If you need to specify only one limit, you can use `None` for another one. + ### Choosing the Test Name The name of the test starts with a five-digit prefix followed by a descriptive name, such as `00422_hash_function_constexpr.sql`. To choose the prefix, find the largest prefix already present in the directory, and increment it by one. In the meantime, some other tests might be added with the same numeric prefix, but this is OK and does not lead to any problems, you don't have to change it later. diff --git a/tests/clickhouse-test b/tests/clickhouse-test index a29c786e998..ea488e7c3dd 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -39,6 +39,7 @@ from errno import ESRCH from subprocess import PIPE, Popen from time import sleep, time from typing import Dict, List, Optional, Set, Tuple, Union +from ast import literal_eval as make_tuple try: import termcolor # type: ignore @@ -1068,9 +1069,25 @@ class TestCase: return description + "\n" + def apply_random_settings_limits(self, random_settings): + print("Random settings limits:", self.random_settings_limits) + for setting in random_settings: + if setting in self.random_settings_limits: + min = self.random_settings_limits[setting][0] + if min and random_settings[setting] < min: + random_settings[setting] = min + max = self.random_settings_limits[setting][1] + if max and random_settings[setting] > max: + random_settings[setting] = max + def __init__(self, suite, case: str, args, is_concurrent: bool): self.case: str = case # case file name self.tags: Set[str] = suite.all_tags[case] if case in suite.all_tags else set() + self.random_settings_limits = ( + suite.all_random_settings_limits[case] + if case in suite.all_random_settings_limits + else dict() + ) for tag in os.getenv("GLOBAL_TAGS", "").split(","): self.tags.add(tag.strip()) @@ -1112,11 +1129,13 @@ class TestCase: if self.randomize_settings: self.random_settings = SettingsRandomizer.get_random_settings(args) + self.apply_random_settings_limits(self.random_settings) if self.randomize_merge_tree_settings: self.merge_tree_random_settings = ( MergeTreeSettingsRandomizer.get_random_settings(args) ) + self.apply_random_settings_limits(self.merge_tree_random_settings) self.base_url_params = ( os.environ["CLICKHOUSE_URL_PARAMS"] @@ -1900,7 +1919,9 @@ class TestSuite: return test_name @staticmethod - def read_test_tags(suite_dir: str, all_tests: List[str]) -> Dict[str, Set[str]]: + def read_test_tags_and_random_settings_limits( + suite_dir: str, all_tests: List[str] + ) -> (Dict[str, Set[str]], Dict[str, Dict[str, Tuple[int, int]]]): def get_comment_sign(filename): if filename.endswith(".sql") or filename.endswith(".sql.j2"): return "--" @@ -1925,22 +1946,48 @@ class TestSuite: tags = {tag.strip() for tag in tags} return tags + def parse_random_settings_limits_from_line( + line, comment_sign + ) -> Dict[str, Tuple[int, int]]: + if not line.startswith(comment_sign): + return {} + random_settings_limits_str = line[ + len(comment_sign) : + ].lstrip() # noqa: ignore E203 + random_settings_limits_prefix = "Random settings limits:" + if not random_settings_limits_str.startswith(random_settings_limits_prefix): + return {} + random_settings_limits_str = random_settings_limits_str[ + len(random_settings_limits_prefix) : + ] # noqa: ignore E203 + # limits are specified in a form 'setting1=(min, max), setting2=(min,max), ...' + random_settings_limits = re.findall( + "([^=, ]+) *= *(\([^=]+\))", random_settings_limits_str + ) + random_settings_limits = { + pair[0]: make_tuple(pair[1]) for pair in random_settings_limits + } + return random_settings_limits + def is_shebang(line: str) -> bool: return line.startswith("#!") def find_tag_line(file): - for line in file: + line = file.readline() + while line != "": line = line.strip() if line and not is_shebang(line): return line + line = file.readline() return "" - def load_tags_from_file(filepath): + def load_tags_and_random_settings_limits_from_file(filepath): comment_sign = get_comment_sign(filepath) need_query_params = False with open(filepath, "r", encoding="utf-8") as file: try: tag_line = find_tag_line(file) + next_line = file.readline() except UnicodeDecodeError: return [] try: @@ -1950,21 +1997,35 @@ class TestSuite: need_query_params = True except UnicodeDecodeError: pass - parsed_tags = parse_tags_from_line(tag_line, comment_sign) - if need_query_params: - parsed_tags.add("need-query-parameters") - return parsed_tags + parsed_tags = parse_tags_from_line(tag_line, comment_sign) + if need_query_params: + parsed_tags.add("need-query-parameters") + random_settings_limits_line = next_line if parsed_tags else tag_line + random_settings_limits = parse_random_settings_limits_from_line( + random_settings_limits_line, comment_sign + ) + return parsed_tags, random_settings_limits all_tags = {} + all_random_settings_limits = {} start_time = datetime.now() for test_name in all_tests: - tags = load_tags_from_file(os.path.join(suite_dir, test_name)) + ( + tags, + random_settings_limits, + ) = load_tags_and_random_settings_limits_from_file( + os.path.join(suite_dir, test_name) + ) if tags: all_tags[test_name] = tags + if random_settings_limits: + all_random_settings_limits[test_name] = random_settings_limits elapsed = (datetime.now() - start_time).total_seconds() if elapsed > 1: - print(f"Tags for suite {suite_dir} read in {elapsed:.2f} seconds") - return all_tags + print( + f"Tags and random settings limits for suite {suite_dir} read in {elapsed:.2f} seconds" + ) + return all_tags, all_random_settings_limits def __init__(self, args, suite_path: str, suite_tmp_path: str, suite: str): self.args = args @@ -1994,9 +2055,16 @@ class TestSuite: self.all_tests: List[str] = self.get_tests_list( self.tests_in_suite_key_func, filter_func ) - self.all_tags: Dict[str, Set[str]] = self.read_test_tags( - self.suite_path, self.all_tests + + all_tags_and_random_settings_limits = ( + self.read_test_tags_and_random_settings_limits( + self.suite_path, self.all_tests + ) ) + self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] + self.all_random_settings_limits: Dict[ + str, Dict[str, (int, int)] + ] = all_tags_and_random_settings_limits[1] self.sequential_tests = [] self.parallel_tests = [] diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.sql b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.sql index ddfba4418bd..822393d3c78 100644 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.sql @@ -1,4 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.sql b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.sql index 5aac5f7b72f..2394893dc8b 100644 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.sql @@ -1,4 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql index d2c787040e5..7c2e7c3d2be 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql @@ -1,4 +1,6 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) + set allow_experimental_dynamic_type=1; drop table if exists test; @@ -31,3 +33,5 @@ optimize table test final; select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); drop table test; + +select 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql index f99bf771608..aa62435188a 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql @@ -1,4 +1,6 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) + set allow_experimental_dynamic_type=1; drop table if exists test; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql index be81596d043..bfc7bb9d206 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql @@ -1,4 +1,6 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) + set allow_experimental_dynamic_type=1; drop table if exists test; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql index f6396af42a8..233667db0a7 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql @@ -1,4 +1,6 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) + set allow_experimental_dynamic_type=1; drop table if exists test; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql index e133ac3001f..48a6a55378c 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql @@ -1,4 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql index d527081b763..44b298b1c35 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql @@ -1,4 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql index ebccfb77922..f42150720b3 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql @@ -1,4 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql index 104d6018e41..ee4ff6af162 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql @@ -1,4 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql index 1d5c63dcdf1..e0636f053df 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql @@ -1,4 +1,5 @@ -- Tags: long +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql index 2bffe35c577..edfad295e9a 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql @@ -1,4 +1,5 @@ -- Tags: long +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql index fb686091ebb..79d488ec253 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql @@ -1,4 +1,5 @@ -- Tags: long +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql index ed195452d56..e2a453b867a 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql @@ -1,4 +1,5 @@ -- Tags: long +-- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; From d3dc17453377368defd80cda9f4b95dda6adc9df Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 5 Aug 2024 21:15:11 +0000 Subject: [PATCH 027/260] Remove log --- tests/clickhouse-test | 17 +---------------- 1 file changed, 1 insertion(+), 16 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 389193836bf..5fcb9fb80f1 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -682,7 +682,6 @@ class FailureReason(enum.Enum): BUILD = "not running for current build" NO_PARALLEL_REPLICAS = "smth in not supported with parallel replicas" SHARED_MERGE_TREE = "no-shared-merge-tree" - DISTRIBUTED_CACHE = "distributed-cache" # UNKNOWN reasons NO_REFERENCE = "no reference file" @@ -1071,7 +1070,6 @@ class TestCase: return description + "\n" def apply_random_settings_limits(self, random_settings): - print("Random settings limits:", self.random_settings_limits) for setting in random_settings: if setting in self.random_settings_limits: min = self.random_settings_limits[setting][0] @@ -1211,9 +1209,6 @@ class TestCase: elif tags and ("no-replicated-database" in tags) and args.replicated_database: return FailureReason.REPLICATED_DB - elif tags and ("no-distributed-cache" in tags) and args.distributed_cache: - return FailureReason.DISTRIBUTED_CACHE - elif ( tags and ("atomic-database" in tags) @@ -1251,11 +1246,6 @@ class TestCase: ): return FailureReason.SKIP - elif "no-flaky-check" in tags and ( - 1 == int(os.environ.get("IS_FLAKY_CHECK", 0)) - ): - return FailureReason.SKIP - elif tags: for build_flag in args.build_flags: if "no-" + build_flag in tags: @@ -2295,6 +2285,7 @@ def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite, bool args, test_suite, client_options, server_logs_level ) test_result = test_case.process_result(test_result, MESSAGES) + break except TimeoutError: break finally: @@ -3280,12 +3271,6 @@ def parse_args(): default=False, help="Run tests over s3 storage", ) - parser.add_argument( - "--distributed-cache", - action="store_true", - default=False, - help="Run tests with enabled distributed cache", - ) parser.add_argument( "--azure-blob-storage", action="store_true", From 18a7a82458ce7ec3f12f7b6751699f119769ed55 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 5 Aug 2024 21:16:18 +0000 Subject: [PATCH 028/260] Better formatting --- tests/clickhouse-test | 33 +++++++++++---------------------- 1 file changed, 11 insertions(+), 22 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 5fcb9fb80f1..bcb8a12625b 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1950,15 +1950,11 @@ class TestSuite: ) -> Dict[str, Tuple[int, int]]: if not line.startswith(comment_sign): return {} - random_settings_limits_str = line[ - len(comment_sign) : - ].lstrip() # noqa: ignore E203 + random_settings_limits_str = line[len(comment_sign) :].lstrip() # noqa: ignore E203 random_settings_limits_prefix = "Random settings limits:" if not random_settings_limits_str.startswith(random_settings_limits_prefix): return {} - random_settings_limits_str = random_settings_limits_str[ - len(random_settings_limits_prefix) : - ] # noqa: ignore E203 + random_settings_limits_str = random_settings_limits_str[len(random_settings_limits_prefix) :] # noqa: ignore E203 # limits are specified in a form 'setting1=(min, max), setting2=(min,max), ...' random_settings_limits = re.findall( "([^=, ]+) *= *(\([^=]+\))", random_settings_limits_str @@ -1996,25 +1992,20 @@ class TestSuite: need_query_params = True except UnicodeDecodeError: pass - parsed_tags = parse_tags_from_line(tag_line, comment_sign) - if need_query_params: - parsed_tags.add("need-query-parameters") - random_settings_limits_line = next_line if parsed_tags else tag_line - random_settings_limits = parse_random_settings_limits_from_line( - random_settings_limits_line, comment_sign - ) + parsed_tags = parse_tags_from_line(tag_line, comment_sign) + if need_query_params: + parsed_tags.add("need-query-parameters") + random_settings_limits_line = next_line if parsed_tags else tag_line + random_settings_limits = parse_random_settings_limits_from_line( + random_settings_limits_line, comment_sign + ) return parsed_tags, random_settings_limits all_tags = {} all_random_settings_limits = {} start_time = datetime.now() for test_name in all_tests: - ( - tags, - random_settings_limits, - ) = load_tags_and_random_settings_limits_from_file( - os.path.join(suite_dir, test_name) - ) + tags, random_settings_limits = load_tags_and_random_settings_limits_from_file(os.path.join(suite_dir, test_name)) # noqa: ignore E203 if tags: all_tags[test_name] = tags if random_settings_limits: @@ -2061,9 +2052,7 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[ - str, Dict[str, (int, int)] - ] = all_tags_and_random_settings_limits[1] + self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = all_tags_and_random_settings_limits[1] # noqa: ignore E203 self.sequential_tests = [] self.parallel_tests = [] From 74a2976810b86086819ee8e6ee1f110ab1e70a37 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 6 Aug 2024 08:13:03 +0000 Subject: [PATCH 029/260] Fix pylint --- tests/clickhouse-test | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index bcb8a12625b..84f33860484 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1072,12 +1072,12 @@ class TestCase: def apply_random_settings_limits(self, random_settings): for setting in random_settings: if setting in self.random_settings_limits: - min = self.random_settings_limits[setting][0] - if min and random_settings[setting] < min: - random_settings[setting] = min - max = self.random_settings_limits[setting][1] - if max and random_settings[setting] > max: - random_settings[setting] = max + min_value = self.random_settings_limits[setting][0] + if min_value and random_settings[setting] < min_value: + random_settings[setting] = min_value + max_value = self.random_settings_limits[setting][1] + if max_value and random_settings[setting] > max_value: + random_settings[setting] = max_value def __init__(self, suite, case: str, args, is_concurrent: bool): self.case: str = case # case file name @@ -2005,7 +2005,7 @@ class TestSuite: all_random_settings_limits = {} start_time = datetime.now() for test_name in all_tests: - tags, random_settings_limits = load_tags_and_random_settings_limits_from_file(os.path.join(suite_dir, test_name)) # noqa: ignore E203 + tags, random_settings_limits = load_tags_and_random_settings_limits_from_file(os.path.join(suite_dir, test_name)) # noqa: ignore E203 if tags: all_tags[test_name] = tags if random_settings_limits: @@ -2052,7 +2052,7 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = all_tags_and_random_settings_limits[1] # noqa: ignore E203 + self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = all_tags_and_random_settings_limits[1] # noqa: ignore E203 self.sequential_tests = [] self.parallel_tests = [] From 5226792b1d8b4e110c63a813fb68c9dd65ea07b7 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 6 Aug 2024 08:48:06 +0000 Subject: [PATCH 030/260] Fix bad merge with master --- tests/clickhouse-test | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 84f33860484..c4124982442 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1163,6 +1163,9 @@ class TestCase: elif args.cloud and ("no-replicated-database" in tags): return FailureReason.REPLICATED_DB + elif tags and ("no-distributed-cache" in tags) and args.distributed_cache: + return FailureReason.DISTRIBUTED_CACHE + elif args.cloud and self.name in suite.cloud_skip_list: return FailureReason.NOT_SUPPORTED_IN_CLOUD @@ -1246,6 +1249,11 @@ class TestCase: ): return FailureReason.SKIP + elif "no-flaky-check" in tags and ( + 1 == int(os.environ.get("IS_FLAKY_CHECK", 0)) + ): + return FailureReason.SKIP + elif tags: for build_flag in args.build_flags: if "no-" + build_flag in tags: @@ -2274,7 +2282,6 @@ def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite, bool args, test_suite, client_options, server_logs_level ) test_result = test_case.process_result(test_result, MESSAGES) - break except TimeoutError: break finally: @@ -3260,6 +3267,12 @@ def parse_args(): default=False, help="Run tests over s3 storage", ) + parser.add_argument( + "--distributed-cache", + action="store_true", + default=False, + help="Run tests with enabled distributed cache", + ) parser.add_argument( "--azure-blob-storage", action="store_true", From bb33dca38470aba044da06938cc96ca55166262d Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 6 Aug 2024 08:49:08 +0000 Subject: [PATCH 031/260] Fix unrelated changes --- tests/clickhouse-test | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index c4124982442..72136404796 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -682,6 +682,7 @@ class FailureReason(enum.Enum): BUILD = "not running for current build" NO_PARALLEL_REPLICAS = "smth in not supported with parallel replicas" SHARED_MERGE_TREE = "no-shared-merge-tree" + DISTRIBUTED_CACHE = "distributed-cache" # UNKNOWN reasons NO_REFERENCE = "no reference file" @@ -1163,9 +1164,6 @@ class TestCase: elif args.cloud and ("no-replicated-database" in tags): return FailureReason.REPLICATED_DB - elif tags and ("no-distributed-cache" in tags) and args.distributed_cache: - return FailureReason.DISTRIBUTED_CACHE - elif args.cloud and self.name in suite.cloud_skip_list: return FailureReason.NOT_SUPPORTED_IN_CLOUD @@ -1212,6 +1210,9 @@ class TestCase: elif tags and ("no-replicated-database" in tags) and args.replicated_database: return FailureReason.REPLICATED_DB + elif tags and ("no-distributed-cache" in tags) and args.distributed_cache: + return FailureReason.DISTRIBUTED_CACHE + elif ( tags and ("atomic-database" in tags) @@ -1250,7 +1251,7 @@ class TestCase: return FailureReason.SKIP elif "no-flaky-check" in tags and ( - 1 == int(os.environ.get("IS_FLAKY_CHECK", 0)) + 1 == int(os.environ.get("IS_FLAKY_CHECK", 0)) ): return FailureReason.SKIP From 71c06b40cbf65abda49579bf5ac08e46575c7d29 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 6 Aug 2024 09:07:21 +0000 Subject: [PATCH 032/260] Avoid regexp --- docs/en/development/tests.md | 4 ++-- tests/clickhouse-test | 14 +++++++------- ...amic_merges_1_horizontal_compact_merge_tree.sql | 2 +- ...namic_merges_1_horizontal_compact_wide_tree.sql | 2 +- ...ynamic_merges_1_vertical_compact_merge_tree.sql | 2 +- ...7_dynamic_merges_1_vertical_wide_merge_tree.sql | 2 +- ...amic_merges_2_horizontal_compact_merge_tree.sql | 2 +- ...dynamic_merges_2_horizontal_wide_merge_tree.sql | 2 +- ...ynamic_merges_2_vertical_compact_merge_tree.sql | 2 +- ...7_dynamic_merges_2_vertical_wide_merge_tree.sql | 2 +- ...38_nested_dynamic_merges_compact_horizontal.sql | 2 +- ...3038_nested_dynamic_merges_compact_vertical.sql | 2 +- ...03038_nested_dynamic_merges_wide_horizontal.sql | 2 +- .../03038_nested_dynamic_merges_wide_vertical.sql | 2 +- 14 files changed, 21 insertions(+), 21 deletions(-) diff --git a/docs/en/development/tests.md b/docs/en/development/tests.md index f0afa983fec..bc9f85ef323 100644 --- a/docs/en/development/tests.md +++ b/docs/en/development/tests.md @@ -100,14 +100,14 @@ For `.sh` tests limits are written as a comment on the line next to tags or on t ```bash #!/usr/bin/env bash # Tags: no-fasttest -# Random settings limits: max_block_size=(1000, 10000), index_granularity=(100, None) +# Random settings limits: max_block_size=(1000, 10000); index_granularity=(100, None) ``` For `.sql` tests tags are placed as a SQL comment in the line next to tags or in the first line: ```sql -- Tags: no-fasttest --- Random settings limits: max_block_size=(1000, 10000), index_granularity=(100, None) +-- Random settings limits: max_block_size=(1000, 10000); index_granularity=(100, None) SELECT 1 ``` diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 72136404796..e5378e8c7f3 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1964,13 +1964,13 @@ class TestSuite: if not random_settings_limits_str.startswith(random_settings_limits_prefix): return {} random_settings_limits_str = random_settings_limits_str[len(random_settings_limits_prefix) :] # noqa: ignore E203 - # limits are specified in a form 'setting1=(min, max), setting2=(min,max), ...' - random_settings_limits = re.findall( - "([^=, ]+) *= *(\([^=]+\))", random_settings_limits_str - ) - random_settings_limits = { - pair[0]: make_tuple(pair[1]) for pair in random_settings_limits - } + # limits are specified in a form 'setting1=(min, max); setting2=(min,max); ...' + random_settings_limits = {} + for setting_and_limit in random_settings_limits_str.split(';'): + setting_and_limit = setting_and_limit.split('=') + random_settings_limits[setting_and_limit[0].strip()] = make_tuple( + setting_and_limit[1] + ) return random_settings_limits def is_shebang(line: str) -> bool: diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql index 7c2e7c3d2be..46f1c78b255 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type=1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql index aa62435188a..bf0c6ef0374 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type=1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql index bfc7bb9d206..fb82369a7a3 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type=1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql index 233667db0a7..c026bc04a56 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type=1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql index 71c6841515a..7f1934091f2 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql index 94ae1d867f5..f1f387fae9d 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql index 98ae230636a..cc11c454d38 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql index f8f5bd5d9e1..ffb2aca8b35 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_dynamic_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql index 13c1fd8b485..9ec4e4f949b 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql index daa95071cdb..ed4de931841 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql index dea7e7c0971..bd3c4b58a8f 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql index bf1323f2ea9..81bcda5443d 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql @@ -1,5 +1,5 @@ -- Tags: long, no-tsan, no-msan, no-ubsan, no-asan --- Random settings limits: index_granularity=(100, None), merge_max_block_size=(100, None) +-- Random settings limits: index_granularity=(100, None); merge_max_block_size=(100, None) set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; From 621f4bbf9e04b62628a9c053b3f39c6b8a67a52d Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Tue, 6 Aug 2024 14:13:20 +0200 Subject: [PATCH 033/260] Update SettingsChangesHistory.cpp --- src/Core/SettingsChangesHistory.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index f815a21b6a1..bb062deaab0 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -79,6 +79,7 @@ static std::initializer_list Date: Tue, 6 Aug 2024 15:01:10 +0200 Subject: [PATCH 034/260] Fix pylint --- tests/clickhouse-test | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index e5378e8c7f3..dea303ecdfb 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1086,7 +1086,7 @@ class TestCase: self.random_settings_limits = ( suite.all_random_settings_limits[case] if case in suite.all_random_settings_limits - else dict() + else {} ) for tag in os.getenv("GLOBAL_TAGS", "").split(","): From d124de847b44344d9346c4d1b76ada03b31c58c8 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 6 Aug 2024 16:06:59 +0000 Subject: [PATCH 035/260] Fix style --- tests/clickhouse-test | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index dea303ecdfb..c3b1d4d907c 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1959,15 +1959,17 @@ class TestSuite: ) -> Dict[str, Tuple[int, int]]: if not line.startswith(comment_sign): return {} - random_settings_limits_str = line[len(comment_sign) :].lstrip() # noqa: ignore E203 + random_settings_limits_str = line[len(comment_sign) :].lstrip() random_settings_limits_prefix = "Random settings limits:" if not random_settings_limits_str.startswith(random_settings_limits_prefix): return {} - random_settings_limits_str = random_settings_limits_str[len(random_settings_limits_prefix) :] # noqa: ignore E203 + random_settings_limits_str = random_settings_limits_str[ + len(random_settings_limits_prefix) : + ] # limits are specified in a form 'setting1=(min, max); setting2=(min,max); ...' random_settings_limits = {} - for setting_and_limit in random_settings_limits_str.split(';'): - setting_and_limit = setting_and_limit.split('=') + for setting_and_limit in random_settings_limits_str.split(";"): + setting_and_limit = setting_and_limit.split("=") random_settings_limits[setting_and_limit[0].strip()] = make_tuple( setting_and_limit[1] ) @@ -2014,7 +2016,12 @@ class TestSuite: all_random_settings_limits = {} start_time = datetime.now() for test_name in all_tests: - tags, random_settings_limits = load_tags_and_random_settings_limits_from_file(os.path.join(suite_dir, test_name)) # noqa: ignore E203 + ( + tags, + random_settings_limits, + ) = load_tags_and_random_settings_limits_from_file( + os.path.join(suite_dir, test_name) + ) # noqa: ignore E203 if tags: all_tags[test_name] = tags if random_settings_limits: @@ -2061,7 +2068,9 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = all_tags_and_random_settings_limits[1] # noqa: ignore E203 + self.all_random_settings_limits: Dict[ + str, Dict[str, (int, int)] + ] = all_tags_and_random_settings_limits[1] self.sequential_tests = [] self.parallel_tests = [] From 0ebe8e35511f764b61cb2428433132644f7deb96 Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Tue, 6 Aug 2024 18:38:23 +0200 Subject: [PATCH 036/260] Fix style --- src/Core/SettingsChangesHistory.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index e6949dd4fba..1ebc9b07748 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -79,7 +79,7 @@ static std::initializer_list Date: Wed, 7 Aug 2024 14:42:42 +0200 Subject: [PATCH 037/260] Fix style check --- tests/clickhouse-test | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index c3b1d4d907c..5946e561949 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -2068,10 +2068,9 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[ - str, Dict[str, (int, int)] - ] = all_tags_and_random_settings_limits[1] - + self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = ( + all_tags_and_random_settings_limits[1] + ) self.sequential_tests = [] self.parallel_tests = [] for test_name in self.all_tests: From f2731841de804c30ece1c75e84c8ca8d3eb62ef8 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 8 Aug 2024 20:20:21 +0200 Subject: [PATCH 038/260] init --- src/Core/callOnTypeIndex.h | 3 + src/DataTypes/getLeastSupertype.cpp | 41 +++++ src/DataTypes/getLeastSupertype.h | 22 +++ src/Functions/FunctionsConversion.cpp | 61 +++++++- ...23_interval_data_type_comparison.reference | 99 ++++++++++++ .../03223_interval_data_type_comparison.sql | 142 ++++++++++++++++++ 6 files changed, 365 insertions(+), 3 deletions(-) create mode 100644 tests/queries/0_stateless/03223_interval_data_type_comparison.reference create mode 100644 tests/queries/0_stateless/03223_interval_data_type_comparison.sql diff --git a/src/Core/callOnTypeIndex.h b/src/Core/callOnTypeIndex.h index f5f67df563b..ae5afce36be 100644 --- a/src/Core/callOnTypeIndex.h +++ b/src/Core/callOnTypeIndex.h @@ -3,6 +3,7 @@ #include #include +#include namespace DB @@ -212,6 +213,8 @@ static bool callOnIndexAndDataType(TypeIndex number, F && f, ExtraArgs && ... ar case TypeIndex::IPv4: return f(TypePair(), std::forward(args)...); case TypeIndex::IPv6: return f(TypePair(), std::forward(args)...); + case TypeIndex::Interval: return f(TypePair(), std::forward(args)...); + default: break; } diff --git a/src/DataTypes/getLeastSupertype.cpp b/src/DataTypes/getLeastSupertype.cpp index a71b19d6c92..0b9c744c091 100644 --- a/src/DataTypes/getLeastSupertype.cpp +++ b/src/DataTypes/getLeastSupertype.cpp @@ -228,6 +228,40 @@ void convertUInt64toInt64IfPossible(const DataTypes & types, TypeIndexSet & type } } +DataTypePtr findSmallestIntervalSuperType(const DataTypes &types, TypeIndexSet &types_set) +{ + const auto& granularity_map = getGranularityMap(); + int min_granularity = std::get<0>(granularity_map.at(IntervalKind::Kind::Year)); + DataTypePtr smallest_type; + + bool is_higher_interval = false; // For Years, Quarters and Months + + for (const auto &type : types) + { + if (const auto * interval_type = typeid_cast(type.get())) + { + int current_granularity = std::get<0>(granularity_map.at(interval_type->getKind())); + if (current_granularity > 8) + is_higher_interval = true; + if (current_granularity < min_granularity) + { + min_granularity = current_granularity; + smallest_type = type; + } + } + } + + if (is_higher_interval && min_granularity <= 8) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Cannot compare intervals {} and {} because the amount of days in month is not determined", types[0]->getName(), types[1]->getName()); + + if (smallest_type) + { + types_set.clear(); + types_set.insert(smallest_type->getTypeId()); + } + + return smallest_type; +} } template @@ -652,6 +686,13 @@ DataTypePtr getLeastSupertype(const DataTypes & types) return numeric_type; } + /// For interval data types. + { + auto res = findSmallestIntervalSuperType(types, type_ids); + if (res) + return res; + } + /// All other data types (UUID, AggregateFunction, Enum...) are compatible only if they are the same (checked in trivial cases). return throwOrReturn(types, "", ErrorCodes::NO_COMMON_TYPE); } diff --git a/src/DataTypes/getLeastSupertype.h b/src/DataTypes/getLeastSupertype.h index 2ae1e52ca96..c584eb83011 100644 --- a/src/DataTypes/getLeastSupertype.h +++ b/src/DataTypes/getLeastSupertype.h @@ -1,5 +1,7 @@ #pragma once #include +#include +#include namespace DB { @@ -48,4 +50,24 @@ DataTypePtr getLeastSupertypeOrString(const TypeIndexSet & types); DataTypePtr tryGetLeastSupertype(const TypeIndexSet & types); +/// A map that enumerated all interval kinds in ascending order with a conversion value to a next interval +inline const std::unordered_map> & getGranularityMap() +{ + static std::unordered_map> granularity_map = + { + {IntervalKind::Kind::Nanosecond, {1, 1000}}, + {IntervalKind::Kind::Microsecond, {2, 1000}}, + {IntervalKind::Kind::Millisecond, {3, 1000}}, + {IntervalKind::Kind::Second, {4, 60}}, + {IntervalKind::Kind::Minute, {5, 60}}, + {IntervalKind::Kind::Hour, {6, 24}}, + {IntervalKind::Kind::Day, {7, 7}}, + {IntervalKind::Kind::Week, {8, 4}}, + {IntervalKind::Kind::Month, {9, 3}}, + {IntervalKind::Kind::Quarter, {10, 4}}, + {IntervalKind::Kind::Year, {11, 1}} + }; + return granularity_map; +} + } diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 675283d011e..0ab1858dc97 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -45,6 +45,7 @@ #include #include #include +#include #include #include #include @@ -1573,6 +1574,55 @@ struct ConvertImpl arguments, result_type, input_rows_count, additions); } } + else if constexpr (std::is_same_v && std::is_same_v) + { + IntervalKind to = typeid_cast(result_type.get())->getKind(); + IntervalKind from = typeid_cast(arguments[0].type.get())->getKind(); + + if (from == to) + return arguments[0].column; + + const auto &map = getGranularityMap(); + Int64 conversion_factor = 1; + Int64 result_value; + + int from_position = map.at(from).first; + int to_position = map.at(to).first; // Positions of each interval according to granurality map + + if (from_position < to_position) + { + for (int i = from_position - 1; i <= to_position; ++i) + { + // Find the kind that matches this position + for (const auto &entry : map) + { + if (entry.second.first == i) + { + conversion_factor *= entry.second.second; + break; + } + } + } + result_value = arguments[0].column->getInt(0) / conversion_factor; + } + else + { + for (int i = from_position - 1; i >= to_position; --i) + { + for (const auto &entry : map) + { + if (entry.second.first == i) + { + conversion_factor *= entry.second.second; + break; + } + } + } + result_value = arguments[0].column->getInt(0) * conversion_factor; + } + + return ColumnConst::create(ColumnInt64::create(1, result_value), input_rows_count); + } else { using FromFieldType = typename FromDataType::FieldType; @@ -2181,7 +2231,7 @@ private: const DataTypePtr from_type = removeNullable(arguments[0].type); ColumnPtr result_column; - [[maybe_unused]] FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior; + FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior = default_date_time_overflow_behavior; if (context) date_time_overflow_behavior = context->getSettingsRef().date_time_overflow_behavior.value; @@ -2277,7 +2327,7 @@ private: } } else - result_column = ConvertImpl::execute(arguments, result_type, input_rows_count, from_string_tag); + result_column = ConvertImpl::execute(arguments, result_type, input_rows_count, from_string_tag); return true; }; @@ -2334,6 +2384,11 @@ private: else done = callOnIndexAndDataType(from_type->getTypeId(), call, BehaviourOnErrorFromString::ConvertDefaultBehaviorTag); } + + if constexpr (std::is_same_v) + { + done = callOnIndexAndDataType(from_type->getTypeId(), call, BehaviourOnErrorFromString::ConvertDefaultBehaviorTag); + } } if (!done) @@ -5224,7 +5279,7 @@ REGISTER_FUNCTION(Conversion) /// MySQL compatibility alias. Cannot be registered as alias, /// because we don't want it to be normalized to toDate in queries, /// otherwise CREATE DICTIONARY query breaks. - factory.registerFunction("DATE", &FunctionToDate::create, {}, FunctionFactory::Case::Insensitive); + factory.registerFunction("DATE", &FunctionToDate::create, {}, FunctionFactory::CaseInsensitive); factory.registerFunction(); factory.registerFunction(); diff --git a/tests/queries/0_stateless/03223_interval_data_type_comparison.reference b/tests/queries/0_stateless/03223_interval_data_type_comparison.reference new file mode 100644 index 00000000000..e98f792e4b2 --- /dev/null +++ b/tests/queries/0_stateless/03223_interval_data_type_comparison.reference @@ -0,0 +1,99 @@ +Comparing nanoseconds +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +Comparing microseconds +1 +1 +1 +1 +1 +1 +1 +0 +0 +0 +0 +0 +0 +0 +Comparing milliseconds +1 +1 +1 +1 +1 +1 +0 +0 +0 +0 +0 +0 +Comparing seconds +1 +1 +1 +1 +1 +0 +0 +0 +0 +0 +Comparing minutes +1 +1 +1 +1 +0 +0 +0 +0 +Comparing hours +1 +1 +1 +0 +0 +0 +Comparing days +1 +1 +0 +0 +Comparing weeks +1 +0 +Comparing months +1 +1 +1 +0 +0 +0 +Comparing quarters +1 +1 +0 +0 +Comparing years +1 +0 diff --git a/tests/queries/0_stateless/03223_interval_data_type_comparison.sql b/tests/queries/0_stateless/03223_interval_data_type_comparison.sql new file mode 100644 index 00000000000..6e4862bf2d2 --- /dev/null +++ b/tests/queries/0_stateless/03223_interval_data_type_comparison.sql @@ -0,0 +1,142 @@ +SELECT('Comparing nanoseconds'); +SELECT toIntervalNanosecond(500) > toIntervalNanosecond(300); +SELECT toIntervalNanosecond(1000) < toIntervalNanosecond(1500); +SELECT toIntervalNanosecond(2000) = toIntervalNanosecond(2000); +SELECT toIntervalNanosecond(1000) >= toIntervalMicrosecond(1); +SELECT toIntervalNanosecond(1000001) > toIntervalMillisecond(1); +SELECT toIntervalNanosecond(2000000001) > toIntervalSecond(2); +SELECT toIntervalNanosecond(60000000000) = toIntervalMinute(1); +SELECT toIntervalNanosecond(7199999999999) < toIntervalHour(2); +SELECT toIntervalNanosecond(1) < toIntervalDay(2); +SELECT toIntervalNanosecond(5) < toIntervalWeek(1); + +SELECT toIntervalNanosecond(500) < toIntervalNanosecond(300); +SELECT toIntervalNanosecond(1000) > toIntervalNanosecond(1500); +SELECT toIntervalNanosecond(2000) != toIntervalNanosecond(2000); +SELECT toIntervalNanosecond(1000) < toIntervalMicrosecond(1); +SELECT toIntervalNanosecond(1000001) < toIntervalMillisecond(1); +SELECT toIntervalNanosecond(2000000001) < toIntervalSecond(2); +SELECT toIntervalNanosecond(60000000000) != toIntervalMinute(1); +SELECT toIntervalNanosecond(7199999999999) > toIntervalHour(2); +SELECT toIntervalNanosecond(1) > toIntervalDay(2); +SELECT toIntervalNanosecond(5) > toIntervalWeek(1); + +SELECT toIntervalNanosecond(1) < toIntervalMonth(2); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing microseconds'); +SELECT toIntervalMicrosecond(1) < toIntervalMicrosecond(999); +SELECT toIntervalMicrosecond(1001) > toIntervalMillisecond(1); +SELECT toIntervalMicrosecond(2000000) = toIntervalSecond(2); +SELECT toIntervalMicrosecond(179999999) < toIntervalMinute(3); +SELECT toIntervalMicrosecond(3600000000) = toIntervalHour(1); +SELECT toIntervalMicrosecond(36000000000000) > toIntervalDay(2); +SELECT toIntervalMicrosecond(1209600000000) = toIntervalWeek(2); + +SELECT toIntervalMicrosecond(1) > toIntervalMicrosecond(999); +SELECT toIntervalMicrosecond(1001) < toIntervalMillisecond(1); +SELECT toIntervalMicrosecond(2000000) != toIntervalSecond(2); +SELECT toIntervalMicrosecond(179999999) > toIntervalMinute(3); +SELECT toIntervalMicrosecond(3600000000) != toIntervalHour(1); +SELECT toIntervalMicrosecond(36000000000000) < toIntervalDay(2); +SELECT toIntervalMicrosecond(1209600000000) != toIntervalWeek(2); + +SELECT toIntervalMicrosecond(36000000000000) < toIntervalQuarter(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing milliseconds'); +SELECT toIntervalMillisecond(2000) > toIntervalMillisecond(2); +SELECT toIntervalMillisecond(2000) = toIntervalSecond(2); +SELECT toIntervalMillisecond(170000) < toIntervalMinute(3); +SELECT toIntervalMillisecond(144000001) > toIntervalHour(40); +SELECT toIntervalMillisecond(1728000000) = toIntervalDay(20); +SELECT toIntervalMillisecond(1198599999) < toIntervalWeek(2); + +SELECT toIntervalMillisecond(2000) < toIntervalMillisecond(2); +SELECT toIntervalMillisecond(2000) != toIntervalSecond(2); +SELECT toIntervalMillisecond(170000) > toIntervalMinute(3); +SELECT toIntervalMillisecond(144000001) < toIntervalHour(40); +SELECT toIntervalMillisecond(1728000000) != toIntervalDay(20); +SELECT toIntervalMillisecond(1198599999) > toIntervalWeek(2); + +SELECT toIntervalMillisecond(36000000000000) < toIntervalYear(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing seconds'); +SELECT toIntervalSecond(120) > toIntervalSecond(2); +SELECT toIntervalSecond(120) = toIntervalMinute(2); +SELECT toIntervalSecond(1) < toIntervalHour(2); +SELECT toIntervalSecond(86401) >= toIntervalDay(1); +SELECT toIntervalSecond(1209600) = toIntervalWeek(2); + +SELECT toIntervalSecond(120) < toIntervalSecond(2); +SELECT toIntervalSecond(120) != toIntervalMinute(2); +SELECT toIntervalSecond(1) > toIntervalHour(2); +SELECT toIntervalSecond(86401) < toIntervalDay(1); +SELECT toIntervalSecond(1209600) != toIntervalWeek(2); + +SELECT toIntervalSecond(36000000000000) < toIntervalMonth(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing minutes'); +SELECT toIntervalMinute(1) < toIntervalMinute(59); +SELECT toIntervalMinute(1) < toIntervalHour(59); +SELECT toIntervalMinute(1440) = toIntervalDay(1); +SELECT toIntervalMinute(30241) > toIntervalWeek(3); + +SELECT toIntervalMinute(1) > toIntervalMinute(59); +SELECT toIntervalMinute(1) > toIntervalHour(59); +SELECT toIntervalMinute(1440) != toIntervalDay(1); +SELECT toIntervalMinute(30241) < toIntervalWeek(3); + +SELECT toIntervalMinute(2) = toIntervalQuarter(120); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing hours'); +SELECT toIntervalHour(48) > toIntervalHour(2); +SELECT toIntervalHour(48) >= toIntervalDay(2); +SELECT toIntervalHour(672) = toIntervalWeek(4); + +SELECT toIntervalHour(48) < toIntervalHour(2); +SELECT toIntervalHour(48) < toIntervalDay(2); +SELECT toIntervalHour(672) != toIntervalWeek(4); + +SELECT toIntervalHour(2) < toIntervalYear(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing days'); +SELECT toIntervalDay(1) < toIntervalDay(23); +SELECT toIntervalDay(25) > toIntervalWeek(3); + +SELECT toIntervalDay(1) > toIntervalDay(23); +SELECT toIntervalDay(25) < toIntervalWeek(3); + +SELECT toIntervalDay(2) = toIntervalMonth(48); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing weeks'); +SELECT toIntervalWeek(1) < toIntervalWeek(6); + +SELECT toIntervalWeek(1) > toIntervalWeek(6); + +SELECT toIntervalWeek(124) > toIntervalQuarter(8); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing months'); +SELECT toIntervalMonth(1) < toIntervalMonth(3); +SELECT toIntervalMonth(124) > toIntervalQuarter(5); +SELECT toIntervalMonth(36) = toIntervalYear(3); + +SELECT toIntervalMonth(1) > toIntervalMonth(3); +SELECT toIntervalMonth(124) < toIntervalQuarter(5); +SELECT toIntervalMonth(36) != toIntervalYear(3); + +SELECT toIntervalMonth(6) = toIntervalMicrosecond(26); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing quarters'); +SELECT toIntervalQuarter(5) > toIntervalQuarter(4); +SELECT toIntervalQuarter(20) = toIntervalYear(5); + +SELECT toIntervalQuarter(5) < toIntervalQuarter(4); +SELECT toIntervalQuarter(20) != toIntervalYear(5); + +SELECT toIntervalQuarter(2) = toIntervalNanosecond(6); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +SELECT('Comparing years'); +SELECT toIntervalYear(1) < toIntervalYear(3); + +SELECT toIntervalYear(1) > toIntervalYear(3); + +SELECT toIntervalYear(2) = toIntervalSecond(8); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } From e9659626adc29d237d23e0f3ced9c8712d472a73 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 8 Aug 2024 20:41:15 +0200 Subject: [PATCH 039/260] fix style + add docs --- .../data-types/special-data-types/interval.md | 21 +++++++++---------- .../data-types/special-data-types/interval.md | 21 +++++++++---------- .../data-types/special-data-types/interval.md | 21 +++++++++---------- src/DataTypes/getLeastSupertype.cpp | 2 +- 4 files changed, 31 insertions(+), 34 deletions(-) diff --git a/docs/en/sql-reference/data-types/special-data-types/interval.md b/docs/en/sql-reference/data-types/special-data-types/interval.md index bedbcf0bd28..be26053580b 100644 --- a/docs/en/sql-reference/data-types/special-data-types/interval.md +++ b/docs/en/sql-reference/data-types/special-data-types/interval.md @@ -53,29 +53,28 @@ SELECT now() as current_date_time, current_date_time + INTERVAL 4 DAY └─────────────────────┴───────────────────────────────┘ ``` -Intervals with different types can’t be combined. You can’t use intervals like `4 DAY 1 HOUR`. Specify intervals in units that are smaller or equal to the smallest unit of the interval, for example, the interval `1 day and an hour` interval can be expressed as `25 HOUR` or `90000 SECOND`. - -You can’t perform arithmetical operations with `Interval`-type values, but you can add intervals of different types consequently to values in `Date` or `DateTime` data types. For example: +Also it is possible to use multiple intervals simultaneously: ``` sql -SELECT now() AS current_date_time, current_date_time + INTERVAL 4 DAY + INTERVAL 3 HOUR +SELECT now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) ``` ``` text -┌───current_date_time─┬─plus(plus(now(), toIntervalDay(4)), toIntervalHour(3))─┐ -│ 2019-10-23 11:16:28 │ 2019-10-27 14:16:28 │ -└─────────────────────┴────────────────────────────────────────────────────────┘ +┌───current_date_time─┬─plus(current_date_time, plus(toIntervalDay(4), toIntervalHour(3)))─┐ +│ 2024-08-08 18:31:39 │ 2024-08-12 21:31:39 │ +└─────────────────────┴────────────────────────────────────────────────────────────────────┘ ``` -The following query causes an exception: +And to compare values with different intevals: ``` sql -select now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) +SELECT toIntervalMicrosecond(3600000000) = toIntervalHour(1); ``` ``` text -Received exception from server (version 19.14.1): -Code: 43. DB::Exception: Received from localhost:9000. DB::Exception: Wrong argument types for function plus: if one argument is Interval, then another must be Date or DateTime.. +┌─less(toIntervalMicrosecond(179999999), toIntervalMinute(3))─┐ +│ 1 │ +└─────────────────────────────────────────────────────────────┘ ``` ## See Also diff --git a/docs/ru/sql-reference/data-types/special-data-types/interval.md b/docs/ru/sql-reference/data-types/special-data-types/interval.md index 867a6665f4b..5064391f582 100644 --- a/docs/ru/sql-reference/data-types/special-data-types/interval.md +++ b/docs/ru/sql-reference/data-types/special-data-types/interval.md @@ -54,29 +54,28 @@ SELECT now() as current_date_time, current_date_time + INTERVAL 4 DAY └─────────────────────┴───────────────────────────────┘ ``` -Нельзя объединять интервалы различных типов. Нельзя использовать интервалы вида `4 DAY 1 HOUR`. Вместо этого выражайте интервал в единицах меньших или равных минимальной единице интервала, например, интервал «1 день и 1 час» можно выразить как `25 HOUR` или `90000 SECOND`. - -Арифметические операции со значениями типов `Interval` не доступны, однако можно последовательно добавлять различные интервалы к значениям типов `Date` и `DateTime`. Например: +Также можно использовать различные типы интервалов одновременно: ``` sql -SELECT now() AS current_date_time, current_date_time + INTERVAL 4 DAY + INTERVAL 3 HOUR +SELECT now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) ``` ``` text -┌───current_date_time─┬─plus(plus(now(), toIntervalDay(4)), toIntervalHour(3))─┐ -│ 2019-10-23 11:16:28 │ 2019-10-27 14:16:28 │ -└─────────────────────┴────────────────────────────────────────────────────────┘ +┌───current_date_time─┬─plus(current_date_time, plus(toIntervalDay(4), toIntervalHour(3)))─┐ +│ 2024-08-08 18:31:39 │ 2024-08-12 21:31:39 │ +└─────────────────────┴────────────────────────────────────────────────────────────────────┘ ``` -Следующий запрос приведёт к генерированию исключения: +И сравнивать значения из разными интервалами: ``` sql -select now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) +SELECT toIntervalMicrosecond(3600000000) = toIntervalHour(1); ``` ``` text -Received exception from server (version 19.14.1): -Code: 43. DB::Exception: Received from localhost:9000. DB::Exception: Wrong argument types for function plus: if one argument is Interval, then another must be Date or DateTime.. +┌─less(toIntervalMicrosecond(179999999), toIntervalMinute(3))─┐ +│ 1 │ +└─────────────────────────────────────────────────────────────┘ ``` ## Смотрите также {#smotrite-takzhe} diff --git a/docs/zh/sql-reference/data-types/special-data-types/interval.md b/docs/zh/sql-reference/data-types/special-data-types/interval.md index e05869b2df8..e16f6d5f84f 100644 --- a/docs/zh/sql-reference/data-types/special-data-types/interval.md +++ b/docs/zh/sql-reference/data-types/special-data-types/interval.md @@ -55,29 +55,28 @@ SELECT now() as current_date_time, current_date_time + INTERVAL 4 DAY └─────────────────────┴───────────────────────────────┘ ``` -不同类型的间隔不能合并。 你不能使用诸如 `4 DAY 1 HOUR` 的时间间隔. 以小于或等于时间间隔最小单位的单位来指定间隔，例如，时间间隔 `1 day and an hour` 可以表示为 `25 HOUR` 或 `90000 SECOND`. - -你不能对 `Interval` 类型的值执行算术运算，但你可以向 `Date` 或 `DateTime` 数据类型的值添加不同类型的时间间隔，例如: +也可以同時使用多個間隔： ``` sql -SELECT now() AS current_date_time, current_date_time + INTERVAL 4 DAY + INTERVAL 3 HOUR +SELECT now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) ``` ``` text -┌───current_date_time─┬─plus(plus(now(), toIntervalDay(4)), toIntervalHour(3))─┐ -│ 2019-10-23 11:16:28 │ 2019-10-27 14:16:28 │ -└─────────────────────┴────────────────────────────────────────────────────────┘ +┌───current_date_time─┬─plus(current_date_time, plus(toIntervalDay(4), toIntervalHour(3)))─┐ +│ 2024-08-08 18:31:39 │ 2024-08-12 21:31:39 │ +└─────────────────────┴────────────────────────────────────────────────────────────────────┘ ``` -以下查询将导致异常: +並比較不同直數的值： ``` sql -select now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) +SELECT toIntervalMicrosecond(3600000000) = toIntervalHour(1); ``` ``` text -Received exception from server (version 19.14.1): -Code: 43. DB::Exception: Received from localhost:9000. DB::Exception: Wrong argument types for function plus: if one argument is Interval, then another must be Date or DateTime.. +┌─less(toIntervalMicrosecond(179999999), toIntervalMinute(3))─┐ +│ 1 │ +└─────────────────────────────────────────────────────────────┘ ``` ## 另请参阅 {#see-also} diff --git a/src/DataTypes/getLeastSupertype.cpp b/src/DataTypes/getLeastSupertype.cpp index 0b9c744c091..674284460dc 100644 --- a/src/DataTypes/getLeastSupertype.cpp +++ b/src/DataTypes/getLeastSupertype.cpp @@ -252,7 +252,7 @@ DataTypePtr findSmallestIntervalSuperType(const DataTypes &types, TypeIndexSet & } if (is_higher_interval && min_granularity <= 8) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Cannot compare intervals {} and {} because the amount of days in month is not determined", types[0]->getName(), types[1]->getName()); + throw Exception(ErrorCodes::NO_COMMON_TYPE, "Cannot compare intervals {} and {} because the amount of days in month is not determined", types[0]->getName(), types[1]->getName()); if (smallest_type) { From 0ad6aa09acb72a67fc88e0cd8186afd32fefd6bf Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 8 Aug 2024 22:51:27 +0200 Subject: [PATCH 040/260] fix style --- docs/en/sql-reference/data-types/special-data-types/interval.md | 2 +- src/Functions/FunctionsConversion.cpp | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/en/sql-reference/data-types/special-data-types/interval.md b/docs/en/sql-reference/data-types/special-data-types/interval.md index be26053580b..4ef1a7e6238 100644 --- a/docs/en/sql-reference/data-types/special-data-types/interval.md +++ b/docs/en/sql-reference/data-types/special-data-types/interval.md @@ -65,7 +65,7 @@ SELECT now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVA └─────────────────────┴────────────────────────────────────────────────────────────────────┘ ``` -And to compare values with different intevals: +And to compare values with different intervals: ``` sql SELECT toIntervalMicrosecond(3600000000) = toIntervalHour(1); diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 0ab1858dc97..1708991af74 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -1606,7 +1606,7 @@ struct ConvertImpl result_value = arguments[0].column->getInt(0) / conversion_factor; } else - { + { for (int i = from_position - 1; i >= to_position; --i) { for (const auto &entry : map) From 94efbb0bf9ab62a5399d4918e7bcfd358421a879 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 8 Aug 2024 23:26:24 +0200 Subject: [PATCH 041/260] fix build --- src/Functions/FunctionsConversion.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 1708991af74..43ebe573582 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -5279,7 +5279,7 @@ REGISTER_FUNCTION(Conversion) /// MySQL compatibility alias. Cannot be registered as alias, /// because we don't want it to be normalized to toDate in queries, /// otherwise CREATE DICTIONARY query breaks. - factory.registerFunction("DATE", &FunctionToDate::create, {}, FunctionFactory::CaseInsensitive); + factory.registerFunction("DATE", &FunctionToDate::create, {}, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); From b4c553718353eb2302f85ea4d096a92036ce832c Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 8 Aug 2024 23:49:56 +0200 Subject: [PATCH 042/260] fix errorcodes in test --- .../03223_interval_data_type_comparison.sql | 22 +++++++++---------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/tests/queries/0_stateless/03223_interval_data_type_comparison.sql b/tests/queries/0_stateless/03223_interval_data_type_comparison.sql index 6e4862bf2d2..5d01addae45 100644 --- a/tests/queries/0_stateless/03223_interval_data_type_comparison.sql +++ b/tests/queries/0_stateless/03223_interval_data_type_comparison.sql @@ -21,7 +21,7 @@ SELECT toIntervalNanosecond(7199999999999) > toIntervalHour(2); SELECT toIntervalNanosecond(1) > toIntervalDay(2); SELECT toIntervalNanosecond(5) > toIntervalWeek(1); -SELECT toIntervalNanosecond(1) < toIntervalMonth(2); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalNanosecond(1) < toIntervalMonth(2); -- { serverError NO_COMMON_TYPE } SELECT('Comparing microseconds'); SELECT toIntervalMicrosecond(1) < toIntervalMicrosecond(999); @@ -40,7 +40,7 @@ SELECT toIntervalMicrosecond(3600000000) != toIntervalHour(1); SELECT toIntervalMicrosecond(36000000000000) < toIntervalDay(2); SELECT toIntervalMicrosecond(1209600000000) != toIntervalWeek(2); -SELECT toIntervalMicrosecond(36000000000000) < toIntervalQuarter(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalMicrosecond(36000000000000) < toIntervalQuarter(1); -- { serverError NO_COMMON_TYPE } SELECT('Comparing milliseconds'); SELECT toIntervalMillisecond(2000) > toIntervalMillisecond(2); @@ -57,7 +57,7 @@ SELECT toIntervalMillisecond(144000001) < toIntervalHour(40); SELECT toIntervalMillisecond(1728000000) != toIntervalDay(20); SELECT toIntervalMillisecond(1198599999) > toIntervalWeek(2); -SELECT toIntervalMillisecond(36000000000000) < toIntervalYear(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalMillisecond(36000000000000) < toIntervalYear(1); -- { serverError NO_COMMON_TYPE } SELECT('Comparing seconds'); SELECT toIntervalSecond(120) > toIntervalSecond(2); @@ -72,7 +72,7 @@ SELECT toIntervalSecond(1) > toIntervalHour(2); SELECT toIntervalSecond(86401) < toIntervalDay(1); SELECT toIntervalSecond(1209600) != toIntervalWeek(2); -SELECT toIntervalSecond(36000000000000) < toIntervalMonth(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalSecond(36000000000000) < toIntervalMonth(1); -- { serverError NO_COMMON_TYPE } SELECT('Comparing minutes'); SELECT toIntervalMinute(1) < toIntervalMinute(59); @@ -85,7 +85,7 @@ SELECT toIntervalMinute(1) > toIntervalHour(59); SELECT toIntervalMinute(1440) != toIntervalDay(1); SELECT toIntervalMinute(30241) < toIntervalWeek(3); -SELECT toIntervalMinute(2) = toIntervalQuarter(120); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalMinute(2) = toIntervalQuarter(120); -- { serverError NO_COMMON_TYPE } SELECT('Comparing hours'); SELECT toIntervalHour(48) > toIntervalHour(2); @@ -96,7 +96,7 @@ SELECT toIntervalHour(48) < toIntervalHour(2); SELECT toIntervalHour(48) < toIntervalDay(2); SELECT toIntervalHour(672) != toIntervalWeek(4); -SELECT toIntervalHour(2) < toIntervalYear(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalHour(2) < toIntervalYear(1); -- { serverError NO_COMMON_TYPE } SELECT('Comparing days'); SELECT toIntervalDay(1) < toIntervalDay(23); @@ -105,14 +105,14 @@ SELECT toIntervalDay(25) > toIntervalWeek(3); SELECT toIntervalDay(1) > toIntervalDay(23); SELECT toIntervalDay(25) < toIntervalWeek(3); -SELECT toIntervalDay(2) = toIntervalMonth(48); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalDay(2) = toIntervalMonth(48); -- { serverError NO_COMMON_TYPE } SELECT('Comparing weeks'); SELECT toIntervalWeek(1) < toIntervalWeek(6); SELECT toIntervalWeek(1) > toIntervalWeek(6); -SELECT toIntervalWeek(124) > toIntervalQuarter(8); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalWeek(124) > toIntervalQuarter(8); -- { serverError NO_COMMON_TYPE } SELECT('Comparing months'); SELECT toIntervalMonth(1) < toIntervalMonth(3); @@ -123,7 +123,7 @@ SELECT toIntervalMonth(1) > toIntervalMonth(3); SELECT toIntervalMonth(124) < toIntervalQuarter(5); SELECT toIntervalMonth(36) != toIntervalYear(3); -SELECT toIntervalMonth(6) = toIntervalMicrosecond(26); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalMonth(6) = toIntervalMicrosecond(26); -- { serverError NO_COMMON_TYPE } SELECT('Comparing quarters'); SELECT toIntervalQuarter(5) > toIntervalQuarter(4); @@ -132,11 +132,11 @@ SELECT toIntervalQuarter(20) = toIntervalYear(5); SELECT toIntervalQuarter(5) < toIntervalQuarter(4); SELECT toIntervalQuarter(20) != toIntervalYear(5); -SELECT toIntervalQuarter(2) = toIntervalNanosecond(6); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalQuarter(2) = toIntervalNanosecond(6); -- { serverError NO_COMMON_TYPE } SELECT('Comparing years'); SELECT toIntervalYear(1) < toIntervalYear(3); SELECT toIntervalYear(1) > toIntervalYear(3); -SELECT toIntervalYear(2) = toIntervalSecond(8); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT toIntervalYear(2) = toIntervalSecond(8); -- { serverError NO_COMMON_TYPE } From 3357275fa8c55bcc5371b4ff9c9a5d80e51ab689 Mon Sep 17 00:00:00 2001 From: JackyWoo Date: Fri, 9 Aug 2024 18:33:45 +0800 Subject: [PATCH 043/260] Fix MSAN issue caused by incorrect date format. --- src/IO/ReadHelpers.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/IO/ReadHelpers.cpp b/src/IO/ReadHelpers.cpp index c771fced73a..dd4aef23a25 100644 --- a/src/IO/ReadHelpers.cpp +++ b/src/IO/ReadHelpers.cpp @@ -1432,7 +1432,7 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D s_pos[size] = 0; if constexpr (throw_exception) - throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse time component of DateTime {}", s); + throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse time component of DateTime {}", String(s, date_broken_down_length + 1 + size)); else return false; } From 35f19522e745ef2267b4c6f99dfc5d7c1f7e78c3 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Fri, 9 Aug 2024 12:56:14 +0200 Subject: [PATCH 044/260] fix fuzzer --- src/Functions/FunctionsConversion.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 43ebe573582..c25bc44450f 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -1579,7 +1579,7 @@ struct ConvertImpl IntervalKind to = typeid_cast(result_type.get())->getKind(); IntervalKind from = typeid_cast(arguments[0].type.get())->getKind(); - if (from == to) + if (from == to || arguments[0].column->empty()) return arguments[0].column; const auto &map = getGranularityMap(); From ca4041847e4aa8acccd6ea31c0a18f2160c0dc7a Mon Sep 17 00:00:00 2001 From: JackyWoo Date: Fri, 9 Aug 2024 19:15:41 +0800 Subject: [PATCH 045/260] Add tests --- src/IO/ReadHelpers.cpp | 4 ++-- ...215_fix_datetime_implicit_conversion.reference | 1 + .../03215_fix_datetime_implicit_conversion.sql | 15 +++++++++++++++ 3 files changed, 18 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.reference create mode 100644 tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.sql diff --git a/src/IO/ReadHelpers.cpp b/src/IO/ReadHelpers.cpp index dd4aef23a25..e69b4187b37 100644 --- a/src/IO/ReadHelpers.cpp +++ b/src/IO/ReadHelpers.cpp @@ -1402,7 +1402,7 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D s_pos[size] = 0; if constexpr (throw_exception) - throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse DateTime {}", s); + throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse DateTime {}", String(s, already_read_length)); else return false; } @@ -1432,7 +1432,7 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D s_pos[size] = 0; if constexpr (throw_exception) - throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse time component of DateTime {}", String(s, date_broken_down_length + 1 + size)); + throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse time component of DateTime {}", String(s, size)); else return false; } diff --git a/tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.reference b/tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.reference new file mode 100644 index 00000000000..573541ac970 --- /dev/null +++ b/tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.reference @@ -0,0 +1 @@ +0 diff --git a/tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.sql b/tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.sql new file mode 100644 index 00000000000..70a8a3432a6 --- /dev/null +++ b/tests/queries/0_stateless/03215_fix_datetime_implicit_conversion.sql @@ -0,0 +1,15 @@ +DROP TABLE IF EXISTS tab SYNC; + +CREATE TABLE tab +( + a DateTime, + pk String +) Engine = MergeTree() ORDER BY pk; + +INSERT INTO tab select cast(number, 'DateTime'), generateUUIDv4() FROM system.numbers LIMIT 1; + +SELECT count(*) FROM tab WHERE a = '2024-08-06 09:58:09'; +SELECT count(*) FROM tab WHERE a = '2024-08-06 09:58:0'; -- { serverError CANNOT_PARSE_DATETIME } +SELECT count(*) FROM tab WHERE a = '2024-08-0 09:58:09'; -- { serverError TYPE_MISMATCH } + +DROP TABLE IF EXISTS tab SYNC; From 6ded5e1c8b994ad2332468e605b17a74e8d5675f Mon Sep 17 00:00:00 2001 From: JackyWoo Date: Fri, 9 Aug 2024 23:50:03 +0800 Subject: [PATCH 046/260] Some fixups --- src/IO/ReadHelpers.cpp | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/src/IO/ReadHelpers.cpp b/src/IO/ReadHelpers.cpp index e69b4187b37..b484f80250d 100644 --- a/src/IO/ReadHelpers.cpp +++ b/src/IO/ReadHelpers.cpp @@ -1399,10 +1399,8 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D size_t size = buf.read(s_pos, remaining_date_size); if (size != remaining_date_size) { - s_pos[size] = 0; - if constexpr (throw_exception) - throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse DateTime {}", String(s, already_read_length)); + throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse DateTime {}", std::string_view(s, already_read_length + size)); else return false; } @@ -1429,10 +1427,8 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D if (size != time_broken_down_length) { - s_pos[size] = 0; - if constexpr (throw_exception) - throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse time component of DateTime {}", String(s, size)); + throw Exception(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot parse time component of DateTime {}", std::string_view(s, size)); else return false; } From a3d8db6e1eb27d6a8fa81bbf43c8ffb171714c0b Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Fri, 9 Aug 2024 19:05:37 +0200 Subject: [PATCH 047/260] updates due to review --- .../data-types/special-data-types/interval.md | 21 +++++++------- src/DataTypes/getLeastSupertype.cpp | 13 ++++----- src/DataTypes/getLeastSupertype.h | 21 ++------------ src/Functions/FunctionsConversion.cpp | 28 +++---------------- 4 files changed, 24 insertions(+), 59 deletions(-) diff --git a/docs/zh/sql-reference/data-types/special-data-types/interval.md b/docs/zh/sql-reference/data-types/special-data-types/interval.md index e16f6d5f84f..e05869b2df8 100644 --- a/docs/zh/sql-reference/data-types/special-data-types/interval.md +++ b/docs/zh/sql-reference/data-types/special-data-types/interval.md @@ -55,28 +55,29 @@ SELECT now() as current_date_time, current_date_time + INTERVAL 4 DAY └─────────────────────┴───────────────────────────────┘ ``` -也可以同時使用多個間隔： +不同类型的间隔不能合并。 你不能使用诸如 `4 DAY 1 HOUR` 的时间间隔. 以小于或等于时间间隔最小单位的单位来指定间隔，例如，时间间隔 `1 day and an hour` 可以表示为 `25 HOUR` 或 `90000 SECOND`. + +你不能对 `Interval` 类型的值执行算术运算，但你可以向 `Date` 或 `DateTime` 数据类型的值添加不同类型的时间间隔，例如: ``` sql -SELECT now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) +SELECT now() AS current_date_time, current_date_time + INTERVAL 4 DAY + INTERVAL 3 HOUR ``` ``` text -┌───current_date_time─┬─plus(current_date_time, plus(toIntervalDay(4), toIntervalHour(3)))─┐ -│ 2024-08-08 18:31:39 │ 2024-08-12 21:31:39 │ -└─────────────────────┴────────────────────────────────────────────────────────────────────┘ +┌───current_date_time─┬─plus(plus(now(), toIntervalDay(4)), toIntervalHour(3))─┐ +│ 2019-10-23 11:16:28 │ 2019-10-27 14:16:28 │ +└─────────────────────┴────────────────────────────────────────────────────────┘ ``` -並比較不同直數的值： +以下查询将导致异常: ``` sql -SELECT toIntervalMicrosecond(3600000000) = toIntervalHour(1); +select now() AS current_date_time, current_date_time + (INTERVAL 4 DAY + INTERVAL 3 HOUR) ``` ``` text -┌─less(toIntervalMicrosecond(179999999), toIntervalMinute(3))─┐ -│ 1 │ -└─────────────────────────────────────────────────────────────┘ +Received exception from server (version 19.14.1): +Code: 43. DB::Exception: Received from localhost:9000. DB::Exception: Wrong argument types for function plus: if one argument is Interval, then another must be Date or DateTime.. ``` ## 另请参阅 {#see-also} diff --git a/src/DataTypes/getLeastSupertype.cpp b/src/DataTypes/getLeastSupertype.cpp index 674284460dc..8bcec49815f 100644 --- a/src/DataTypes/getLeastSupertype.cpp +++ b/src/DataTypes/getLeastSupertype.cpp @@ -230,8 +230,7 @@ void convertUInt64toInt64IfPossible(const DataTypes & types, TypeIndexSet & type DataTypePtr findSmallestIntervalSuperType(const DataTypes &types, TypeIndexSet &types_set) { - const auto& granularity_map = getGranularityMap(); - int min_granularity = std::get<0>(granularity_map.at(IntervalKind::Kind::Year)); + auto min_interval = IntervalKind::Kind::Year; DataTypePtr smallest_type; bool is_higher_interval = false; // For Years, Quarters and Months @@ -240,18 +239,18 @@ DataTypePtr findSmallestIntervalSuperType(const DataTypes &types, TypeIndexSet & { if (const auto * interval_type = typeid_cast(type.get())) { - int current_granularity = std::get<0>(granularity_map.at(interval_type->getKind())); - if (current_granularity > 8) + auto current_interval = interval_type->getKind().kind; + if (current_interval > IntervalKind::Kind::Week) is_higher_interval = true; - if (current_granularity < min_granularity) + if (current_interval < min_interval) { - min_granularity = current_granularity; + min_interval = current_interval; smallest_type = type; } } } - if (is_higher_interval && min_granularity <= 8) + if (is_higher_interval && min_interval <= IntervalKind::Kind::Week) throw Exception(ErrorCodes::NO_COMMON_TYPE, "Cannot compare intervals {} and {} because the amount of days in month is not determined", types[0]->getName(), types[1]->getName()); if (smallest_type) diff --git a/src/DataTypes/getLeastSupertype.h b/src/DataTypes/getLeastSupertype.h index c584eb83011..5ea2b6417b2 100644 --- a/src/DataTypes/getLeastSupertype.h +++ b/src/DataTypes/getLeastSupertype.h @@ -50,24 +50,9 @@ DataTypePtr getLeastSupertypeOrString(const TypeIndexSet & types); DataTypePtr tryGetLeastSupertype(const TypeIndexSet & types); -/// A map that enumerated all interval kinds in ascending order with a conversion value to a next interval -inline const std::unordered_map> & getGranularityMap() -{ - static std::unordered_map> granularity_map = - { - {IntervalKind::Kind::Nanosecond, {1, 1000}}, - {IntervalKind::Kind::Microsecond, {2, 1000}}, - {IntervalKind::Kind::Millisecond, {3, 1000}}, - {IntervalKind::Kind::Second, {4, 60}}, - {IntervalKind::Kind::Minute, {5, 60}}, - {IntervalKind::Kind::Hour, {6, 24}}, - {IntervalKind::Kind::Day, {7, 7}}, - {IntervalKind::Kind::Week, {8, 4}}, - {IntervalKind::Kind::Month, {9, 3}}, - {IntervalKind::Kind::Quarter, {10, 4}}, - {IntervalKind::Kind::Year, {11, 1}} - }; - return granularity_map; +/// A vector that shows the conversion rates to the next Interval type starting from NanoSecond +static std::vector interval_conversions = {1000, 1000, 1000, 60, 60, 24, 7, 4, 3, 4, 1}; + } } diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index c25bc44450f..25c6bbcbfef 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -1582,42 +1582,22 @@ struct ConvertImpl if (from == to || arguments[0].column->empty()) return arguments[0].column; - const auto &map = getGranularityMap(); Int64 conversion_factor = 1; Int64 result_value; - int from_position = map.at(from).first; - int to_position = map.at(to).first; // Positions of each interval according to granurality map + int from_position = static_cast(from.kind); + int to_position = static_cast(to.kind); // Positions of each interval according to granurality map if (from_position < to_position) { for (int i = from_position - 1; i <= to_position; ++i) - { - // Find the kind that matches this position - for (const auto &entry : map) - { - if (entry.second.first == i) - { - conversion_factor *= entry.second.second; - break; - } - } - } + conversion_factor *= interval_conversions[i]; result_value = arguments[0].column->getInt(0) / conversion_factor; } else { for (int i = from_position - 1; i >= to_position; --i) - { - for (const auto &entry : map) - { - if (entry.second.first == i) - { - conversion_factor *= entry.second.second; - break; - } - } - } + conversion_factor *= interval_conversions[i]; result_value = arguments[0].column->getInt(0) * conversion_factor; } From 384aedccaeece56456ad1e5ea17a8da4f56a69a4 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Sat, 10 Aug 2024 00:09:50 +0200 Subject: [PATCH 048/260] Update getLeastSupertype.h --- src/DataTypes/getLeastSupertype.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/src/DataTypes/getLeastSupertype.h b/src/DataTypes/getLeastSupertype.h index 5ea2b6417b2..8dd1685e6e9 100644 --- a/src/DataTypes/getLeastSupertype.h +++ b/src/DataTypes/getLeastSupertype.h @@ -54,5 +54,3 @@ DataTypePtr tryGetLeastSupertype(const TypeIndexSet & types); static std::vector interval_conversions = {1000, 1000, 1000, 60, 60, 24, 7, 4, 3, 4, 1}; } - -} From ece707c4436ab65fcb142f0eaae72f7eb2c3d8db Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Sat, 10 Aug 2024 19:43:29 +0200 Subject: [PATCH 049/260] Better test for Not-ready Set is passed in system.* tables - system.distribution_queue - system.replication_queue - system.rocksdb - system.databases - system.mutations - test for system.part_moves_between_shards will not be provided since it is a likely deprecated feature and the test requires some code (I've fixed it differently from #66018, but it does not make sense anymore, so I'm submitting only the test) --- ...3223_system_tables_set_not_ready.reference | 5 ++++ .../03223_system_tables_set_not_ready.sql | 30 +++++++++++++++++++ 2 files changed, 35 insertions(+) create mode 100644 tests/queries/0_stateless/03223_system_tables_set_not_ready.reference create mode 100644 tests/queries/0_stateless/03223_system_tables_set_not_ready.sql diff --git a/tests/queries/0_stateless/03223_system_tables_set_not_ready.reference b/tests/queries/0_stateless/03223_system_tables_set_not_ready.reference new file mode 100644 index 00000000000..e39523ed4f5 --- /dev/null +++ b/tests/queries/0_stateless/03223_system_tables_set_not_ready.reference @@ -0,0 +1,5 @@ +system.distribution_queue 1 +system.rocksdb 1 +system.databases 1 +system.mutations 1 +system.replication_queue 1 diff --git a/tests/queries/0_stateless/03223_system_tables_set_not_ready.sql b/tests/queries/0_stateless/03223_system_tables_set_not_ready.sql new file mode 100644 index 00000000000..907fa47143c --- /dev/null +++ b/tests/queries/0_stateless/03223_system_tables_set_not_ready.sql @@ -0,0 +1,30 @@ +-- Tags: no-fasttest +-- Tag no-fasttest -- due to EmbeddedRocksDB + +drop table if exists null; +drop table if exists dist; +create table null as system.one engine=Null; +create table dist as null engine=Distributed(test_cluster_two_shards, currentDatabase(), 'null', rand()); +insert into dist settings prefer_localhost_replica=0 values (1); +select 'system.distribution_queue', count() from system.distribution_queue where exists(select 1) and database = currentDatabase(); + +drop table if exists rocksdb; +create table rocksdb (key Int) engine=EmbeddedRocksDB() primary key key; +insert into rocksdb values (1); +select 'system.rocksdb', count()>0 from system.rocksdb where exists(select 1) and database = currentDatabase(); + +select 'system.databases', count() from system.databases where exists(select 1) and database = currentDatabase(); + +drop table if exists mt; +create table mt (key Int) engine=MergeTree() order by key; +alter table mt delete where 1; +select 'system.mutations', count() from system.mutations where exists(select 1) and database = currentDatabase(); + +drop table if exists rep1; +drop table if exists rep2; +create table rep1 (key Int) engine=ReplicatedMergeTree('/{database}/rep', '{table}') order by key; +create table rep2 (key Int) engine=ReplicatedMergeTree('/{database}/rep', '{table}') order by key; +system stop fetches rep2; +insert into rep1 values (1); +system sync replica rep2 pull; +select 'system.replication_queue', count()>0 from system.replication_queue where exists(select 1) and database = currentDatabase(); From 1cc845726842f388c4524d55b248f210e28d979d Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Mon, 12 Aug 2024 16:57:47 +0200 Subject: [PATCH 050/260] fix reviews, fix crash in fuzzer --- src/DataTypes/getLeastSupertype.cpp | 2 +- src/DataTypes/getLeastSupertype.h | 2 +- src/Functions/FunctionsConversion.cpp | 11 +- .../03223_interval_data_type_comparison.sql | 198 +++++++++--------- 4 files changed, 106 insertions(+), 107 deletions(-) diff --git a/src/DataTypes/getLeastSupertype.cpp b/src/DataTypes/getLeastSupertype.cpp index 8bcec49815f..65df529e78b 100644 --- a/src/DataTypes/getLeastSupertype.cpp +++ b/src/DataTypes/getLeastSupertype.cpp @@ -251,7 +251,7 @@ DataTypePtr findSmallestIntervalSuperType(const DataTypes &types, TypeIndexSet & } if (is_higher_interval && min_interval <= IntervalKind::Kind::Week) - throw Exception(ErrorCodes::NO_COMMON_TYPE, "Cannot compare intervals {} and {} because the amount of days in month is not determined", types[0]->getName(), types[1]->getName()); + throw Exception(ErrorCodes::NO_COMMON_TYPE, "Cannot compare intervals {} and {} because the number of days in a month is not fixed", types[0]->getName(), types[1]->getName()); if (smallest_type) { diff --git a/src/DataTypes/getLeastSupertype.h b/src/DataTypes/getLeastSupertype.h index 8dd1685e6e9..55d8e8fff0d 100644 --- a/src/DataTypes/getLeastSupertype.h +++ b/src/DataTypes/getLeastSupertype.h @@ -51,6 +51,6 @@ DataTypePtr getLeastSupertypeOrString(const TypeIndexSet & types); DataTypePtr tryGetLeastSupertype(const TypeIndexSet & types); /// A vector that shows the conversion rates to the next Interval type starting from NanoSecond -static std::vector interval_conversions = {1000, 1000, 1000, 60, 60, 24, 7, 4, 3, 4, 1}; +static std::vector interval_conversions = {1, 1000, 1000, 1000, 60, 60, 24, 7, 4, 3, 4}; } diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 25c6bbcbfef..b6102cb7ecf 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -1586,17 +1586,17 @@ struct ConvertImpl Int64 result_value; int from_position = static_cast(from.kind); - int to_position = static_cast(to.kind); // Positions of each interval according to granurality map + int to_position = static_cast(to.kind); /// Positions of each interval according to granularity map if (from_position < to_position) { - for (int i = from_position - 1; i <= to_position; ++i) + for (int i = from_position; i < to_position; ++i) conversion_factor *= interval_conversions[i]; result_value = arguments[0].column->getInt(0) / conversion_factor; } else { - for (int i = from_position - 1; i >= to_position; --i) + for (int i = from_position; i > to_position; --i) conversion_factor *= interval_conversions[i]; result_value = arguments[0].column->getInt(0) * conversion_factor; } @@ -2366,9 +2366,8 @@ private: } if constexpr (std::is_same_v) - { - done = callOnIndexAndDataType(from_type->getTypeId(), call, BehaviourOnErrorFromString::ConvertDefaultBehaviorTag); - } + if (WhichDataType(from_type).isInterval()) + done = callOnIndexAndDataType(from_type->getTypeId(), call, BehaviourOnErrorFromString::ConvertDefaultBehaviorTag); } if (!done) diff --git a/tests/queries/0_stateless/03223_interval_data_type_comparison.sql b/tests/queries/0_stateless/03223_interval_data_type_comparison.sql index 5d01addae45..77b6e2fa3dc 100644 --- a/tests/queries/0_stateless/03223_interval_data_type_comparison.sql +++ b/tests/queries/0_stateless/03223_interval_data_type_comparison.sql @@ -1,142 +1,142 @@ SELECT('Comparing nanoseconds'); -SELECT toIntervalNanosecond(500) > toIntervalNanosecond(300); -SELECT toIntervalNanosecond(1000) < toIntervalNanosecond(1500); -SELECT toIntervalNanosecond(2000) = toIntervalNanosecond(2000); -SELECT toIntervalNanosecond(1000) >= toIntervalMicrosecond(1); -SELECT toIntervalNanosecond(1000001) > toIntervalMillisecond(1); -SELECT toIntervalNanosecond(2000000001) > toIntervalSecond(2); -SELECT toIntervalNanosecond(60000000000) = toIntervalMinute(1); -SELECT toIntervalNanosecond(7199999999999) < toIntervalHour(2); -SELECT toIntervalNanosecond(1) < toIntervalDay(2); -SELECT toIntervalNanosecond(5) < toIntervalWeek(1); +SELECT INTERVAL 500 NANOSECOND > INTERVAL 300 NANOSECOND; +SELECT INTERVAL 1000 NANOSECOND < INTERVAL 1500 NANOSECOND; +SELECT INTERVAL 2000 NANOSECOND = INTERVAL 2000 NANOSECOND; +SELECT INTERVAL 1000 NANOSECOND >= INTERVAL 1 MICROSECOND; +SELECT INTERVAL 1000001 NANOSECOND > INTERVAL 1 MILLISECOND; +SELECT INTERVAL 2000000001 NANOSECOND > INTERVAL 2 SECOND; +SELECT INTERVAL 60000000000 NANOSECOND = INTERVAL 1 MINUTE; +SELECT INTERVAL 7199999999999 NANOSECOND < INTERVAL 2 HOUR; +SELECT INTERVAL 1 NANOSECOND < INTERVAL 2 DAY; +SELECT INTERVAL 5 NANOSECOND < INTERVAL 1 WEEK; -SELECT toIntervalNanosecond(500) < toIntervalNanosecond(300); -SELECT toIntervalNanosecond(1000) > toIntervalNanosecond(1500); -SELECT toIntervalNanosecond(2000) != toIntervalNanosecond(2000); -SELECT toIntervalNanosecond(1000) < toIntervalMicrosecond(1); -SELECT toIntervalNanosecond(1000001) < toIntervalMillisecond(1); -SELECT toIntervalNanosecond(2000000001) < toIntervalSecond(2); -SELECT toIntervalNanosecond(60000000000) != toIntervalMinute(1); -SELECT toIntervalNanosecond(7199999999999) > toIntervalHour(2); -SELECT toIntervalNanosecond(1) > toIntervalDay(2); -SELECT toIntervalNanosecond(5) > toIntervalWeek(1); +SELECT INTERVAL 500 NANOSECOND < INTERVAL 300 NANOSECOND; +SELECT INTERVAL 1000 NANOSECOND > INTERVAL 1500 NANOSECOND; +SELECT INTERVAL 2000 NANOSECOND != INTERVAL 2000 NANOSECOND; +SELECT INTERVAL 1000 NANOSECOND < INTERVAL 1 MICROSECOND; +SELECT INTERVAL 1000001 NANOSECOND < INTERVAL 1 MILLISECOND; +SELECT INTERVAL 2000000001 NANOSECOND < INTERVAL 2 SECOND; +SELECT INTERVAL 60000000000 NANOSECOND != INTERVAL 1 MINUTE; +SELECT INTERVAL 7199999999999 NANOSECOND > INTERVAL 2 HOUR; +SELECT INTERVAL 1 NANOSECOND > INTERVAL 2 DAY; +SELECT INTERVAL 5 NANOSECOND > INTERVAL 1 WEEK; -SELECT toIntervalNanosecond(1) < toIntervalMonth(2); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 1 NANOSECOND < INTERVAL 2 MONTH; -- { serverError NO_COMMON_TYPE } SELECT('Comparing microseconds'); -SELECT toIntervalMicrosecond(1) < toIntervalMicrosecond(999); -SELECT toIntervalMicrosecond(1001) > toIntervalMillisecond(1); -SELECT toIntervalMicrosecond(2000000) = toIntervalSecond(2); -SELECT toIntervalMicrosecond(179999999) < toIntervalMinute(3); -SELECT toIntervalMicrosecond(3600000000) = toIntervalHour(1); -SELECT toIntervalMicrosecond(36000000000000) > toIntervalDay(2); -SELECT toIntervalMicrosecond(1209600000000) = toIntervalWeek(2); +SELECT INTERVAL 1 MICROSECOND < INTERVAL 999 MICROSECOND; +SELECT INTERVAL 1001 MICROSECOND > INTERVAL 1 MILLISECOND; +SELECT INTERVAL 2000000 MICROSECOND = INTERVAL 2 SECOND; +SELECT INTERVAL 179999999 MICROSECOND < INTERVAL 3 MINUTE; +SELECT INTERVAL 3600000000 MICROSECOND = INTERVAL 1 HOUR; +SELECT INTERVAL 36000000000000 MICROSECOND > INTERVAL 2 DAY; +SELECT INTERVAL 1209600000000 MICROSECOND = INTERVAL 2 WEEK; -SELECT toIntervalMicrosecond(1) > toIntervalMicrosecond(999); -SELECT toIntervalMicrosecond(1001) < toIntervalMillisecond(1); -SELECT toIntervalMicrosecond(2000000) != toIntervalSecond(2); -SELECT toIntervalMicrosecond(179999999) > toIntervalMinute(3); -SELECT toIntervalMicrosecond(3600000000) != toIntervalHour(1); -SELECT toIntervalMicrosecond(36000000000000) < toIntervalDay(2); -SELECT toIntervalMicrosecond(1209600000000) != toIntervalWeek(2); +SELECT INTERVAL 1 MICROSECOND > INTERVAL 999 MICROSECOND; +SELECT INTERVAL 1001 MICROSECOND < INTERVAL 1 MILLISECOND; +SELECT INTERVAL 2000000 MICROSECOND != INTERVAL 2 SECOND; +SELECT INTERVAL 179999999 MICROSECOND > INTERVAL 3 MINUTE; +SELECT INTERVAL 3600000000 MICROSECOND != INTERVAL 1 HOUR; +SELECT INTERVAL 36000000000000 MICROSECOND < INTERVAL 2 DAY; +SELECT INTERVAL 1209600000000 MICROSECOND != INTERVAL 2 WEEK; -SELECT toIntervalMicrosecond(36000000000000) < toIntervalQuarter(1); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 36000000000000 MICROSECOND < INTERVAL 1 QUARTER; -- { serverError NO_COMMON_TYPE } SELECT('Comparing milliseconds'); -SELECT toIntervalMillisecond(2000) > toIntervalMillisecond(2); -SELECT toIntervalMillisecond(2000) = toIntervalSecond(2); -SELECT toIntervalMillisecond(170000) < toIntervalMinute(3); -SELECT toIntervalMillisecond(144000001) > toIntervalHour(40); -SELECT toIntervalMillisecond(1728000000) = toIntervalDay(20); -SELECT toIntervalMillisecond(1198599999) < toIntervalWeek(2); +SELECT INTERVAL 2000 MILLISECOND > INTERVAL 2 MILLISECOND; +SELECT INTERVAL 2000 MILLISECOND = INTERVAL 2 SECOND; +SELECT INTERVAL 170000 MILLISECOND < INTERVAL 3 MINUTE; +SELECT INTERVAL 144000001 MILLISECOND > INTERVAL 40 HOUR; +SELECT INTERVAL 1728000000 MILLISECOND = INTERVAL 20 DAY; +SELECT INTERVAL 1198599999 MILLISECOND < INTERVAL 2 WEEK; -SELECT toIntervalMillisecond(2000) < toIntervalMillisecond(2); -SELECT toIntervalMillisecond(2000) != toIntervalSecond(2); -SELECT toIntervalMillisecond(170000) > toIntervalMinute(3); -SELECT toIntervalMillisecond(144000001) < toIntervalHour(40); -SELECT toIntervalMillisecond(1728000000) != toIntervalDay(20); -SELECT toIntervalMillisecond(1198599999) > toIntervalWeek(2); +SELECT INTERVAL 2000 MILLISECOND < INTERVAL 2 MILLISECOND; +SELECT INTERVAL 2000 MILLISECOND != INTERVAL 2 SECOND; +SELECT INTERVAL 170000 MILLISECOND > INTERVAL 3 MINUTE; +SELECT INTERVAL 144000001 MILLISECOND < INTERVAL 40 HOUR; +SELECT INTERVAL 1728000000 MILLISECOND != INTERVAL 20 DAY; +SELECT INTERVAL 1198599999 MILLISECOND > INTERVAL 2 WEEK; -SELECT toIntervalMillisecond(36000000000000) < toIntervalYear(1); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 36000000000000 MILLISECOND < INTERVAL 1 YEAR; -- { serverError NO_COMMON_TYPE } SELECT('Comparing seconds'); -SELECT toIntervalSecond(120) > toIntervalSecond(2); -SELECT toIntervalSecond(120) = toIntervalMinute(2); -SELECT toIntervalSecond(1) < toIntervalHour(2); -SELECT toIntervalSecond(86401) >= toIntervalDay(1); -SELECT toIntervalSecond(1209600) = toIntervalWeek(2); +SELECT INTERVAL 120 SECOND > INTERVAL 2 SECOND; +SELECT INTERVAL 120 SECOND = INTERVAL 2 MINUTE; +SELECT INTERVAL 1 SECOND < INTERVAL 2 HOUR; +SELECT INTERVAL 86401 SECOND >= INTERVAL 1 DAY; +SELECT INTERVAL 1209600 SECOND = INTERVAL 2 WEEK; -SELECT toIntervalSecond(120) < toIntervalSecond(2); -SELECT toIntervalSecond(120) != toIntervalMinute(2); -SELECT toIntervalSecond(1) > toIntervalHour(2); -SELECT toIntervalSecond(86401) < toIntervalDay(1); -SELECT toIntervalSecond(1209600) != toIntervalWeek(2); +SELECT INTERVAL 120 SECOND < INTERVAL 2 SECOND; +SELECT INTERVAL 120 SECOND != INTERVAL 2 MINUTE; +SELECT INTERVAL 1 SECOND > INTERVAL 2 HOUR; +SELECT INTERVAL 86401 SECOND < INTERVAL 1 DAY; +SELECT INTERVAL 1209600 SECOND != INTERVAL 2 WEEK; -SELECT toIntervalSecond(36000000000000) < toIntervalMonth(1); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 36000000000000 SECOND < INTERVAL 1 MONTH; -- { serverError NO_COMMON_TYPE } SELECT('Comparing minutes'); -SELECT toIntervalMinute(1) < toIntervalMinute(59); -SELECT toIntervalMinute(1) < toIntervalHour(59); -SELECT toIntervalMinute(1440) = toIntervalDay(1); -SELECT toIntervalMinute(30241) > toIntervalWeek(3); +SELECT INTERVAL 1 MINUTE < INTERVAL 59 MINUTE; +SELECT INTERVAL 1 MINUTE < INTERVAL 59 HOUR; +SELECT INTERVAL 1440 MINUTE = INTERVAL 1 DAY; +SELECT INTERVAL 30241 MINUTE > INTERVAL 3 WEEK; -SELECT toIntervalMinute(1) > toIntervalMinute(59); -SELECT toIntervalMinute(1) > toIntervalHour(59); -SELECT toIntervalMinute(1440) != toIntervalDay(1); -SELECT toIntervalMinute(30241) < toIntervalWeek(3); +SELECT INTERVAL 1 MINUTE > INTERVAL 59 MINUTE; +SELECT INTERVAL 1 MINUTE > INTERVAL 59 HOUR; +SELECT INTERVAL 1440 MINUTE != INTERVAL 1 DAY; +SELECT INTERVAL 30241 MINUTE < INTERVAL 3 WEEK; -SELECT toIntervalMinute(2) = toIntervalQuarter(120); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 2 MINUTE = INTERVAL 120 QUARTER; -- { serverError NO_COMMON_TYPE } SELECT('Comparing hours'); -SELECT toIntervalHour(48) > toIntervalHour(2); -SELECT toIntervalHour(48) >= toIntervalDay(2); -SELECT toIntervalHour(672) = toIntervalWeek(4); +SELECT INTERVAL 48 HOUR > INTERVAL 2 HOUR; +SELECT INTERVAL 48 HOUR >= INTERVAL 2 DAY; +SELECT INTERVAL 672 HOUR = INTERVAL 4 WEEK; -SELECT toIntervalHour(48) < toIntervalHour(2); -SELECT toIntervalHour(48) < toIntervalDay(2); -SELECT toIntervalHour(672) != toIntervalWeek(4); +SELECT INTERVAL 48 HOUR < INTERVAL 2 HOUR; +SELECT INTERVAL 48 HOUR < INTERVAL 2 DAY; +SELECT INTERVAL 672 HOUR != INTERVAL 4 WEEK; -SELECT toIntervalHour(2) < toIntervalYear(1); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 2 HOUR < INTERVAL 1 YEAR; -- { serverError NO_COMMON_TYPE } SELECT('Comparing days'); -SELECT toIntervalDay(1) < toIntervalDay(23); -SELECT toIntervalDay(25) > toIntervalWeek(3); +SELECT INTERVAL 1 DAY < INTERVAL 23 DAY; +SELECT INTERVAL 25 DAY > INTERVAL 3 WEEK; -SELECT toIntervalDay(1) > toIntervalDay(23); -SELECT toIntervalDay(25) < toIntervalWeek(3); +SELECT INTERVAL 1 DAY > INTERVAL 23 DAY; +SELECT INTERVAL 25 DAY < INTERVAL 3 WEEK; -SELECT toIntervalDay(2) = toIntervalMonth(48); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 2 DAY = INTERVAL 48 MONTH; -- { serverError NO_COMMON_TYPE } SELECT('Comparing weeks'); -SELECT toIntervalWeek(1) < toIntervalWeek(6); +SELECT INTERVAL 1 WEEK < INTERVAL 6 WEEK; -SELECT toIntervalWeek(1) > toIntervalWeek(6); +SELECT INTERVAL 1 WEEK > INTERVAL 6 WEEK; -SELECT toIntervalWeek(124) > toIntervalQuarter(8); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 124 WEEK > INTERVAL 8 QUARTER; -- { serverError NO_COMMON_TYPE } SELECT('Comparing months'); -SELECT toIntervalMonth(1) < toIntervalMonth(3); -SELECT toIntervalMonth(124) > toIntervalQuarter(5); -SELECT toIntervalMonth(36) = toIntervalYear(3); +SELECT INTERVAL 1 MONTH < INTERVAL 3 MONTH; +SELECT INTERVAL 124 MONTH > INTERVAL 5 QUARTER; +SELECT INTERVAL 36 MONTH = INTERVAL 3 YEAR; -SELECT toIntervalMonth(1) > toIntervalMonth(3); -SELECT toIntervalMonth(124) < toIntervalQuarter(5); -SELECT toIntervalMonth(36) != toIntervalYear(3); +SELECT INTERVAL 1 MONTH > INTERVAL 3 MONTH; +SELECT INTERVAL 124 MONTH < INTERVAL 5 QUARTER; +SELECT INTERVAL 36 MONTH != INTERVAL 3 YEAR; -SELECT toIntervalMonth(6) = toIntervalMicrosecond(26); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 6 MONTH = INTERVAL 26 MICROSECOND; -- { serverError NO_COMMON_TYPE } SELECT('Comparing quarters'); -SELECT toIntervalQuarter(5) > toIntervalQuarter(4); -SELECT toIntervalQuarter(20) = toIntervalYear(5); +SELECT INTERVAL 5 QUARTER > INTERVAL 4 QUARTER; +SELECT INTERVAL 20 QUARTER = INTERVAL 5 YEAR; -SELECT toIntervalQuarter(5) < toIntervalQuarter(4); -SELECT toIntervalQuarter(20) != toIntervalYear(5); +SELECT INTERVAL 5 QUARTER < INTERVAL 4 QUARTER; +SELECT INTERVAL 20 QUARTER != INTERVAL 5 YEAR; -SELECT toIntervalQuarter(2) = toIntervalNanosecond(6); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 2 QUARTER = INTERVAL 6 NANOSECOND; -- { serverError NO_COMMON_TYPE } SELECT('Comparing years'); -SELECT toIntervalYear(1) < toIntervalYear(3); +SELECT INTERVAL 1 YEAR < INTERVAL 3 YEAR; -SELECT toIntervalYear(1) > toIntervalYear(3); +SELECT INTERVAL 1 YEAR > INTERVAL 3 YEAR; -SELECT toIntervalYear(2) = toIntervalSecond(8); -- { serverError NO_COMMON_TYPE } +SELECT INTERVAL 2 YEAR = INTERVAL 8 SECOND; -- { serverError NO_COMMON_TYPE } \ No newline at end of file From 4c043301e6dde6b0c83394d6721e112c9c7bf4ce Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Tue, 13 Aug 2024 10:30:31 +0200 Subject: [PATCH 051/260] Avoid ignoring errors of execute_process() (set COMMAND_ERROR_IS_FATAL=ANY) This will fix with issues like this [1]: Aug 12 09:58:44 '/usr/bin/cmake' '--build' '/build/build_docker/native' '--target' 'pre_compressor' Aug 12 09:58:44 sccache: error: Server startup failed: cache storage failed to read: Unexpected (temporary) at stat Aug 12 09:58:45 ninja: build stopped: subcommand failed. Aug 12 09:58:45 -- Configuring done (77.7s) Aug 12 09:58:47 -- Generating done (1.8s) Aug 12 09:58:47 -- Build files have been written to: /build/build_docker So as you can see even if ninja fails it still wrote build files, while it should fail. [1]: https://s3.amazonaws.com/clickhouse-test-reports/64955/0af41e32a5822d25ac3760f1ebb2313557474701/builds/report.html [2]: https://s3.amazonaws.com/clickhouse-builds/PRs/64955/0af41e32a5822d25ac3760f1ebb2313557474701/binary_darwin_aarch64/build_log.log Note, COMMAND_ERROR_IS_FATAL is 3.19+, and the requirement for now is 3.20 Signed-off-by: Azat Khuzhin --- CMakeLists.txt | 12 ++++++-- PreLoad.cmake | 10 +++++-- cmake/freebsd/default_libs.cmake | 12 ++++++-- cmake/linux/default_libs.cmake | 6 +++- cmake/tools.cmake | 6 +++- cmake/utils.cmake | 5 +++- contrib/cctz-cmake/CMakeLists.txt | 4 ++- contrib/google-protobuf-cmake/CMakeLists.txt | 12 ++++++-- contrib/grpc-cmake/CMakeLists.txt | 30 +++++++++++++------ .../completions/CMakeLists.txt | 1 + 10 files changed, 75 insertions(+), 23 deletions(-) diff --git a/CMakeLists.txt b/CMakeLists.txt index 7b4e0484ab1..8e2302e6c52 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -609,7 +609,9 @@ if (NATIVE_BUILD_TARGETS execute_process( COMMAND ${CMAKE_COMMAND} -E make_directory "${NATIVE_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) execute_process( COMMAND ${CMAKE_COMMAND} @@ -621,9 +623,13 @@ if (NATIVE_BUILD_TARGETS "-DENABLE_CLICKHOUSE_SELF_EXTRACTING=${ENABLE_CLICKHOUSE_SELF_EXTRACTING}" ${PROJECT_SOURCE_DIR} WORKING_DIRECTORY "${NATIVE_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) execute_process( COMMAND ${CMAKE_COMMAND} --build "${NATIVE_BUILD_DIR}" --target ${NATIVE_BUILD_TARGETS} - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) endif () diff --git a/PreLoad.cmake b/PreLoad.cmake index e0fd37b2fd6..92b221c9f63 100644 --- a/PreLoad.cmake +++ b/PreLoad.cmake @@ -51,8 +51,14 @@ if (NOT "$ENV{CFLAGS}" STREQUAL "" endif() # Default toolchain - this is needed to avoid dependency on OS files. -execute_process(COMMAND uname -s OUTPUT_VARIABLE OS) -execute_process(COMMAND uname -m OUTPUT_VARIABLE ARCH) +execute_process(COMMAND uname -s + OUTPUT_VARIABLE OS + COMMAND_ERROR_IS_FATAL ANY +) +execute_process(COMMAND uname -m + OUTPUT_VARIABLE ARCH + COMMAND_ERROR_IS_FATAL ANY +) # By default, prefer clang on Linux # But note, that you still may change the compiler with -DCMAKE_C_COMPILER/-DCMAKE_CXX_COMPILER. diff --git a/cmake/freebsd/default_libs.cmake b/cmake/freebsd/default_libs.cmake index 6bde75f8c9a..3f5b3829877 100644 --- a/cmake/freebsd/default_libs.cmake +++ b/cmake/freebsd/default_libs.cmake @@ -9,10 +9,18 @@ endif () file(GLOB bprefix "/usr/local/llvm${COMPILER_VERSION_MAJOR}/lib/clang/${COMPILER_VERSION_MAJOR}/lib/${system_processor}-portbld-freebsd*/") message(STATUS "-Bprefix: ${bprefix}") -execute_process(COMMAND ${CMAKE_CXX_COMPILER} -Bprefix=${bprefix} --print-file-name=libclang_rt.builtins-${system_processor}.a OUTPUT_VARIABLE BUILTINS_LIBRARY OUTPUT_STRIP_TRAILING_WHITESPACE) +execute_process(COMMAND + ${CMAKE_CXX_COMPILER} -Bprefix=${bprefix} --print-file-name=libclang_rt.builtins-${system_processor}.a + OUTPUT_VARIABLE BUILTINS_LIBRARY + COMMAND_ERROR_IS_FATAL ANY + OUTPUT_STRIP_TRAILING_WHITESPACE) # --print-file-name simply prints what you passed in case of nothing was resolved, so let's try one other possible option if (BUILTINS_LIBRARY STREQUAL "libclang_rt.builtins-${system_processor}.a") - execute_process(COMMAND ${CMAKE_CXX_COMPILER} -Bprefix=${bprefix} --print-file-name=libclang_rt.builtins.a OUTPUT_VARIABLE BUILTINS_LIBRARY OUTPUT_STRIP_TRAILING_WHITESPACE) + execute_process(COMMAND + ${CMAKE_CXX_COMPILER} -Bprefix=${bprefix} --print-file-name=libclang_rt.builtins.a + OUTPUT_VARIABLE BUILTINS_LIBRARY + COMMAND_ERROR_IS_FATAL ANY + OUTPUT_STRIP_TRAILING_WHITESPACE) endif() if (BUILTINS_LIBRARY STREQUAL "libclang_rt.builtins.a") message(FATAL_ERROR "libclang_rt.builtins had not been found") diff --git a/cmake/linux/default_libs.cmake b/cmake/linux/default_libs.cmake index 4a06243243e..51620bc9f33 100644 --- a/cmake/linux/default_libs.cmake +++ b/cmake/linux/default_libs.cmake @@ -5,7 +5,11 @@ set (DEFAULT_LIBS "-nodefaultlibs") # We need builtins from Clang's RT even without libcxx - for ubsan+int128. # See https://bugs.llvm.org/show_bug.cgi?id=16404 -execute_process (COMMAND ${CMAKE_CXX_COMPILER} --target=${CMAKE_CXX_COMPILER_TARGET} --print-libgcc-file-name --rtlib=compiler-rt OUTPUT_VARIABLE BUILTINS_LIBRARY OUTPUT_STRIP_TRAILING_WHITESPACE) +execute_process (COMMAND + ${CMAKE_CXX_COMPILER} --target=${CMAKE_CXX_COMPILER_TARGET} --print-libgcc-file-name --rtlib=compiler-rt + OUTPUT_VARIABLE BUILTINS_LIBRARY + COMMAND_ERROR_IS_FATAL ANY + OUTPUT_STRIP_TRAILING_WHITESPACE) # Apparently, in clang-19, the UBSan support library for C++ was moved out into ubsan_standalone_cxx.a, so we have to include both. if (SANITIZE STREQUAL undefined) diff --git a/cmake/tools.cmake b/cmake/tools.cmake index 7aa5d4c51ce..5c7da54b779 100644 --- a/cmake/tools.cmake +++ b/cmake/tools.cmake @@ -5,7 +5,11 @@ if (NOT CMAKE_CXX_COMPILER_ID MATCHES "Clang") endif () # Print details to output -execute_process(COMMAND ${CMAKE_CXX_COMPILER} --version OUTPUT_VARIABLE COMPILER_SELF_IDENTIFICATION OUTPUT_STRIP_TRAILING_WHITESPACE) +execute_process(COMMAND ${CMAKE_CXX_COMPILER} --version + OUTPUT_VARIABLE COMPILER_SELF_IDENTIFICATION + COMMAND_ERROR_IS_FATAL ANY + OUTPUT_STRIP_TRAILING_WHITESPACE +) message (STATUS "Using compiler:\n${COMPILER_SELF_IDENTIFICATION}") # Require minimum compiler versions diff --git a/cmake/utils.cmake b/cmake/utils.cmake index a318408098a..a99d8e050a8 100644 --- a/cmake/utils.cmake +++ b/cmake/utils.cmake @@ -90,7 +90,10 @@ endfunction() # Function get_cmake_properties returns list of all propreties that cmake supports function(get_cmake_properties outvar) - execute_process(COMMAND cmake --help-property-list OUTPUT_VARIABLE cmake_properties) + execute_process(COMMAND cmake --help-property-list + OUTPUT_VARIABLE cmake_properties + COMMAND_ERROR_IS_FATAL ANY + ) # Convert command output into a CMake list string(REGEX REPLACE ";" "\\\\;" cmake_properties "${cmake_properties}") string(REGEX REPLACE "\n" ";" cmake_properties "${cmake_properties}") diff --git a/contrib/cctz-cmake/CMakeLists.txt b/contrib/cctz-cmake/CMakeLists.txt index 7161f743de1..fadf948b053 100644 --- a/contrib/cctz-cmake/CMakeLists.txt +++ b/contrib/cctz-cmake/CMakeLists.txt @@ -37,7 +37,9 @@ message(STATUS "Packaging with tzdata version: ${TZDATA_VERSION}") execute_process(COMMAND bash -c "cd ${TZDIR} && find * -type f -and ! -name '*.tab' -and ! -name 'localtime' | LC_ALL=C sort | paste -sd ';' -" OUTPUT_STRIP_TRAILING_WHITESPACE - OUTPUT_VARIABLE TIMEZONES) + OUTPUT_VARIABLE TIMEZONES + COMMAND_ERROR_IS_FATAL ANY +) file(APPEND ${TIMEZONES_FILE} "// autogenerated by ClickHouse/contrib/cctz-cmake/CMakeLists.txt\n") file(APPEND ${TIMEZONES_FILE} "#include \n") diff --git a/contrib/google-protobuf-cmake/CMakeLists.txt b/contrib/google-protobuf-cmake/CMakeLists.txt index e44f737cfc3..f1a744f851f 100644 --- a/contrib/google-protobuf-cmake/CMakeLists.txt +++ b/contrib/google-protobuf-cmake/CMakeLists.txt @@ -359,7 +359,9 @@ else () execute_process( COMMAND mkdir -p ${PROTOC_BUILD_DIR} - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) execute_process( COMMAND ${CMAKE_COMMAND} @@ -375,11 +377,15 @@ else () "-DABSL_ENABLE_INSTALL=0" "${protobuf_source_dir}" WORKING_DIRECTORY "${PROTOC_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) execute_process( COMMAND ${CMAKE_COMMAND} --build "${PROTOC_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) endif () add_executable(protoc IMPORTED GLOBAL) diff --git a/contrib/grpc-cmake/CMakeLists.txt b/contrib/grpc-cmake/CMakeLists.txt index 1c0bf41ff78..975774d1990 100644 --- a/contrib/grpc-cmake/CMakeLists.txt +++ b/contrib/grpc-cmake/CMakeLists.txt @@ -51,8 +51,9 @@ if (NOT CMAKE_HOST_SYSTEM_NAME STREQUAL CMAKE_SYSTEM_NAME set(OPENSSL_SOURCE_DIR "${ClickHouse_SOURCE_DIR}/contrib/openssl-cmake") execute_process( - COMMAND mkdir -p ${OPENSSL_BUILD_DIR} - COMMAND_ECHO STDOUT + COMMAND mkdir -p ${OPENSSL_BUILD_DIR} + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY ) if (CMAKE_HOST_SYSTEM_PROCESSOR MATCHES "amd64|x86_64") @@ -89,15 +90,21 @@ if (NOT CMAKE_HOST_SYSTEM_NAME STREQUAL CMAKE_SYSTEM_NAME "-DClickHouse_SOURCE_DIR=${ClickHouse_SOURCE_DIR}" "${OPENSSL_SOURCE_DIR}" WORKING_DIRECTORY "${OPENSSL_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) execute_process( COMMAND ${CMAKE_COMMAND} --build "${OPENSSL_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) execute_process( COMMAND ${CMAKE_COMMAND} --install "${OPENSSL_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) # It's not important on which file we depend, we just want to specify right order add_library(openssl_for_grpc STATIC IMPORTED GLOBAL) @@ -108,8 +115,9 @@ if (NOT CMAKE_HOST_SYSTEM_NAME STREQUAL CMAKE_SYSTEM_NAME set (GRPC_CPP_PLUGIN_BUILD_DIR "${_gRPC_BINARY_DIR}/build") execute_process( - COMMAND mkdir -p ${GRPC_CPP_PLUGIN_BUILD_DIR} - COMMAND_ECHO STDOUT + COMMAND mkdir -p ${GRPC_CPP_PLUGIN_BUILD_DIR} + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY ) set(abseil_source_dir "${ClickHouse_SOURCE_DIR}/contrib/abseil-cpp") @@ -140,11 +148,15 @@ if (NOT CMAKE_HOST_SYSTEM_NAME STREQUAL CMAKE_SYSTEM_NAME "-DgRPC_SSL_PROVIDER=package" "${_gRPC_SOURCE_DIR}" WORKING_DIRECTORY "${GRPC_CPP_PLUGIN_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) execute_process( COMMAND ${CMAKE_COMMAND} --build "${GRPC_CPP_PLUGIN_BUILD_DIR}" - COMMAND_ECHO STDOUT) + COMMAND_ECHO STDOUT + COMMAND_ERROR_IS_FATAL ANY + ) add_executable(grpc_cpp_plugin IMPORTED GLOBAL) set_target_properties (grpc_cpp_plugin PROPERTIES IMPORTED_LOCATION "${GRPC_CPP_PLUGIN_BUILD_DIR}/grpc_cpp_plugin") diff --git a/programs/bash-completion/completions/CMakeLists.txt b/programs/bash-completion/completions/CMakeLists.txt index d364e07ef6e..2e911e81981 100644 --- a/programs/bash-completion/completions/CMakeLists.txt +++ b/programs/bash-completion/completions/CMakeLists.txt @@ -6,6 +6,7 @@ macro(configure_bash_completion) COMMAND ${PKG_CONFIG_BIN} --variable=completionsdir bash-completion OUTPUT_VARIABLE ${out} OUTPUT_STRIP_TRAILING_WHITESPACE + COMMAND_ERROR_IS_FATAL ANY ) endif() string(REPLACE /usr "${CMAKE_INSTALL_PREFIX}" out "${out}") From 7f005a6ca48d4f193470d3a71bc1d97ff55f4a2f Mon Sep 17 00:00:00 2001 From: shiyer7474 Date: Tue, 13 Aug 2024 08:38:30 +0000 Subject: [PATCH 052/260] Fix small value DateTime64 constant folding in nested query --- src/Analyzer/ConstantNode.cpp | 10 ++++- ...222_datetime64_small_value_const.reference | 18 +++++++++ .../03222_datetime64_small_value_const.sql | 39 +++++++++++++++++++ 3 files changed, 65 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03222_datetime64_small_value_const.reference create mode 100644 tests/queries/0_stateless/03222_datetime64_small_value_const.sql diff --git a/src/Analyzer/ConstantNode.cpp b/src/Analyzer/ConstantNode.cpp index c65090f5b55..3d0f448da4b 100644 --- a/src/Analyzer/ConstantNode.cpp +++ b/src/Analyzer/ConstantNode.cpp @@ -177,9 +177,15 @@ ASTPtr ConstantNode::toASTImpl(const ConvertToASTOptions & options) const * It could also lead to ambiguous parsing because we don't know if the string literal represents a date or a Decimal64 literal. * For this reason, we use a string literal representing a date instead of a Decimal64 literal. */ - if (WhichDataType(constant_value_type->getTypeId()).isDateTime64()) + if ((WhichDataType(constant_value_type->getTypeId()).isDateTime64()) || + (WhichDataType(constant_value_type->getTypeId()).isNullable() && WhichDataType((typeid_cast(constant_value_type.get()))->getNestedType()->getTypeId()).isDateTime64())) { - const auto * date_time_type = typeid_cast(constant_value_type.get()); + const DataTypeDateTime64 * date_time_type = nullptr; + if (WhichDataType(constant_value_type->getTypeId()).isNullable()) + date_time_type = typeid_cast((typeid_cast(constant_value_type.get()))->getNestedType().get()); + else + date_time_type = typeid_cast(constant_value_type.get()); + DecimalField decimal_value; if (constant_value_literal.tryGet>(decimal_value)) { diff --git a/tests/queries/0_stateless/03222_datetime64_small_value_const.reference b/tests/queries/0_stateless/03222_datetime64_small_value_const.reference new file mode 100644 index 00000000000..ae36c08acc5 --- /dev/null +++ b/tests/queries/0_stateless/03222_datetime64_small_value_const.reference @@ -0,0 +1,18 @@ +0 1970-01-01 00:00:00.000 +0 1970-01-01 00:00:05.000 +0 1970-01-01 00:45:25.456789 +0 1970-01-01 00:53:25.456789123 +0 \N +1 1970-01-01 00:00:00.000 +5 1970-01-01 00:00:00.000 +2 1970-01-01 00:00:02.456 +3 1970-01-01 00:00:04.811 +4 1970-01-01 00:10:05.000 +4 1970-01-01 00:10:05.000 +1 1970-01-01 00:00:00.000 +2 1970-01-01 00:00:02.456 +3 1970-01-01 00:00:04.811 +5 1970-01-01 00:00:00.000 +0 +0 +5 diff --git a/tests/queries/0_stateless/03222_datetime64_small_value_const.sql b/tests/queries/0_stateless/03222_datetime64_small_value_const.sql new file mode 100644 index 00000000000..6999ba9662a --- /dev/null +++ b/tests/queries/0_stateless/03222_datetime64_small_value_const.sql @@ -0,0 +1,39 @@ +-- Tags: shard + +select *, (select toDateTime64(0, 3)) from remote('127.0.0.1', system.one) settings prefer_localhost_replica=0; +select *, (select toDateTime64(5, 3)) from remote('127.0.0.1', system.one) settings prefer_localhost_replica=0; +select *, (select toDateTime64('1970-01-01 00:45:25.456789', 6)) from remote('127.0.0.1', system.one) settings prefer_localhost_replica=0; +select *, (select toDateTime64('1970-01-01 00:53:25.456789123', 9)) from remote('127.0.0.1', system.one) settings prefer_localhost_replica=0; +select *, (select toDateTime64(null,3)) from remote('127.0.0.1', system.one) settings prefer_localhost_replica=0; + +create database if not exists shard_0; +create database if not exists shard_1; + +drop table if exists shard_0.dt64_03222; +drop table if exists shard_1.dt64_03222; +drop table if exists distr_03222_dt64; + +create table shard_0.dt64_03222(id UInt64, dt DateTime64(3)) engine = MergeTree order by id; +create table shard_1.dt64_03222(id UInt64, dt DateTime64(3)) engine = MergeTree order by id; +create table distr_03222_dt64 (id UInt64, dt DateTime64(3)) engine = Distributed(test_cluster_two_shards_different_databases, '', dt64_03222); + +insert into shard_0.dt64_03222 values(1, toDateTime64('1970-01-01 00:00:00.000',3)) +insert into shard_0.dt64_03222 values(2, toDateTime64('1970-01-01 00:00:02.456',3)); +insert into shard_1.dt64_03222 values(3, toDateTime64('1970-01-01 00:00:04.811',3)); +insert into shard_1.dt64_03222 values(4, toDateTime64('1970-01-01 00:10:05',3)); +insert into shard_1.dt64_03222 values(5, toDateTime64(0,3)); + +--Output : 1,5 2,3,4 4 1,2,3,5 0 0 5 +select id, dt from distr_03222_dt64 where dt = (select toDateTime64(0,3)) order by id; +select id, dt from distr_03222_dt64 where dt > (select toDateTime64(0,3)) order by id; +select id, dt from distr_03222_dt64 where dt > (select toDateTime64('1970-01-01 00:10:00.000',3)) order by id; +select id, dt from distr_03222_dt64 where dt < (select toDateTime64(5,3)) order by id; + +select count(*) from distr_03222_dt64 where dt > (select toDateTime64('2024-07-20 00:00:00',3)); +select count(*) from distr_03222_dt64 where dt > (select now()); +select count(*) from distr_03222_dt64 where dt < (select toDateTime64('2004-07-20 00:00:00',3)); + + +drop table if exists shard_0.dt64_03222; +drop table if exists shard_1.dt64_03222; +drop table if exists distr_03222_dt64; From e46c5a75ef6b5488834add56bea4cab327515bfb Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Tue, 13 Aug 2024 17:59:05 +0800 Subject: [PATCH 053/260] fix building issue --- src/Functions/{FunctionOverlay.cpp => overlay.cpp} | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) rename src/Functions/{FunctionOverlay.cpp => overlay.cpp} (99%) diff --git a/src/Functions/FunctionOverlay.cpp b/src/Functions/overlay.cpp similarity index 99% rename from src/Functions/FunctionOverlay.cpp rename to src/Functions/overlay.cpp index 61d2df88ab1..094da27a71d 100644 --- a/src/Functions/FunctionOverlay.cpp +++ b/src/Functions/overlay.cpp @@ -728,8 +728,8 @@ private: REGISTER_FUNCTION(Overlay) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); - factory.registerFunction>({}, FunctionFactory::CaseSensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); + factory.registerFunction>({}, FunctionFactory::Case::Sensitive); } } From 0414cdbbbf32efe10a92c9dd93ba47743ceeb848 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Tue, 13 Aug 2024 15:58:49 +0200 Subject: [PATCH 054/260] Fix unpack error --- tests/clickhouse-test | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 5946e561949..5bde4686d3a 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1995,7 +1995,7 @@ class TestSuite: tag_line = find_tag_line(file) next_line = file.readline() except UnicodeDecodeError: - return [] + return [], {} try: if filepath.endswith(".sql"): for line in file: From 6dfed409f460311f133e30e70f839f9865d71861 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 13 Aug 2024 16:09:45 +0000 Subject: [PATCH 055/260] Fix seraching for query params --- tests/clickhouse-test | 35 ++++++++++++++++++++++------------- 1 file changed, 22 insertions(+), 13 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 5bde4686d3a..515b519af3e 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1978,13 +1978,20 @@ class TestSuite: def is_shebang(line: str) -> bool: return line.startswith("#!") - def find_tag_line(file): - line = file.readline() - while line != "": - line = line.strip() - if line and not is_shebang(line): + def find_tag_line(lines, comment_sign): + for line in lines: + if line.startswith(comment_sign) and line[ + len(comment_sign) : + ].lstrip().startswith("Tags:"): + return line + return "" + + def find_random_settings_limits_line(lines, comment_sign): + for line in lines: + if line.startswith(comment_sign) and line[ + len(comment_sign) : + ].lstrip().startswith("Random settings limits:"): return line - line = file.readline() return "" def load_tags_and_random_settings_limits_from_file(filepath): @@ -1992,13 +1999,16 @@ class TestSuite: need_query_params = False with open(filepath, "r", encoding="utf-8") as file: try: - tag_line = find_tag_line(file) - next_line = file.readline() + lines = file.readlines() + tag_line = find_tag_line(lines, comment_sign) + random_settings_limits_line = find_random_settings_limits_line( + lines, comment_sign + ) except UnicodeDecodeError: return [], {} try: if filepath.endswith(".sql"): - for line in file: + for line in lines: if "{CLICKHOUSE_DATABASE" in line: need_query_params = True except UnicodeDecodeError: @@ -2006,7 +2016,6 @@ class TestSuite: parsed_tags = parse_tags_from_line(tag_line, comment_sign) if need_query_params: parsed_tags.add("need-query-parameters") - random_settings_limits_line = next_line if parsed_tags else tag_line random_settings_limits = parse_random_settings_limits_from_line( random_settings_limits_line, comment_sign ) @@ -2068,9 +2077,9 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = ( - all_tags_and_random_settings_limits[1] - ) + self.all_random_settings_limits: Dict[ + str, Dict[str, (int, int)] + ] = all_tags_and_random_settings_limits[1] self.sequential_tests = [] self.parallel_tests = [] for test_name in self.all_tests: From 3cfb921befa895e445e8d7b98e639015e1e41aa0 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 13 Aug 2024 18:41:53 +0000 Subject: [PATCH 056/260] Fix using schema_inference_make_columns_nullable=0 --- src/Core/Settings.h | 2 +- src/Formats/FormatFactory.cpp | 2 +- src/Formats/FormatSettings.h | 2 +- src/Formats/SchemaInferenceUtils.cpp | 4 +++ src/Processors/Formats/ISchemaReader.cpp | 2 +- .../Formats/Impl/ArrowBlockInputFormat.cpp | 7 +++-- .../Formats/Impl/ArrowColumnToCHColumn.cpp | 9 ++++--- .../Formats/Impl/ArrowColumnToCHColumn.h | 3 ++- .../Impl/NativeORCBlockInputFormat.cpp | 2 +- .../Formats/Impl/ORCBlockInputFormat.cpp | 7 +++-- .../Formats/Impl/ParquetBlockInputFormat.cpp | 7 +++-- .../03036_parquet_arrow_nullable.reference | 26 +++++++++++++++++++ .../03036_parquet_arrow_nullable.sh | 7 +++++ 13 files changed, 65 insertions(+), 15 deletions(-) diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 0808e8eb49f..bc9c6daab1b 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -1116,7 +1116,7 @@ class IColumn; M(String, column_names_for_schema_inference, "", "The list of column names to use in schema inference for formats without column names. The format: 'column1,column2,column3,...'", 0) \ M(String, schema_inference_hints, "", "The list of column names and types to use in schema inference for formats without column names. The format: 'column_name1 column_type1, column_name2 column_type2, ...'", 0) \ M(SchemaInferenceMode, schema_inference_mode, "default", "Mode of schema inference. 'default' - assume that all files have the same schema and schema can be inferred from any file, 'union' - files can have different schemas and the resulting schema should be the a union of schemas of all files", 0) \ - M(Bool, schema_inference_make_columns_nullable, true, "If set to true, all inferred types will be Nullable in schema inference for formats without information about nullability.", 0) \ + M(UInt64Auto, schema_inference_make_columns_nullable, 1, "If set to true, all inferred types will be Nullable in schema inference. When set to false, no columns will be converted to Nullable. When set to 'auto', ClickHosue will use information about nullability from the data..", 0) \ M(Bool, input_format_json_read_bools_as_numbers, true, "Allow to parse bools as numbers in JSON input formats", 0) \ M(Bool, input_format_json_read_bools_as_strings, true, "Allow to parse bools as strings in JSON input formats", 0) \ M(Bool, input_format_json_try_infer_numbers_from_strings, false, "Try to infer numbers from string fields while schema inference", 0) \ diff --git a/src/Formats/FormatFactory.cpp b/src/Formats/FormatFactory.cpp index a78836ff63c..8d8257b9abc 100644 --- a/src/Formats/FormatFactory.cpp +++ b/src/Formats/FormatFactory.cpp @@ -255,7 +255,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.max_bytes_to_read_for_schema_inference = settings.input_format_max_bytes_to_read_for_schema_inference; format_settings.column_names_for_schema_inference = settings.column_names_for_schema_inference; format_settings.schema_inference_hints = settings.schema_inference_hints; - format_settings.schema_inference_make_columns_nullable = settings.schema_inference_make_columns_nullable; + format_settings.schema_inference_make_columns_nullable = settings.schema_inference_make_columns_nullable.valueOr(2); format_settings.mysql_dump.table_name = settings.input_format_mysql_dump_table_name; format_settings.mysql_dump.map_column_names = settings.input_format_mysql_dump_map_column_names; format_settings.sql_insert.max_batch_size = settings.output_format_sql_insert_max_batch_size; diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index f0359218775..479b1a89adf 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -75,7 +75,7 @@ struct FormatSettings Raw }; - bool schema_inference_make_columns_nullable = true; + UInt64 schema_inference_make_columns_nullable = true; DateTimeOutputFormat date_time_output_format = DateTimeOutputFormat::Simple; diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 3c374ada9e6..c04682e8765 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -1262,7 +1262,11 @@ namespace if (checkCharCaseInsensitive('n', buf)) { if (checkStringCaseInsensitive("ull", buf)) + { + if (settings.schema_inference_make_columns_nullable == 0) + return std::make_shared(); return makeNullable(std::make_shared()); + } else if (checkStringCaseInsensitive("an", buf)) return std::make_shared(); } diff --git a/src/Processors/Formats/ISchemaReader.cpp b/src/Processors/Formats/ISchemaReader.cpp index 45523700a5d..569d4bb39e7 100644 --- a/src/Processors/Formats/ISchemaReader.cpp +++ b/src/Processors/Formats/ISchemaReader.cpp @@ -54,7 +54,7 @@ void checkFinalInferredType( type = default_type; } - if (settings.schema_inference_make_columns_nullable) + if (settings.schema_inference_make_columns_nullable == 1) type = makeNullableRecursively(type); /// In case when data for some column could contain nulls and regular values, /// resulting inferred type is Nullable. diff --git a/src/Processors/Formats/Impl/ArrowBlockInputFormat.cpp b/src/Processors/Formats/Impl/ArrowBlockInputFormat.cpp index 72a93002669..cf079e52db0 100644 --- a/src/Processors/Formats/Impl/ArrowBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/ArrowBlockInputFormat.cpp @@ -204,8 +204,11 @@ NamesAndTypesList ArrowSchemaReader::readSchema() schema = file_reader->schema(); auto header = ArrowColumnToCHColumn::arrowSchemaToCHHeader( - *schema, stream ? "ArrowStream" : "Arrow", format_settings.arrow.skip_columns_with_unsupported_types_in_schema_inference); - if (format_settings.schema_inference_make_columns_nullable) + *schema, + stream ? "ArrowStream" : "Arrow", + format_settings.arrow.skip_columns_with_unsupported_types_in_schema_inference, + format_settings.schema_inference_make_columns_nullable != 0); + if (format_settings.schema_inference_make_columns_nullable == 1) return getNamesAndRecursivelyNullableTypes(header); return header.getNamesAndTypesList(); } diff --git a/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp b/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp index ed91913de4d..bcc8bfecdc6 100644 --- a/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp +++ b/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp @@ -712,6 +712,7 @@ struct ReadColumnFromArrowColumnSettings FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior; bool allow_arrow_null_type; bool skip_columns_with_unsupported_types; + bool allow_inferring_nullable_columns; }; static ColumnWithTypeAndName readColumnFromArrowColumn( @@ -1085,7 +1086,7 @@ static ColumnWithTypeAndName readColumnFromArrowColumn( bool is_map_nested_column, const ReadColumnFromArrowColumnSettings & settings) { - bool read_as_nullable_column = arrow_column->null_count() || is_nullable_column || (type_hint && type_hint->isNullable()); + bool read_as_nullable_column = (arrow_column->null_count() || is_nullable_column || (type_hint && type_hint->isNullable())) && settings.allow_inferring_nullable_columns; if (read_as_nullable_column && arrow_column->type()->id() != arrow::Type::LIST && arrow_column->type()->id() != arrow::Type::LARGE_LIST && @@ -1149,14 +1150,16 @@ static std::shared_ptr createArrowColumn(const std::shared_ Block ArrowColumnToCHColumn::arrowSchemaToCHHeader( const arrow::Schema & schema, const std::string & format_name, - bool skip_columns_with_unsupported_types) + bool skip_columns_with_unsupported_types, + bool allow_inferring_nullable_columns) { ReadColumnFromArrowColumnSettings settings { .format_name = format_name, .date_time_overflow_behavior = FormatSettings::DateTimeOverflowBehavior::Ignore, .allow_arrow_null_type = false, - .skip_columns_with_unsupported_types = skip_columns_with_unsupported_types + .skip_columns_with_unsupported_types = skip_columns_with_unsupported_types, + .allow_inferring_nullable_columns = allow_inferring_nullable_columns, }; ColumnsWithTypeAndName sample_columns; diff --git a/src/Processors/Formats/Impl/ArrowColumnToCHColumn.h b/src/Processors/Formats/Impl/ArrowColumnToCHColumn.h index 27e9afdf763..8521cd2f410 100644 --- a/src/Processors/Formats/Impl/ArrowColumnToCHColumn.h +++ b/src/Processors/Formats/Impl/ArrowColumnToCHColumn.h @@ -34,7 +34,8 @@ public: static Block arrowSchemaToCHHeader( const arrow::Schema & schema, const std::string & format_name, - bool skip_columns_with_unsupported_types = false); + bool skip_columns_with_unsupported_types = false, + bool allow_inferring_nullable_columns = true); struct DictionaryInfo { diff --git a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp index 58bec8120f1..b0fd6789d1a 100644 --- a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp @@ -1002,7 +1002,7 @@ NamesAndTypesList NativeORCSchemaReader::readSchema() header.insert(ColumnWithTypeAndName{type, name}); } - if (format_settings.schema_inference_make_columns_nullable) + if (format_settings.schema_inference_make_columns_nullable == 1) return getNamesAndRecursivelyNullableTypes(header); return header.getNamesAndTypesList(); } diff --git a/src/Processors/Formats/Impl/ORCBlockInputFormat.cpp b/src/Processors/Formats/Impl/ORCBlockInputFormat.cpp index a3c218fa26e..2266c0b488c 100644 --- a/src/Processors/Formats/Impl/ORCBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/ORCBlockInputFormat.cpp @@ -160,8 +160,11 @@ NamesAndTypesList ORCSchemaReader::readSchema() { initializeIfNeeded(); auto header = ArrowColumnToCHColumn::arrowSchemaToCHHeader( - *schema, "ORC", format_settings.orc.skip_columns_with_unsupported_types_in_schema_inference); - if (format_settings.schema_inference_make_columns_nullable) + *schema, + "ORC", + format_settings.orc.skip_columns_with_unsupported_types_in_schema_inference, + format_settings.schema_inference_make_columns_nullable != 0); + if (format_settings.schema_inference_make_columns_nullable == 1) return getNamesAndRecursivelyNullableTypes(header); return header.getNamesAndTypesList(); } diff --git a/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp b/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp index bc5e8292192..b116070b8df 100644 --- a/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp @@ -810,8 +810,11 @@ NamesAndTypesList ParquetSchemaReader::readSchema() THROW_ARROW_NOT_OK(parquet::arrow::FromParquetSchema(metadata->schema(), &schema)); auto header = ArrowColumnToCHColumn::arrowSchemaToCHHeader( - *schema, "Parquet", format_settings.parquet.skip_columns_with_unsupported_types_in_schema_inference); - if (format_settings.schema_inference_make_columns_nullable) + *schema, + "Parquet", + format_settings.parquet.skip_columns_with_unsupported_types_in_schema_inference, + format_settings.schema_inference_make_columns_nullable != 0); + if (format_settings.schema_inference_make_columns_nullable == 1) return getNamesAndRecursivelyNullableTypes(header); return header.getNamesAndTypesList(); } diff --git a/tests/queries/0_stateless/03036_parquet_arrow_nullable.reference b/tests/queries/0_stateless/03036_parquet_arrow_nullable.reference index 985f8192f26..d15f0d8365d 100644 --- a/tests/queries/0_stateless/03036_parquet_arrow_nullable.reference +++ b/tests/queries/0_stateless/03036_parquet_arrow_nullable.reference @@ -1,40 +1,66 @@ Parquet a UInt64 a_nullable Nullable(UInt64) +a UInt64 +a_nullable UInt64 Arrow a UInt64 a_nullable Nullable(UInt64) +a UInt64 +a_nullable UInt64 Parquet b Array(UInt64) b_nullable Array(Nullable(UInt64)) +b Array(UInt64) +b_nullable Array(UInt64) Arrow b Array(Nullable(UInt64)) b_nullable Array(Nullable(UInt64)) +b Array(UInt64) +b_nullable Array(UInt64) Parquet c Tuple(\n a UInt64,\n b String) c_nullable Tuple(\n a Nullable(UInt64),\n b Nullable(String)) +c Tuple(\n a UInt64,\n b String) +c_nullable Tuple(\n a UInt64,\n b String) Arrow c Tuple(\n a UInt64,\n b String) c_nullable Tuple(\n a Nullable(UInt64),\n b Nullable(String)) +c Tuple(\n a UInt64,\n b String) +c_nullable Tuple(\n a UInt64,\n b String) Parquet d Tuple(\n a UInt64,\n b Tuple(\n a UInt64,\n b String),\n d_nullable Tuple(\n a UInt64,\n b Tuple(\n a Nullable(UInt64),\n b Nullable(String)))) +d Tuple(\n a UInt64,\n b Tuple(\n a UInt64,\n b String),\n d_nullable Tuple(\n a UInt64,\n b Tuple(\n a UInt64,\n b String))) Arrow d Tuple(\n a UInt64,\n b Tuple(\n a UInt64,\n b String),\n d_nullable Tuple(\n a UInt64,\n b Tuple(\n a Nullable(UInt64),\n b Nullable(String)))) +d Tuple(\n a UInt64,\n b Tuple(\n a UInt64,\n b String),\n d_nullable Tuple(\n a UInt64,\n b Tuple(\n a UInt64,\n b String))) Parquet e Map(UInt64, String) e_nullable Map(UInt64, Nullable(String)) +e Map(UInt64, String) +e_nullable Map(UInt64, String) Arrow e Map(UInt64, Nullable(String)) e_nullable Map(UInt64, Nullable(String)) +e Map(UInt64, String) +e_nullable Map(UInt64, String) Parquet f Map(UInt64, Map(UInt64, String)) f_nullables Map(UInt64, Map(UInt64, Nullable(String))) +f Map(UInt64, Map(UInt64, String)) +f_nullables Map(UInt64, Map(UInt64, String)) Arrow f Map(UInt64, Map(UInt64, Nullable(String))) f_nullables Map(UInt64, Map(UInt64, Nullable(String))) +f Map(UInt64, Map(UInt64, String)) +f_nullables Map(UInt64, Map(UInt64, String)) Parquet g String g_nullable Nullable(String) +g String +g_nullable String Arrow g LowCardinality(String) g_nullable LowCardinality(String) +g LowCardinality(String) +g_nullable LowCardinality(String) diff --git a/tests/queries/0_stateless/03036_parquet_arrow_nullable.sh b/tests/queries/0_stateless/03036_parquet_arrow_nullable.sh index bdd641e2b94..379756f78f3 100755 --- a/tests/queries/0_stateless/03036_parquet_arrow_nullable.sh +++ b/tests/queries/0_stateless/03036_parquet_arrow_nullable.sh @@ -14,6 +14,7 @@ for format in $formats do echo $format $CLICKHOUSE_LOCAL -q "select * from generateRandom('a UInt64, a_nullable Nullable(UInt64)', 42) limit 10 format $format" > $DATA_FILE + $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'" $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0" done @@ -21,6 +22,7 @@ for format in $formats do echo $format $CLICKHOUSE_LOCAL -q "select * from generateRandom('b Array(UInt64), b_nullable Array(Nullable(UInt64))', 42) limit 10 format $format" > $DATA_FILE + $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'" $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0" done @@ -28,6 +30,7 @@ for format in $formats do echo $format $CLICKHOUSE_LOCAL -q "select * from generateRandom('c Tuple(a UInt64, b String), c_nullable Tuple(a Nullable(UInt64), b Nullable(String))', 42) limit 10 format $format" > $DATA_FILE + $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'" $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0" done @@ -35,6 +38,7 @@ for format in $formats do echo $format $CLICKHOUSE_LOCAL -q "select * from generateRandom('d Tuple(a UInt64, b Tuple(a UInt64, b String), d_nullable Tuple(a UInt64, b Tuple(a Nullable(UInt64), b Nullable(String))))', 42) limit 10 format $format" > $DATA_FILE + $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'" $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0" done @@ -42,6 +46,7 @@ for format in $formats do echo $format $CLICKHOUSE_LOCAL -q "select * from generateRandom('e Map(UInt64, String), e_nullable Map(UInt64, Nullable(String))', 42) limit 10 format $format" > $DATA_FILE + $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'" $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0" done @@ -49,6 +54,7 @@ for format in $formats do echo $format $CLICKHOUSE_LOCAL -q "select * from generateRandom('f Map(UInt64, Map(UInt64, String)), f_nullables Map(UInt64, Map(UInt64, Nullable(String)))', 42) limit 10 format $format" > $DATA_FILE + $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'" $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0" done @@ -56,6 +62,7 @@ for format in $formats do echo $format $CLICKHOUSE_LOCAL -q "select * from generateRandom('g LowCardinality(String), g_nullable LowCardinality(Nullable(String))', 42) limit 10 settings output_format_arrow_low_cardinality_as_dictionary=1, allow_suspicious_low_cardinality_types=1 format $format" > $DATA_FILE + $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'" $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0" done From 70708fd5dcf633d4d3147240195554587f4fb14f Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 13 Aug 2024 19:19:02 +0000 Subject: [PATCH 057/260] Update docs, make better --- docs/en/interfaces/schema-inference.md | 7 ++----- docs/en/operations/settings/settings-formats.md | 4 ++-- src/Processors/Formats/ISchemaReader.cpp | 5 ----- 3 files changed, 4 insertions(+), 12 deletions(-) diff --git a/docs/en/interfaces/schema-inference.md b/docs/en/interfaces/schema-inference.md index 05fae994cbe..5b3cd179e21 100644 --- a/docs/en/interfaces/schema-inference.md +++ b/docs/en/interfaces/schema-inference.md @@ -1385,7 +1385,7 @@ DESC format(JSONEachRow, '{"id" : 1, "age" : 25, "name" : "Josh", "status" : nul #### schema_inference_make_columns_nullable Controls making inferred types `Nullable` in schema inference for formats without information about nullability. -If the setting is enabled, all inferred type will be `Nullable`, if disabled, the inferred type will be `Nullable` only if `input_format_null_as_default` is disabled and the column contains `NULL` in a sample that is parsed during schema inference. +If the setting is enabled, all inferred type will be `Nullable`, if disabled, the inferred type will never be `Nullable`, if set to `auto`, the inferred type will be `Nullable` only if the column contains `NULL` in a sample that is parsed during schema inference or file metadata contains information about column nullability. Enabled by default. @@ -1408,15 +1408,13 @@ DESC format(JSONEachRow, $$ └─────────┴─────────────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘ ``` ```sql -SET schema_inference_make_columns_nullable = 0; -SET input_format_null_as_default = 0; +SET schema_inference_make_columns_nullable = 'auto'; DESC format(JSONEachRow, $$ {"id" : 1, "age" : 25, "name" : "Josh", "status" : null, "hobbies" : ["football", "cooking"]} {"id" : 2, "age" : 19, "name" : "Alan", "status" : "married", "hobbies" : ["tennis", "art"]} $$) ``` ```response - ┌─name────┬─type─────────────┬─default_type─┬─default_expression─┬─comment─┬─codec_expression─┬─ttl_expression─┐ │ id │ Int64 │ │ │ │ │ │ │ age │ Int64 │ │ │ │ │ │ @@ -1428,7 +1426,6 @@ DESC format(JSONEachRow, $$ ```sql SET schema_inference_make_columns_nullable = 0; -SET input_format_null_as_default = 1; DESC format(JSONEachRow, $$ {"id" : 1, "age" : 25, "name" : "Josh", "status" : null, "hobbies" : ["football", "cooking"]} {"id" : 2, "age" : 19, "name" : "Alan", "status" : "married", "hobbies" : ["tennis", "art"]} diff --git a/docs/en/operations/settings/settings-formats.md b/docs/en/operations/settings/settings-formats.md index f8b40cd81ac..57812ef0e03 100644 --- a/docs/en/operations/settings/settings-formats.md +++ b/docs/en/operations/settings/settings-formats.md @@ -171,8 +171,8 @@ If the `schema_inference_hints` is not formated properly, or if there is a typo ## schema_inference_make_columns_nullable {#schema_inference_make_columns_nullable} -Controls making inferred types `Nullable` in schema inference for formats without information about nullability. -If the setting is enabled, the inferred type will be `Nullable` only if column contains `NULL` in a sample that is parsed during schema inference. +Controls making inferred types `Nullable` in schema inference. +If the setting is enabled, all inferred type will be `Nullable`, if disabled, the inferred type will never be `Nullable`, if set to `auto`, the inferred type will be `Nullable` only if the column contains `NULL` in a sample that is parsed during schema inference or file metadata contains information about column nullability. Default value: `true`. diff --git a/src/Processors/Formats/ISchemaReader.cpp b/src/Processors/Formats/ISchemaReader.cpp index 569d4bb39e7..e002e64b7e5 100644 --- a/src/Processors/Formats/ISchemaReader.cpp +++ b/src/Processors/Formats/ISchemaReader.cpp @@ -56,11 +56,6 @@ void checkFinalInferredType( if (settings.schema_inference_make_columns_nullable == 1) type = makeNullableRecursively(type); - /// In case when data for some column could contain nulls and regular values, - /// resulting inferred type is Nullable. - /// If input_format_null_as_default is enabled, we should remove Nullable type. - else if (settings.null_as_default) - type = removeNullable(type); } void ISchemaReader::transformTypesIfNeeded(DB::DataTypePtr & type, DB::DataTypePtr & new_type) From 0abca8b7ddbafa37da5b1196b21fb816999fd334 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 14 Aug 2024 09:57:59 +0800 Subject: [PATCH 058/260] fix doc --- .../sql-reference/functions/string-replace-functions.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/en/sql-reference/functions/string-replace-functions.md b/docs/en/sql-reference/functions/string-replace-functions.md index 1caa6215b6b..d086c9ee64b 100644 --- a/docs/en/sql-reference/functions/string-replace-functions.md +++ b/docs/en/sql-reference/functions/string-replace-functions.md @@ -263,8 +263,8 @@ overlay(s, replace, position[, length]) - `s`: A string type [String](../data-types/string.md). - `replace`: A string type [String](../data-types/string.md). -- `position`: An integer type [Int](../data-types/int.md). -- `length`: Optional. An integer type [Int](../data-types/int.md). +- `position`: An integer type [Int](../data-types/int-uint.md). +- `length`: Optional. An integer type [Int](../data-types/int-uint.md). **Returned value** @@ -300,8 +300,8 @@ overlayUTF8(s, replace, position[, length]) - `s`: A string type [String](../data-types/string.md). - `replace`: A string type [String](../data-types/string.md). -- `position`: An integer type [Int](../data-types/int.md). -- `length`: Optional. An integer type [Int](../data-types/int.md). +- `position`: An integer type [Int](../data-types/int-uint.md). +- `length`: Optional. An integer type [Int](../data-types/int-uint.md). **Returned value** From f740cf4eaa71621fb518c6d5668e8356f452a979 Mon Sep 17 00:00:00 2001 From: Antonio Andelic Date: Wed, 14 Aug 2024 09:54:03 +0200 Subject: [PATCH 059/260] Fix data race on SampleKey --- src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp index a6ef0063069..6efd3a5c97f 100644 --- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp +++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp @@ -369,7 +369,7 @@ MergeTreeDataSelectSamplingData MergeTreeDataSelectExecutor::getSampling( /// If sample and final are used together no need to calculate sampling expression twice. /// The first time it was calculated for final, because sample key is a part of the PK. /// So, assume that we already have calculated column. - ASTPtr sampling_key_ast = metadata_snapshot->getSamplingKeyAST(); + ASTPtr sampling_key_ast; if (final) { @@ -377,6 +377,12 @@ MergeTreeDataSelectSamplingData MergeTreeDataSelectExecutor::getSampling( /// We do spoil available_real_columns here, but it is not used later. available_real_columns.emplace_back(sampling_key.column_names[0], std::move(sampling_column_type)); } + else + { + sampling_key_ast = metadata_snapshot->getSamplingKeyAST()->clone(); + } + + chassert(sampling_key_ast != nullptr); if (has_lower_limit) { From e2feaefcaf0e88f86f303c068edcbdacaeb67252 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Wed, 14 Aug 2024 15:13:49 +0200 Subject: [PATCH 060/260] Update src/Core/Settings.h Co-authored-by: Alexey Katsman --- src/Core/Settings.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/Settings.h b/src/Core/Settings.h index bc9c6daab1b..2417ddd39e8 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -1116,7 +1116,7 @@ class IColumn; M(String, column_names_for_schema_inference, "", "The list of column names to use in schema inference for formats without column names. The format: 'column1,column2,column3,...'", 0) \ M(String, schema_inference_hints, "", "The list of column names and types to use in schema inference for formats without column names. The format: 'column_name1 column_type1, column_name2 column_type2, ...'", 0) \ M(SchemaInferenceMode, schema_inference_mode, "default", "Mode of schema inference. 'default' - assume that all files have the same schema and schema can be inferred from any file, 'union' - files can have different schemas and the resulting schema should be the a union of schemas of all files", 0) \ - M(UInt64Auto, schema_inference_make_columns_nullable, 1, "If set to true, all inferred types will be Nullable in schema inference. When set to false, no columns will be converted to Nullable. When set to 'auto', ClickHosue will use information about nullability from the data..", 0) \ + M(UInt64Auto, schema_inference_make_columns_nullable, 1, "If set to true, all inferred types will be Nullable in schema inference. When set to false, no columns will be converted to Nullable. When set to 'auto', ClickHouse will use information about nullability from the data.", 0) \ M(Bool, input_format_json_read_bools_as_numbers, true, "Allow to parse bools as numbers in JSON input formats", 0) \ M(Bool, input_format_json_read_bools_as_strings, true, "Allow to parse bools as strings in JSON input formats", 0) \ M(Bool, input_format_json_try_infer_numbers_from_strings, false, "Try to infer numbers from string fields while schema inference", 0) \ From ccb7ecb9a22ddeabe93a5b907e3ad688b04966b4 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Wed, 14 Aug 2024 15:13:57 +0200 Subject: [PATCH 061/260] Update src/Formats/FormatSettings.h Co-authored-by: Alexey Katsman --- src/Formats/FormatSettings.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index 479b1a89adf..81b34ff0c55 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -75,7 +75,7 @@ struct FormatSettings Raw }; - UInt64 schema_inference_make_columns_nullable = true; + UInt64 schema_inference_make_columns_nullable = 1; DateTimeOutputFormat date_time_output_format = DateTimeOutputFormat::Simple; From 28b0aad3f9e54beed27ee384ab81312233abaa84 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Wed, 14 Aug 2024 15:16:34 +0200 Subject: [PATCH 062/260] Fix python style --- tests/clickhouse-test | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 515b519af3e..8c2da7334d4 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -1975,9 +1975,6 @@ class TestSuite: ) return random_settings_limits - def is_shebang(line: str) -> bool: - return line.startswith("#!") - def find_tag_line(lines, comment_sign): for line in lines: if line.startswith(comment_sign) and line[ @@ -2077,9 +2074,9 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[ - str, Dict[str, (int, int)] - ] = all_tags_and_random_settings_limits[1] + self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = ( + all_tags_and_random_settings_limits[1] + ) self.sequential_tests = [] self.parallel_tests = [] for test_name in self.all_tests: From 844cdd8937cce17060ea8b54fdfc2428d3015f44 Mon Sep 17 00:00:00 2001 From: Blargian Date: Wed, 14 Aug 2024 20:38:09 +0200 Subject: [PATCH 063/260] update toInterval functions --- .../functions/type-conversion-functions.md | 400 +++++++++++++++++- 1 file changed, 380 insertions(+), 20 deletions(-) diff --git a/docs/en/sql-reference/functions/type-conversion-functions.md b/docs/en/sql-reference/functions/type-conversion-functions.md index 1e618b8cdab..cd6fd9ab839 100644 --- a/docs/en/sql-reference/functions/type-conversion-functions.md +++ b/docs/en/sql-reference/functions/type-conversion-functions.md @@ -4866,30 +4866,23 @@ Result: └───────┴───────────────┴──────┴──────────────┴──────────────┴──────────────────────┘ ``` -## toInterval(Year\|Quarter\|Month\|Week\|Day\|Hour\|Minute\|Second) +## toIntervalYear -Converts a Number type argument to an [Interval](../data-types/special-data-types/interval.md) data type. +Returns an interval of `n` years of data type [IntervalYear](../data-types/special-data-types/interval.md). **Syntax** ``` sql -toIntervalSecond(number) -toIntervalMinute(number) -toIntervalHour(number) -toIntervalDay(number) -toIntervalWeek(number) -toIntervalMonth(number) -toIntervalQuarter(number) -toIntervalYear(number) +toIntervalYear(n) ``` **Arguments** -- `number` — Duration of interval. Positive integer number. +- `n` — Number of years. Positive integer number. [Int*](../data-types/int-uint.md). **Returned values** -- The value in `Interval` data type. +- Interval of `n` years. [IntervalYear](../data-types/special-data-types/interval.md). **Example** @@ -4898,19 +4891,386 @@ Query: ``` sql WITH toDate('2019-01-01') AS date, - INTERVAL 1 WEEK AS interval_week, - toIntervalWeek(1) AS interval_to_week -SELECT - date + interval_week, - date + interval_to_week; + toIntervalYear(1) AS interval_to_year +SELECT date + interval_to_year ``` Result: ```response -┌─plus(date, interval_week)─┬─plus(date, interval_to_week)─┐ -│ 2019-01-08 │ 2019-01-08 │ -└───────────────────────────┴──────────────────────────────┘ +┌─plus(date, interval_to_year)─┐ +│ 2020-01-01 │ +└──────────────────────────────┘ +``` + +## toIntervalQuarter + +Returns an interval of `n` quarters of data type [IntervalQuarter](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalQuarter(n) +``` + +**Arguments** + +- `n` — Number of quarters. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` quarters. [IntervalQuarter](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDate('2019-01-01') AS date, + toIntervalQuarter(1) AS interval_to_quarter +SELECT date + interval_to_quarter +``` + +Result: + +```response +┌─plus(date, interval_to_quarter)─┐ +│ 2019-04-01 │ +└─────────────────────────────────┘ +``` + +## toIntervalMonth + +Returns an interval of `n` months of data type [IntervalMonth](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalMonth(n) +``` + +**Arguments** + +- `n` — Number of m. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` months. [IntervalMonth](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDate('2019-01-01') AS date, + toIntervalMonth(1) AS interval_to_month +SELECT date + interval_to_month +``` + +Result: + +```response +┌─plus(date, interval_to_month)─┐ +│ 2019-02-01 │ +└───────────────────────────────┘ +``` + +## toIntervalWeek + +Returns an interval of `n` weeks of data type [IntervalWeek](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalWeek(n) +``` + +**Arguments** + +- `n` — Number of weeks. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` weeks. [IntervalWeek](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDate('2019-01-01') AS date, + toIntervalWeek(1) AS interval_to_week +SELECT date + interval_to_week +``` + +Result: + +```response +┌─plus(date, interval_to_week)─┐ +│ 2019-01-08 │ +└──────────────────────────────┘ +``` + +## toIntervalDay + +Returns an interval of `n` days of data type [IntervalDay](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalDay(n) +``` + +**Arguments** + +- `n` — Number of days. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` days. [IntervalDay](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDate('2019-01-01') AS date, + toIntervalDay(5) AS interval_to_days +SELECT date + interval_to_days +``` + +Result: + +```response +┌─plus(date, interval_to_days)─┐ +│ 2019-01-06 │ +└──────────────────────────────┘ +``` + +## toIntervalHour + +Returns an interval of `n` hours of data type [IntervalHour](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalHour(n) +``` + +**Arguments** + +- `n` — Number of hours. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` hours. [IntervalHour](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDate('2019-01-01') AS date, + toIntervalHour(12) AS interval_to_hours +SELECT date + interval_to_hours +``` + +Result: + +```response +┌─plus(date, interval_to_hours)─┐ +│ 2019-01-01 12:00:00 │ +└───────────────────────────────┘ +``` + +## toIntervalMinute + +Returns an interval of `n` minutes of data type [IntervalMinute](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalMinute(n) +``` + +**Arguments** + +- `n` — Number of minutes. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` minutes. [IntervalMinute](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDate('2019-01-01') AS date, + toIntervalMinute(12) AS interval_to_minutes +SELECT date + interval_to_minutes +``` + +Result: + +```response +┌─plus(date, interval_to_minutes)─┐ +│ 2019-01-01 00:12:00 │ +└─────────────────────────────────┘ +``` + +## toIntervalSecond + +Returns an interval of `n` seconds of data type [IntervalSecond](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalSecond(n) +``` + +**Arguments** + +- `n` — Number of seconds. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` seconds. [IntervalSecond](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDate('2019-01-01') AS date, + toIntervalSecond(30) AS interval_to_seconds +SELECT date + interval_to_seconds +``` + +Result: + +```response +┌─plus(date, interval_to_seconds)─┐ +│ 2019-01-01 00:00:30 │ +└─────────────────────────────────┘ +``` + +## toIntervalMillisecond + +Returns an interval of `n` milliseconds of data type [IntervalMillisecond](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalMillisecond(n) +``` + +**Arguments** + +- `n` — Number of milliseconds. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` milliseconds. [IntervalMilliseconds](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDateTime('2019-01-01') AS date, + toIntervalMillisecond(30) AS interval_to_milliseconds +SELECT date + interval_to_milliseconds +``` + +Result: + +```response +┌─plus(date, interval_to_milliseconds)─┐ +│ 2019-01-01 00:00:00.030 │ +└──────────────────────────────────────┘ +``` + +## toIntervalMicrosecond + +Returns an interval of `n` microseconds of data type [IntervalMicrosecond](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalMicrosecond(n) +``` + +**Arguments** + +- `n` — Number of microseconds. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` microseconds. [IntervalMicrosecond](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDateTime('2019-01-01') AS date, + toIntervalMicrosecond(30) AS interval_to_microseconds +SELECT date + interval_to_microseconds +``` + +Result: + +```response +┌─plus(date, interval_to_microseconds)─┐ +│ 2019-01-01 00:00:00.000030 │ +└──────────────────────────────────────┘ +``` + +## toIntervalNanosecond + +Returns an interval of `n` nanoseconds of data type [IntervalNanosecond](../data-types/special-data-types/interval.md). + +**Syntax** + +``` sql +toIntervalNanosecond(n) +``` + +**Arguments** + +- `n` — Number of nanoseconds. Positive integer number. [Int*](../data-types/int-uint.md). + +**Returned values** + +- Interval of `n` nanoseconds. [IntervalNanosecond](../data-types/special-data-types/interval.md). + +**Example** + +Query: + +``` sql +WITH + toDateTime('2019-01-01') AS date, + toIntervalNanosecond(30) AS interval_to_nanoseconds +SELECT date + interval_to_nanoseconds +``` + +Result: + +```response +┌─plus(date, interval_to_nanoseconds)─┐ +│ 2019-01-01 00:00:00.000000030 │ +└─────────────────────────────────────┘ ``` ## parseDateTime From b82c231886f2496c01b288a138663c4d430fc7b2 Mon Sep 17 00:00:00 2001 From: shiyer7474 Date: Thu, 15 Aug 2024 10:37:13 +0000 Subject: [PATCH 064/260] Code review feedback - used removeNullable() --- src/Analyzer/ConstantNode.cpp | 11 +++-------- .../03222_datetime64_small_value_const.sql | 2 +- 2 files changed, 4 insertions(+), 9 deletions(-) diff --git a/src/Analyzer/ConstantNode.cpp b/src/Analyzer/ConstantNode.cpp index 3d0f448da4b..3a99ad08ad8 100644 --- a/src/Analyzer/ConstantNode.cpp +++ b/src/Analyzer/ConstantNode.cpp @@ -177,15 +177,10 @@ ASTPtr ConstantNode::toASTImpl(const ConvertToASTOptions & options) const * It could also lead to ambiguous parsing because we don't know if the string literal represents a date or a Decimal64 literal. * For this reason, we use a string literal representing a date instead of a Decimal64 literal. */ - if ((WhichDataType(constant_value_type->getTypeId()).isDateTime64()) || - (WhichDataType(constant_value_type->getTypeId()).isNullable() && WhichDataType((typeid_cast(constant_value_type.get()))->getNestedType()->getTypeId()).isDateTime64())) + const auto & constant_value_end_type = removeNullable(constant_value_type); /// if Nullable + if (WhichDataType(constant_value_end_type->getTypeId()).isDateTime64()) { - const DataTypeDateTime64 * date_time_type = nullptr; - if (WhichDataType(constant_value_type->getTypeId()).isNullable()) - date_time_type = typeid_cast((typeid_cast(constant_value_type.get()))->getNestedType().get()); - else - date_time_type = typeid_cast(constant_value_type.get()); - + const auto * date_time_type = typeid_cast(constant_value_end_type.get()); DecimalField decimal_value; if (constant_value_literal.tryGet>(decimal_value)) { diff --git a/tests/queries/0_stateless/03222_datetime64_small_value_const.sql b/tests/queries/0_stateless/03222_datetime64_small_value_const.sql index 6999ba9662a..af06a622f8d 100644 --- a/tests/queries/0_stateless/03222_datetime64_small_value_const.sql +++ b/tests/queries/0_stateless/03222_datetime64_small_value_const.sql @@ -17,7 +17,7 @@ create table shard_0.dt64_03222(id UInt64, dt DateTime64(3)) engine = MergeTree create table shard_1.dt64_03222(id UInt64, dt DateTime64(3)) engine = MergeTree order by id; create table distr_03222_dt64 (id UInt64, dt DateTime64(3)) engine = Distributed(test_cluster_two_shards_different_databases, '', dt64_03222); -insert into shard_0.dt64_03222 values(1, toDateTime64('1970-01-01 00:00:00.000',3)) +insert into shard_0.dt64_03222 values(1, toDateTime64('1970-01-01 00:00:00.000',3)); insert into shard_0.dt64_03222 values(2, toDateTime64('1970-01-01 00:00:02.456',3)); insert into shard_1.dt64_03222 values(3, toDateTime64('1970-01-01 00:00:04.811',3)); insert into shard_1.dt64_03222 values(4, toDateTime64('1970-01-01 00:10:05',3)); From 92a9b29b45c254e670fe9f67114b5af890bfb5cb Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 15 Aug 2024 22:25:21 +0800 Subject: [PATCH 065/260] devirtualize format reader --- .../Formats/Impl/BinaryRowInputFormat.cpp | 4 +- .../Formats/Impl/BinaryRowInputFormat.h | 7 ++- .../Formats/Impl/CSVRowInputFormat.cpp | 2 +- .../Formats/Impl/CSVRowInputFormat.h | 7 ++- .../Impl/CustomSeparatedRowInputFormat.h | 3 +- .../Impl/JSONCompactEachRowRowInputFormat.h | 4 +- .../Impl/JSONCompactRowInputFormat.cpp | 2 +- .../Formats/Impl/JSONCompactRowInputFormat.h | 4 +- .../Formats/Impl/TabSeparatedRowInputFormat.h | 4 +- .../RowInputFormatWithNamesAndTypes.cpp | 61 +++++++++++++------ .../Formats/RowInputFormatWithNamesAndTypes.h | 5 +- 11 files changed, 67 insertions(+), 36 deletions(-) diff --git a/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp b/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp index c5336f3bcc7..b549f2de975 100644 --- a/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp @@ -15,8 +15,8 @@ namespace ErrorCodes } template -BinaryRowInputFormat::BinaryRowInputFormat(ReadBuffer & in_, const Block & header, Params params_, bool with_names_, bool with_types_, const FormatSettings & format_settings_) - : RowInputFormatWithNamesAndTypes( +BinaryRowInputFormat::BinaryRowInputFormat(ReadBuffer & in_, const Block & header, IRowInputFormat::Params params_, bool with_names_, bool with_types_, const FormatSettings & format_settings_) + : RowInputFormatWithNamesAndTypes>( header, in_, params_, diff --git a/src/Processors/Formats/Impl/BinaryRowInputFormat.h b/src/Processors/Formats/Impl/BinaryRowInputFormat.h index 6f2042d1315..6a4ca8f6418 100644 --- a/src/Processors/Formats/Impl/BinaryRowInputFormat.h +++ b/src/Processors/Formats/Impl/BinaryRowInputFormat.h @@ -10,13 +10,16 @@ namespace DB class ReadBuffer; +template +class BinaryFormatReader; + /** A stream for inputting data in a binary line-by-line format. */ template -class BinaryRowInputFormat final : public RowInputFormatWithNamesAndTypes +class BinaryRowInputFormat final : public RowInputFormatWithNamesAndTypes> { public: - BinaryRowInputFormat(ReadBuffer & in_, const Block & header, Params params_, bool with_names_, bool with_types_, const FormatSettings & format_settings_); + BinaryRowInputFormat(ReadBuffer & in_, const Block & header, IRowInputFormat::Params params_, bool with_names_, bool with_types_, const FormatSettings & format_settings_); String getName() const override { return "BinaryRowInputFormat"; } diff --git a/src/Processors/Formats/Impl/CSVRowInputFormat.cpp b/src/Processors/Formats/Impl/CSVRowInputFormat.cpp index b7f84748f61..cf58a4057c8 100644 --- a/src/Processors/Formats/Impl/CSVRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/CSVRowInputFormat.cpp @@ -61,7 +61,7 @@ CSVRowInputFormat::CSVRowInputFormat( bool with_names_, bool with_types_, const FormatSettings & format_settings_, - std::unique_ptr format_reader_) + std::unique_ptr format_reader_) : RowInputFormatWithNamesAndTypes( header_, *in_, diff --git a/src/Processors/Formats/Impl/CSVRowInputFormat.h b/src/Processors/Formats/Impl/CSVRowInputFormat.h index fe4d4e3be08..86af5028438 100644 --- a/src/Processors/Formats/Impl/CSVRowInputFormat.h +++ b/src/Processors/Formats/Impl/CSVRowInputFormat.h @@ -1,7 +1,6 @@ #pragma once #include -#include #include #include @@ -13,10 +12,12 @@ namespace DB { +class CSVFormatReader; + /** A stream for inputting data in csv format. * Does not conform with https://tools.ietf.org/html/rfc4180 because it skips spaces and tabs between values. */ -class CSVRowInputFormat : public RowInputFormatWithNamesAndTypes +class CSVRowInputFormat : public RowInputFormatWithNamesAndTypes { public: /** with_names - in the first line the header with column names @@ -32,7 +33,7 @@ public: protected: CSVRowInputFormat(const Block & header_, std::shared_ptr in_, const Params & params_, - bool with_names_, bool with_types_, const FormatSettings & format_settings_, std::unique_ptr format_reader_); + bool with_names_, bool with_types_, const FormatSettings & format_settings_, std::unique_ptr format_reader_); CSVRowInputFormat(const Block & header_, std::shared_ptr in_buf_, const Params & params_, bool with_names_, bool with_types_, const FormatSettings & format_settings_); diff --git a/src/Processors/Formats/Impl/CustomSeparatedRowInputFormat.h b/src/Processors/Formats/Impl/CustomSeparatedRowInputFormat.h index 58f78e5af42..b1d35947ba8 100644 --- a/src/Processors/Formats/Impl/CustomSeparatedRowInputFormat.h +++ b/src/Processors/Formats/Impl/CustomSeparatedRowInputFormat.h @@ -9,7 +9,8 @@ namespace DB { -class CustomSeparatedRowInputFormat final : public RowInputFormatWithNamesAndTypes +class CustomSeparatedFormatReader; +class CustomSeparatedRowInputFormat final : public RowInputFormatWithNamesAndTypes { public: CustomSeparatedRowInputFormat( diff --git a/src/Processors/Formats/Impl/JSONCompactEachRowRowInputFormat.h b/src/Processors/Formats/Impl/JSONCompactEachRowRowInputFormat.h index ebeb939e7fa..50589329073 100644 --- a/src/Processors/Formats/Impl/JSONCompactEachRowRowInputFormat.h +++ b/src/Processors/Formats/Impl/JSONCompactEachRowRowInputFormat.h @@ -11,7 +11,7 @@ namespace DB { class ReadBuffer; - +class JSONCompactEachRowFormatReader; /** A stream for reading data in a bunch of formats: * - JSONCompactEachRow @@ -20,7 +20,7 @@ class ReadBuffer; * - JSONCompactStringsEachRowWithNamesAndTypes * */ -class JSONCompactEachRowRowInputFormat final : public RowInputFormatWithNamesAndTypes +class JSONCompactEachRowRowInputFormat final : public RowInputFormatWithNamesAndTypes { public: JSONCompactEachRowRowInputFormat( diff --git a/src/Processors/Formats/Impl/JSONCompactRowInputFormat.cpp b/src/Processors/Formats/Impl/JSONCompactRowInputFormat.cpp index 63066fc8220..63ced05dd3a 100644 --- a/src/Processors/Formats/Impl/JSONCompactRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/JSONCompactRowInputFormat.cpp @@ -14,7 +14,7 @@ namespace ErrorCodes JSONCompactRowInputFormat::JSONCompactRowInputFormat( const Block & header_, ReadBuffer & in_, Params params_, const FormatSettings & format_settings_) - : RowInputFormatWithNamesAndTypes( + : RowInputFormatWithNamesAndTypes( header_, in_, params_, false, false, false, format_settings_, std::make_unique(in_, format_settings_)) { } diff --git a/src/Processors/Formats/Impl/JSONCompactRowInputFormat.h b/src/Processors/Formats/Impl/JSONCompactRowInputFormat.h index 3a93e7149b0..eb70f6ec2a3 100644 --- a/src/Processors/Formats/Impl/JSONCompactRowInputFormat.h +++ b/src/Processors/Formats/Impl/JSONCompactRowInputFormat.h @@ -5,8 +5,8 @@ namespace DB { - -class JSONCompactRowInputFormat final : public RowInputFormatWithNamesAndTypes +class JSONCompactFormatReader; +class JSONCompactRowInputFormat final : public RowInputFormatWithNamesAndTypes { public: JSONCompactRowInputFormat(const Block & header_, ReadBuffer & in_, Params params_, const FormatSettings & format_settings_); diff --git a/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.h b/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.h index 32abd532a52..3c6efe9ac4c 100644 --- a/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.h +++ b/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.h @@ -10,9 +10,11 @@ namespace DB { +class TabSeparatedFormatReader; + /** A stream to input data in tsv format. */ -class TabSeparatedRowInputFormat final : public RowInputFormatWithNamesAndTypes +class TabSeparatedRowInputFormat final : public RowInputFormatWithNamesAndTypes { public: /** with_names - the first line is the header with the names of the columns diff --git a/src/Processors/Formats/RowInputFormatWithNamesAndTypes.cpp b/src/Processors/Formats/RowInputFormatWithNamesAndTypes.cpp index ae30d741c2f..5701b80ecc2 100644 --- a/src/Processors/Formats/RowInputFormatWithNamesAndTypes.cpp +++ b/src/Processors/Formats/RowInputFormatWithNamesAndTypes.cpp @@ -1,14 +1,20 @@ -#include -#include -#include -#include #include #include -#include -#include -#include -#include +#include +#include #include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include namespace DB @@ -44,7 +50,8 @@ namespace } } -RowInputFormatWithNamesAndTypes::RowInputFormatWithNamesAndTypes( +template +RowInputFormatWithNamesAndTypes::RowInputFormatWithNamesAndTypes( const Block & header_, ReadBuffer & in_, const Params & params_, @@ -52,7 +59,7 @@ RowInputFormatWithNamesAndTypes::RowInputFormatWithNamesAndTypes( bool with_names_, bool with_types_, const FormatSettings & format_settings_, - std::unique_ptr format_reader_, + std::unique_ptr format_reader_, bool try_detect_header_) : RowInputFormatWithDiagnosticInfo(header_, in_, params_) , format_settings(format_settings_) @@ -66,7 +73,8 @@ RowInputFormatWithNamesAndTypes::RowInputFormatWithNamesAndTypes( column_indexes_by_names = getPort().getHeader().getNamesToIndexesMap(); } -void RowInputFormatWithNamesAndTypes::readPrefix() +template +void RowInputFormatWithNamesAndTypes::readPrefix() { /// Search and remove BOM only in textual formats (CSV, TSV etc), not in binary ones (RowBinary*). /// Also, we assume that column name or type cannot contain BOM, so, if format has header, @@ -138,7 +146,8 @@ void RowInputFormatWithNamesAndTypes::readPrefix() } } -void RowInputFormatWithNamesAndTypes::tryDetectHeader(std::vector & column_names_out, std::vector & type_names_out) +template +void RowInputFormatWithNamesAndTypes::tryDetectHeader(std::vector & column_names_out, std::vector & type_names_out) { auto & read_buf = getReadBuffer(); PeekableReadBuffer * peekable_buf = dynamic_cast(&read_buf); @@ -201,7 +210,8 @@ void RowInputFormatWithNamesAndTypes::tryDetectHeader(std::vector & colu peekable_buf->dropCheckpoint(); } -bool RowInputFormatWithNamesAndTypes::readRow(MutableColumns & columns, RowReadExtension & ext) +template +bool RowInputFormatWithNamesAndTypes::readRow(MutableColumns & columns, RowReadExtension & ext) { if (unlikely(end_of_stream)) return false; @@ -280,7 +290,8 @@ bool RowInputFormatWithNamesAndTypes::readRow(MutableColumns & columns, RowReadE return true; } -size_t RowInputFormatWithNamesAndTypes::countRows(size_t max_block_size) +template +size_t RowInputFormatWithNamesAndTypes::countRows(size_t max_block_size) { if (unlikely(end_of_stream)) return 0; @@ -304,7 +315,8 @@ size_t RowInputFormatWithNamesAndTypes::countRows(size_t max_block_size) return num_rows; } -void RowInputFormatWithNamesAndTypes::resetParser() +template +void RowInputFormatWithNamesAndTypes::resetParser() { RowInputFormatWithDiagnosticInfo::resetParser(); column_mapping->column_indexes_for_input_fields.clear(); @@ -313,7 +325,8 @@ void RowInputFormatWithNamesAndTypes::resetParser() end_of_stream = false; } -void RowInputFormatWithNamesAndTypes::tryDeserializeField(const DataTypePtr & type, IColumn & column, size_t file_column) +template +void RowInputFormatWithNamesAndTypes::tryDeserializeField(const DataTypePtr & type, IColumn & column, size_t file_column) { const auto & index = column_mapping->column_indexes_for_input_fields[file_column]; if (index) @@ -328,7 +341,8 @@ void RowInputFormatWithNamesAndTypes::tryDeserializeField(const DataTypePtr & ty } } -bool RowInputFormatWithNamesAndTypes::parseRowAndPrintDiagnosticInfo(MutableColumns & columns, WriteBuffer & out) +template +bool RowInputFormatWithNamesAndTypes::parseRowAndPrintDiagnosticInfo(MutableColumns & columns, WriteBuffer & out) { if (in->eof()) { @@ -374,12 +388,14 @@ bool RowInputFormatWithNamesAndTypes::parseRowAndPrintDiagnosticInfo(MutableColu return format_reader->parseRowEndWithDiagnosticInfo(out); } -bool RowInputFormatWithNamesAndTypes::isGarbageAfterField(size_t index, ReadBuffer::Position pos) +template +bool RowInputFormatWithNamesAndTypes::isGarbageAfterField(size_t index, ReadBuffer::Position pos) { return format_reader->isGarbageAfterField(index, pos); } -void RowInputFormatWithNamesAndTypes::setReadBuffer(ReadBuffer & in_) +template +void RowInputFormatWithNamesAndTypes::setReadBuffer(ReadBuffer & in_) { format_reader->setReadBuffer(in_); IInputFormat::setReadBuffer(in_); @@ -582,5 +598,12 @@ void FormatWithNamesAndTypesSchemaReader::transformTypesIfNeeded(DB::DataTypePtr transformInferredTypesIfNeeded(type, new_type, format_settings); } +template class RowInputFormatWithNamesAndTypes; +template class RowInputFormatWithNamesAndTypes; +template class RowInputFormatWithNamesAndTypes; +template class RowInputFormatWithNamesAndTypes; +template class RowInputFormatWithNamesAndTypes; +template class RowInputFormatWithNamesAndTypes>; +template class RowInputFormatWithNamesAndTypes>; } diff --git a/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h b/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h index b7d9507151e..cd836cb00dc 100644 --- a/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h +++ b/src/Processors/Formats/RowInputFormatWithNamesAndTypes.h @@ -26,6 +26,7 @@ class FormatWithNamesAndTypesReader; /// will be compared types from header. /// It's important that firstly this class reads/skips names and only /// then reads/skips types. So you can this invariant. +template class RowInputFormatWithNamesAndTypes : public RowInputFormatWithDiagnosticInfo { protected: @@ -41,7 +42,7 @@ protected: bool with_names_, bool with_types_, const FormatSettings & format_settings_, - std::unique_ptr format_reader_, + std::unique_ptr format_reader_, bool try_detect_header_ = false); void resetParser() override; @@ -70,7 +71,7 @@ private: bool is_header_detected = false; protected: - std::unique_ptr format_reader; + std::unique_ptr format_reader; Block::NameMap column_indexes_by_names; }; From 5ff4d990e189dfee42eb57f567a5ff6313cfa8d8 Mon Sep 17 00:00:00 2001 From: Max Kainov Date: Fri, 16 Aug 2024 11:11:11 +0200 Subject: [PATCH 066/260] CI: Auto Releases in prod --- .github/workflows/auto_releases.yml | 76 ++++++++++------------------ .github/workflows/create_release.yml | 2 + tests/ci/auto_release.py | 9 ++++ tests/ci/ci_utils.py | 24 ++++++--- 4 files changed, 54 insertions(+), 57 deletions(-) diff --git a/.github/workflows/auto_releases.yml b/.github/workflows/auto_releases.yml index c159907187c..28483ea136f 100644 --- a/.github/workflows/auto_releases.yml +++ b/.github/workflows/auto_releases.yml @@ -19,13 +19,11 @@ on: jobs: AutoReleaseInfo: - runs-on: [self-hosted, style-checker-aarch64] + runs-on: [self-hosted, release-maker] outputs: data: ${{ steps.info.outputs.AUTO_RELEASE_PARAMS }} dry_run: ${{ steps.info.outputs.DRY_RUN }} steps: - - name: Debug Info - uses: ./.github/actions/debug - name: Set envs run: | cat >> "$GITHUB_ENV" << 'EOF' @@ -36,6 +34,10 @@ jobs: echo "DRY_RUN=true" >> "$GITHUB_ENV" - name: Check out repository code uses: ClickHouse/checkout@v1 + with: + fetch-depth: 0 # full history needed + - name: Debug Info + uses: ./.github/actions/debug - name: Prepare Info id: info run: | @@ -46,12 +48,7 @@ jobs: echo "::endgroup::" { echo 'AUTO_RELEASE_PARAMS<> "$GITHUB_ENV" - { - echo 'AUTO_RELEASE_PARAMS<> "$GITHUB_OUTPUT" echo "DRY_RUN=true" >> "$GITHUB_OUTPUT" @@ -62,48 +59,29 @@ jobs: - name: Clean up uses: ./.github/actions/clean - Release_0: + Releases: needs: AutoReleaseInfo - name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].release_branch }} - if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].ready }} + strategy: + matrix: + release_params: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases }} + max-parallel: 1 + name: Release ${{ matrix.release_params.release_branch }} uses: ./.github/workflows/create_release.yml with: - ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].commit_sha }} + ref: ${{ matrix.release_params.commit_sha }} type: patch - dry-run: ${{ needs.AutoReleaseInfo.outputs.dry_run }} -# -# Release_1: -# needs: [AutoReleaseInfo, Release_0] -# name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[1].release_branch }} -# if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[1] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[1].ready }} -# uses: ./.github/workflows/create_release.yml -# with: -# ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[1].commit_sha }} -# type: patch -# dry-run: ${{ env.DRY_RUN }} -# -# Release_2: -# needs: [AutoReleaseInfo, Release_1] -# name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[2].release_branch }} -# if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[2].ready }} -# uses: ./.github/workflow/create_release.yml -# with: -# ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].commit_sha }} -# type: patch -# dry-run: ${{ env.DRY_RUN }} -# -# Release_3: -# needs: [AutoReleaseInfo, Release_2] -# name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[3].release_branch }} -# if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[3] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[3].ready }} -# uses: ./.github/workflow/create_release.yml -# with: -# ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[3].commit_sha }} -# type: patch -# dry-run: ${{ env.DRY_RUN }} + dry-run: ${{ fromJson(needs.AutoReleaseInfo.outputs.dry_run) }} + secrets: + ROBOT_CLICKHOUSE_COMMIT_TOKEN: ${{ secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN }} -# - name: Post Slack Message -# if: ${{ !cancelled() }} -# run: | -# cd "$GITHUB_WORKSPACE/tests/ci" -# python3 auto_release.py --post-auto-release-complete --wf-status ${{ job.status }} + PostSlackMessage: + needs: [AutoReleaseInfo] + runs-on: [self-hosted, release-maker] + if: ${{ !cancelled() }} + steps: + - name: Check out repository code + uses: ClickHouse/checkout@v1 + - name: Post + run: | + cd "$GITHUB_WORKSPACE/tests/ci" + python3 auto_release.py --post-auto-release-complete --wf-status ${{ job.status }} diff --git a/.github/workflows/create_release.yml b/.github/workflows/create_release.yml index 1553d689227..1fb6cb60e96 100644 --- a/.github/workflows/create_release.yml +++ b/.github/workflows/create_release.yml @@ -47,6 +47,8 @@ concurrency: required: false default: false type: boolean + secrets: + ROBOT_CLICKHOUSE_COMMIT_TOKEN: jobs: CreateRelease: diff --git a/tests/ci/auto_release.py b/tests/ci/auto_release.py index 3cc88634004..58cfc833afe 100644 --- a/tests/ci/auto_release.py +++ b/tests/ci/auto_release.py @@ -1,4 +1,5 @@ import argparse +import copy import dataclasses import json import os @@ -46,6 +47,7 @@ def parse_args(): MAX_NUMBER_OF_COMMITS_TO_CONSIDER_FOR_RELEASE = 5 AUTORELEASE_INFO_FILE = "/tmp/autorelease_info.json" +AUTORELEASE_MATRIX_PARAMS = "/tmp/autorelease_params.json" @dataclasses.dataclass @@ -74,6 +76,12 @@ class AutoReleaseInfo: with open(AUTORELEASE_INFO_FILE, "w", encoding="utf-8") as f: print(json.dumps(dataclasses.asdict(self), indent=2), file=f) + # dump file for GH action matrix that is similar to the file above but with dropped not ready release branches + params = copy.deepcopy(self) + params.releases = [release for release in params.releases if release.ready] + with open(AUTORELEASE_MATRIX_PARAMS, "w", encoding="utf-8") as f: + print(json.dumps(params, indent=2), file=f) + @staticmethod def from_file() -> "AutoReleaseInfo": with open(AUTORELEASE_INFO_FILE, "r", encoding="utf-8") as json_file: @@ -136,6 +144,7 @@ def _prepare(token): commit_ci_status = CI.GH.get_commit_status_by_name( token=token, commit_sha=commit, + # handle old name for old releases status_name=(CI.JobNames.BUILD_CHECK, "ClickHouse build check"), ) commit_sha = commit diff --git a/tests/ci/ci_utils.py b/tests/ci/ci_utils.py index b8778e0cc50..97ab10f1b58 100644 --- a/tests/ci/ci_utils.py +++ b/tests/ci/ci_utils.py @@ -102,21 +102,29 @@ class GH: assert len(commit_sha) == 40 assert Utils.is_hex(commit_sha) assert not Utils.is_hex(token) - url = f"https://api.github.com/repos/{Envs.GITHUB_REPOSITORY}/commits/{commit_sha}/statuses?per_page={200}" + + url = f"https://api.github.com/repos/{Envs.GITHUB_REPOSITORY}/commits/{commit_sha}/statuses" headers = { "Authorization": f"token {token}", "Accept": "application/vnd.github.v3+json", } - response = requests.get(url, headers=headers, timeout=5) if isinstance(status_name, str): status_name = (status_name,) - if response.status_code == 200: - assert "next" not in response.links, "Response truncated" - statuses = response.json() - for status in statuses: - if status["context"] in status_name: - return status["state"] # type: ignore + + while url: + response = requests.get(url, headers=headers, timeout=5) + if response.status_code == 200: + statuses = response.json() + for status in statuses: + if status["context"] in status_name: + return status["state"] + + # Check if there is a next page + url = response.links.get("next", {}).get("url") + else: + break + return "" @staticmethod From dfd17cc2d71555de9c42ad6085c35bb3f1372dd1 Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 16 Aug 2024 13:23:57 +0000 Subject: [PATCH 067/260] Check for invalid regexp in JSON SKIP REGEXP section --- src/DataTypes/DataTypeObject.cpp | 11 +++++++++++ .../0_stateless/03227_json_invalid_regexp.reference | 0 .../queries/0_stateless/03227_json_invalid_regexp.sql | 4 ++++ 3 files changed, 15 insertions(+) create mode 100644 tests/queries/0_stateless/03227_json_invalid_regexp.reference create mode 100644 tests/queries/0_stateless/03227_json_invalid_regexp.sql diff --git a/src/DataTypes/DataTypeObject.cpp b/src/DataTypes/DataTypeObject.cpp index d6395155397..11fffd8769b 100644 --- a/src/DataTypes/DataTypeObject.cpp +++ b/src/DataTypes/DataTypeObject.cpp @@ -49,6 +49,17 @@ DataTypeObject::DataTypeObject( , max_dynamic_paths(max_dynamic_paths_) , max_dynamic_types(max_dynamic_types_) { + /// Check if regular expressions are valid. + for (const auto & regexp_str : path_regexps_to_skip) + { + re2::RE2::Options options; + /// Don't log errors to stderr. + options.set_log_errors(false); + auto regexp = re2::RE2(regexp_str, options); + if (!regexp.error().empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Invalid regexp '{}': {}", regexp_str, regexp.error()); + } + for (const auto & [typed_path, type] : typed_paths) { for (const auto & path_to_skip : paths_to_skip) diff --git a/tests/queries/0_stateless/03227_json_invalid_regexp.reference b/tests/queries/0_stateless/03227_json_invalid_regexp.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03227_json_invalid_regexp.sql b/tests/queries/0_stateless/03227_json_invalid_regexp.sql new file mode 100644 index 00000000000..734dea1aac6 --- /dev/null +++ b/tests/queries/0_stateless/03227_json_invalid_regexp.sql @@ -0,0 +1,4 @@ +set allow_experimental_json_type = 1; +create table test (json JSON(SKIP REGEXP '[]')) engine=Memory(); -- {serverError BAD_ARGUMENTS} +create table test (json JSON(SKIP REGEXP '+')) engine=Memory(); -- {serverError BAD_ARGUMENTS}; + From 370b6bdc7b6d97f0e697e99ccd06a25e97651406 Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 16 Aug 2024 13:38:30 +0000 Subject: [PATCH 068/260] Update tests --- tests/queries/0_stateless/02497_schema_inference_nulls.sql | 4 ++-- .../0_stateless/02784_schema_inference_null_as_default.sql | 4 ++-- ..._max_bytes_to_read_for_schema_inference_in_cache.reference | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/tests/queries/0_stateless/02497_schema_inference_nulls.sql b/tests/queries/0_stateless/02497_schema_inference_nulls.sql index b78b5709dbb..5670b031e8b 100644 --- a/tests/queries/0_stateless/02497_schema_inference_nulls.sql +++ b/tests/queries/0_stateless/02497_schema_inference_nulls.sql @@ -40,7 +40,7 @@ desc format(JSONCompactEachRow, '[[[], [null], [1, 2, 3]]]'); desc format(JSONCompactEachRow, '[[{"a" : null}, {"b" : 1}]]'); desc format(JSONCompactEachRow, '[[["2020-01-01", null, "1234"], ["abcd"]]]'); -set schema_inference_make_columns_nullable=0; +set schema_inference_make_columns_nullable='auto'; desc format(JSONCompactEachRow, '[[1, 2]]'); desc format(JSONCompactEachRow, '[[null, 1]]'); desc format(JSONCompactEachRow, '[[1, 2]], [[3]]'); @@ -59,7 +59,7 @@ desc format(CSV, '"[[], [null], [1, 2, 3]]"'); desc format(CSV, '"[{\'a\' : null}, {\'b\' : 1}]"'); desc format(CSV, '"[[\'2020-01-01\', null, \'1234\'], [\'abcd\']]"'); -set schema_inference_make_columns_nullable=0; +set schema_inference_make_columns_nullable='auto'; desc format(CSV, '"[1,2]"'); desc format(CSV, '"[NULL, 1]"'); desc format(CSV, '"[1, 2]"\n"[3]"'); diff --git a/tests/queries/0_stateless/02784_schema_inference_null_as_default.sql b/tests/queries/0_stateless/02784_schema_inference_null_as_default.sql index 9c9f99d8283..571e3ab4f25 100644 --- a/tests/queries/0_stateless/02784_schema_inference_null_as_default.sql +++ b/tests/queries/0_stateless/02784_schema_inference_null_as_default.sql @@ -1,7 +1,7 @@ desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=1; select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=1; -desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=0; -select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=0; +desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable='auto', input_format_null_as_default=0; +select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable='auto', input_format_null_as_default=0; desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=1; select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=1; diff --git a/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference b/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference index cd109daac52..3b9d88edc19 100644 --- a/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference +++ b/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference @@ -1,2 +1,2 @@ x Nullable(Int64) -schema_inference_hints=, max_rows_to_read_for_schema_inference=25000, max_bytes_to_read_for_schema_inference=1000, schema_inference_make_columns_nullable=true, try_infer_integers=true, try_infer_dates=true, try_infer_datetimes=true, try_infer_numbers_from_strings=false, read_bools_as_numbers=true, read_bools_as_strings=true, read_objects_as_strings=true, read_numbers_as_strings=true, read_arrays_as_strings=true, try_infer_objects_as_tuples=true, infer_incomplete_types_as_strings=true, try_infer_objects=false, use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects=false +schema_inference_hints=, max_rows_to_read_for_schema_inference=25000, max_bytes_to_read_for_schema_inference=1000, schema_inference_make_columns_nullable=1, try_infer_integers=true, try_infer_dates=true, try_infer_datetimes=true, try_infer_numbers_from_strings=false, read_bools_as_numbers=true, read_bools_as_strings=true, read_objects_as_strings=true, read_numbers_as_strings=true, read_arrays_as_strings=true, try_infer_objects_as_tuples=true, infer_incomplete_types_as_strings=true, try_infer_objects=false, use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects=false From 5fe46af4221a36ecb4566ca7bfad314d732f1de2 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Fri, 16 Aug 2024 18:12:51 +0200 Subject: [PATCH 069/260] Update 02497_schema_inference_nulls.sql --- tests/queries/0_stateless/02497_schema_inference_nulls.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/02497_schema_inference_nulls.sql b/tests/queries/0_stateless/02497_schema_inference_nulls.sql index 5670b031e8b..d62fc76d9b9 100644 --- a/tests/queries/0_stateless/02497_schema_inference_nulls.sql +++ b/tests/queries/0_stateless/02497_schema_inference_nulls.sql @@ -18,7 +18,7 @@ desc format(JSONEachRow, '{"x" : [[], [null], [1, 2, 3]]}'); desc format(JSONEachRow, '{"x" : [{"a" : null}, {"b" : 1}]}'); desc format(JSONEachRow, '{"x" : [["2020-01-01", null, "1234"], ["abcd"]]}'); -set schema_inference_make_columns_nullable=0; +set schema_inference_make_columns_nullable='auto'; desc format(JSONEachRow, '{"x" : [1, 2]}'); desc format(JSONEachRow, '{"x" : [null, 1]}'); desc format(JSONEachRow, '{"x" : [1, 2]}, {"x" : [3]}'); From 4f84c82d6d53ded0adda46aac1db1d345b5ba2eb Mon Sep 17 00:00:00 2001 From: Linh Giang <165205637+linhgiang24@users.noreply.github.com> Date: Fri, 16 Aug 2024 11:02:44 -0600 Subject: [PATCH 070/260] Update grant.md to include POSTGRES privilege Added POSTGRES privilege under the SOURCES category as it seems to be missing. --- docs/en/sql-reference/statements/grant.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/docs/en/sql-reference/statements/grant.md b/docs/en/sql-reference/statements/grant.md index 43fa344a16d..6118f4c1d36 100644 --- a/docs/en/sql-reference/statements/grant.md +++ b/docs/en/sql-reference/statements/grant.md @@ -200,6 +200,7 @@ Hierarchy of privileges: - `JDBC` - `HDFS` - `S3` + - `POSTGRES` - [dictGet](#dictget) - [displaySecretsInShowAndSelect](#displaysecretsinshowandselect) - [NAMED COLLECTION ADMIN](#named-collection-admin) @@ -476,6 +477,7 @@ Allows using external data sources. Applies to [table engines](../../engines/tab - `JDBC`. Level: `GLOBAL` - `HDFS`. Level: `GLOBAL` - `S3`. Level: `GLOBAL` + - `POSTGRES`. Level: `GLOBAL` The `SOURCES` privilege enables use of all the sources. Also you can grant a privilege for each source individually. To use sources, you need additional privileges. From cb8d9a05643d3aac5f410c4eac53124224c63bc8 Mon Sep 17 00:00:00 2001 From: Blargian Date: Sat, 17 Aug 2024 20:13:35 +0200 Subject: [PATCH 071/260] fix typo --- docs/en/sql-reference/functions/type-conversion-functions.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/sql-reference/functions/type-conversion-functions.md b/docs/en/sql-reference/functions/type-conversion-functions.md index cd6fd9ab839..a03394be226 100644 --- a/docs/en/sql-reference/functions/type-conversion-functions.md +++ b/docs/en/sql-reference/functions/type-conversion-functions.md @@ -4952,7 +4952,7 @@ toIntervalMonth(n) **Arguments** -- `n` — Number of m. Positive integer number. [Int*](../data-types/int-uint.md). +- `n` — Number of months. Positive integer number. [Int*](../data-types/int-uint.md). **Returned values** From ae389d14ee65ff5fea3543868b6b161f9fcb806e Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Sat, 17 Aug 2024 20:42:00 +0200 Subject: [PATCH 072/260] Fix stylecheck --- src/Formats/SchemaInferenceUtils.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 54352b88578..e8eab3b4453 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -333,7 +333,7 @@ namespace type = variant_type; type_indexes = {TypeIndex::Variant}; } - + /// If we have only date/datetimes types (Date/DateTime/DateTime64), convert all of them to the common type, /// otherwise, convert all Date, DateTime and DateTime64 to String. void transformDatesAndDateTimes(DataTypes & data_types, TypeIndexesSet & type_indexes) From d6e170f77704833fa6655820d55090ba18b0b9fe Mon Sep 17 00:00:00 2001 From: Chang Chen Date: Sat, 17 Aug 2024 18:31:11 +0800 Subject: [PATCH 073/260] repeat field is also compound types, ignore it. --- src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp b/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp index c6167e572df..7b5c29e321f 100644 --- a/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp @@ -286,6 +286,9 @@ static std::vector getHyperrectangleForRowGroup(const parquet::FileMetaDa if (!s) continue; + if(s->descr()->schema_node()->is_repeated()) + continue; + auto path = c->path_in_schema()->ToDotVector(); if (path.size() != 1) continue; // compound types not supported From 858f8b502002661584e6153d39a23edc87b49dda Mon Sep 17 00:00:00 2001 From: Chang Chen Date: Sun, 18 Aug 2024 19:11:30 +0800 Subject: [PATCH 074/260] add test and update codes per commit --- .../Formats/Impl/ParquetBlockInputFormat.cpp | 2 +- .../02841_parquet_filter_pushdown_bug.reference | 1 + .../02841_parquet_filter_pushdown_bug.sh.sh | 8 ++++++++ .../0_stateless/data_parquet/68131.parquet | Bin 0 -> 289 bytes 4 files changed, 10 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.reference create mode 100755 tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.sh.sh create mode 100644 tests/queries/0_stateless/data_parquet/68131.parquet diff --git a/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp b/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp index 7b5c29e321f..1f213fef731 100644 --- a/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp @@ -286,7 +286,7 @@ static std::vector getHyperrectangleForRowGroup(const parquet::FileMetaDa if (!s) continue; - if(s->descr()->schema_node()->is_repeated()) + if (s->descr()->schema_node()->is_repeated()) continue; auto path = c->path_in_schema()->ToDotVector(); diff --git a/tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.reference b/tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.reference new file mode 100644 index 00000000000..6ed63af507a --- /dev/null +++ b/tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.reference @@ -0,0 +1 @@ +[1,2] diff --git a/tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.sh.sh b/tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.sh.sh new file mode 100755 index 00000000000..58eb207b6e6 --- /dev/null +++ b/tests/queries/0_stateless/02841_parquet_filter_pushdown_bug.sh.sh @@ -0,0 +1,8 @@ +#!/usr/bin/env bash +# Tags: no-fasttest + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_LOCAL -q "select f from file('$CURDIR/data_parquet/68131.parquet', Parquet, 'f Array(Int32)')" \ No newline at end of file diff --git a/tests/queries/0_stateless/data_parquet/68131.parquet b/tests/queries/0_stateless/data_parquet/68131.parquet new file mode 100644 index 0000000000000000000000000000000000000000..169f6152003db164c78e33cd69205caa33f906b5 GIT binary patch literal 289 zcmXAl!D_=W42Bgqgq#M0O4Q&(E)5xMp|QKBg Date: Sun, 18 Aug 2024 15:10:35 +0000 Subject: [PATCH 075/260] Add explicit session_timezone to UTC --- tests/queries/0_stateless/03222_datetime64_small_value_const.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03222_datetime64_small_value_const.sql b/tests/queries/0_stateless/03222_datetime64_small_value_const.sql index af06a622f8d..39266ba7992 100644 --- a/tests/queries/0_stateless/03222_datetime64_small_value_const.sql +++ b/tests/queries/0_stateless/03222_datetime64_small_value_const.sql @@ -1,4 +1,5 @@ -- Tags: shard +set session_timezone = 'UTC'; -- don't randomize the session timezone select *, (select toDateTime64(0, 3)) from remote('127.0.0.1', system.one) settings prefer_localhost_replica=0; select *, (select toDateTime64(5, 3)) from remote('127.0.0.1', system.one) settings prefer_localhost_replica=0; From 8eb922036e5b7caa36c1b904b43fdaee8e45acaa Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Mon, 19 Aug 2024 12:45:15 +0800 Subject: [PATCH 076/260] change as request --- .../functions/string-replace-functions.md | 16 +- src/Functions/overlay.cpp | 140 +++++------ .../0_stateless/03205_overlay.reference | 230 +++++------------- tests/queries/0_stateless/03205_overlay.sql | 132 +++++----- .../0_stateless/03206_overlay_utf8.reference | 168 ------------- .../0_stateless/03206_overlay_utf8.sql | 60 ----- 6 files changed, 203 insertions(+), 543 deletions(-) delete mode 100644 tests/queries/0_stateless/03206_overlay_utf8.reference delete mode 100644 tests/queries/0_stateless/03206_overlay_utf8.sql diff --git a/docs/en/sql-reference/functions/string-replace-functions.md b/docs/en/sql-reference/functions/string-replace-functions.md index d086c9ee64b..408a896e607 100644 --- a/docs/en/sql-reference/functions/string-replace-functions.md +++ b/docs/en/sql-reference/functions/string-replace-functions.md @@ -251,24 +251,24 @@ select printf('%%%s %s %d', 'Hello', 'World', 2024); ## overlay -Replace the string `s` with the string `replace` starting from the 1-based `position` for `length` bytes. If `length` is omitted or negative, then it defaults to the length of `replace`. +Replace the string `s` with the string `replace` starting from the 1-based `offset` for `length` bytes. If `length` is omitted or negative, then it defaults to the length of `replace`. **Syntax** ```sql -overlay(s, replace, position[, length]) +overlay(s, replace, offset[, length]) ``` **Parameters** - `s`: A string type [String](../data-types/string.md). - `replace`: A string type [String](../data-types/string.md). -- `position`: An integer type [Int](../data-types/int-uint.md). +- `offset`: An integer type [Int](../data-types/int-uint.md). - `length`: Optional. An integer type [Int](../data-types/int-uint.md). **Returned value** -- A [String](../data-types/string.md) data type value. If `position` is negative the position is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. +- A [String](../data-types/string.md) data type value. If `offset` is negative the offset is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. **Example** @@ -286,26 +286,26 @@ Result: ## overlayUTF8 -Replace the string `s` with the string `replace` starting from the 1-based `position` for `length` UTF-8 characters. If `length` is omitted or negative, then it defaults to the length of `replace`. +Replace the string `s` with the string `replace` starting from the 1-based `offset` for `length` UTF-8 characters. If `length` is omitted or negative, then it defaults to the length of `replace`. Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. **Syntax** ```sql -overlayUTF8(s, replace, position[, length]) +overlayUTF8(s, replace, offset[, length]) ``` **Parameters** - `s`: A string type [String](../data-types/string.md). - `replace`: A string type [String](../data-types/string.md). -- `position`: An integer type [Int](../data-types/int-uint.md). +- `offset`: An integer type [Int](../data-types/int-uint.md). - `length`: Optional. An integer type [Int](../data-types/int-uint.md). **Returned value** -- A [String](../data-types/string.md) data type value. If `position` is negative the position is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. +- A [String](../data-types/string.md) data type value. If `offset` is negative the offset is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. **Example** diff --git a/src/Functions/overlay.cpp b/src/Functions/overlay.cpp index 094da27a71d..73ca0acbb8e 100644 --- a/src/Functions/overlay.cpp +++ b/src/Functions/overlay.cpp @@ -17,13 +17,13 @@ extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; } -using namespace GatherUtils; - namespace { /// If 'is_utf8' - measure offset and length in code points instead of bytes. -/// Syntax: overlay(input, replace, offset[, length]) +/// Syntax: +/// - overlay(input, replace, offset[, length]) +/// - overlayUTF8(input, replace, offset[, length]) - measure offset and length in code points instead of bytes template class FunctionOverlay : public IFunction { @@ -37,63 +37,39 @@ public: bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } bool useDefaultImplementationForConstants() const override { return true; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - const size_t number_of_arguments = arguments.size(); - if (number_of_arguments < 3 || number_of_arguments > 4) - throw Exception( - ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, - "Number of arguments for function {} doesn't match: " - "passed {}, should be 3 or 4", - getName(), - number_of_arguments); + FunctionArgumentDescriptors mandatory_args{ + {"input", static_cast(&isString), nullptr, "String"}, + {"replace", static_cast(&isString), nullptr, "String"}, + {"offset", static_cast(&isNativeInteger), nullptr, "(U)Int8/16/32/64"}, + }; - /// first argument is string - if (!isString(arguments[0])) - throw Exception( - ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Illegal type {} of first argument of function {}, expected String", - arguments[0]->getName(), - getName()); + FunctionArgumentDescriptors optional_args{ + {"length", static_cast(&isNativeInteger), nullptr, "(U)Int8/16/32/64"}, + }; - /// second argument is string - if (!isString(arguments[1])) - throw Exception( - ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Illegal type {} of second argument of function {}, expected String", - arguments[1]->getName(), - getName()); - - if (!isNativeNumber(arguments[2])) - throw Exception( - ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Illegal type {} of third argument of function {}, expected (U)Int8|16|32|64", - arguments[2]->getName(), - getName()); - - if (number_of_arguments == 4 && !isNativeNumber(arguments[3])) - throw Exception( - ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Illegal type {} of second argument of function {}, expected (U)Int8|16|32|64", - arguments[3]->getName(), - getName()); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(); } ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { + if (input_rows_count == 0) + return ColumnString::create(); + const size_t number_of_arguments = arguments.size(); - bool three_args = number_of_arguments == 3; + bool has_three_args = number_of_arguments == 3; ColumnPtr column_offset = arguments[2].column; ColumnPtr column_length; - if (!three_args) + if (!has_three_args) column_length = arguments[3].column; const ColumnConst * column_offset_const = checkAndGetColumn(column_offset.get()); const ColumnConst * column_length_const = nullptr; - if (!three_args) + if (!has_three_args) column_length_const = checkAndGetColumn(column_length.get()); bool offset_is_const = false; @@ -126,7 +102,7 @@ public: if (column_input_const) { StringRef input = column_input_const->getDataAt(0); - res_data.reserve(input.size * input_rows_count); + res_data.reserve((input.size + 1) * input_rows_count); } else { @@ -135,8 +111,8 @@ public: const auto * column_replace_const = checkAndGetColumn(column_replace.get()); const auto * column_replace_string = checkAndGetColumn(column_replace.get()); - bool input_is_const = column_input_const != nullptr; - bool replace_is_const = column_replace_const != nullptr; + bool input_is_const = (column_input_const != nullptr); + bool replace_is_const = (column_replace_const != nullptr); #define OVERLAY_EXECUTE_CASE(THREE_ARGS, OFFSET_IS_CONST, LENGTH_IS_CONST) \ if (input_is_const && replace_is_const) \ @@ -150,8 +126,9 @@ public: length, \ res_data, \ res_offsets); \ - else if (input_is_const) \ + else if (input_is_const && !replace_is_const) \ constantVector( \ + input_rows_count, \ column_input_const->getDataAt(0), \ column_replace_string->getChars(), \ column_replace_string->getOffsets(), \ @@ -161,8 +138,9 @@ public: length, \ res_data, \ res_offsets); \ - else if (replace_is_const) \ + else if (!input_is_const && replace_is_const) \ vectorConstant( \ + input_rows_count, \ column_input_string->getChars(), \ column_input_string->getOffsets(), \ column_replace_const->getDataAt(0), \ @@ -174,6 +152,7 @@ public: res_offsets); \ else \ vectorVector( \ + input_rows_count, \ column_input_string->getChars(), \ column_input_string->getOffsets(), \ column_replace_string->getChars(), \ @@ -185,7 +164,7 @@ public: res_data, \ res_offsets); - if (three_args) + if (has_three_args) { if (offset_is_const) { @@ -251,7 +230,7 @@ private: return bytes; } - template + template void constantConstant( size_t rows, const StringRef & input, @@ -263,7 +242,7 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!three_args && length_is_const && const_length < 0) + if (!has_three_args && length_is_const && const_length < 0) { constantConstant( rows, input, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); @@ -277,12 +256,12 @@ private: size_t replace_size = getSliceSize(reinterpret_cast(replace.data), replace.size); size_t valid_length = 0; // not negative - if constexpr (!three_args && length_is_const) + if constexpr (!has_three_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; } - else if constexpr (three_args) + else if constexpr (has_three_args) { valid_length = replace_size; } @@ -300,7 +279,7 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (!three_args && !length_is_const) + if constexpr (!has_three_args && !length_is_const) { length = column_length->getInt(i); valid_length = length >= 0 ? length : replace_size; @@ -331,10 +310,10 @@ private: } else { - const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + const auto * prefix_end = GatherUtils::UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); size_t prefix_bytes = prefix_end > input_end ? input.size : prefix_end - input_begin; - const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + const auto * suffix_begin = GatherUtils::UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); size_t suffix_bytes = input_end - suffix_begin; size_t new_res_size = res_data.size() + prefix_bytes + replace.size + suffix_bytes + 1; /// +1 for zero terminator @@ -363,8 +342,9 @@ private: } } - template + template void vectorConstant( + size_t rows, const ColumnString::Chars & input_data, const ColumnString::Offsets & input_offsets, const StringRef & replace, @@ -375,27 +355,26 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!three_args && length_is_const && const_length < 0) + if (!has_three_args && length_is_const && const_length < 0) { vectorConstant( - input_data, input_offsets, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); + rows, input_data, input_offsets, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); return; } size_t replace_size = getSliceSize(reinterpret_cast(replace.data), replace.size); Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative - if constexpr (!three_args && length_is_const) + if constexpr (!has_three_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; } - else if constexpr (three_args) + else if constexpr (has_three_args) { valid_length = replace_size; } - size_t rows = input_offsets.size(); Int64 offset = 0; // start from 1, maybe negative size_t valid_offset = 0; // start from 0, not negative size_t res_offset = 0; @@ -415,7 +394,7 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (!three_args && !length_is_const) + if constexpr (!has_three_args && !length_is_const) { length = column_length->getInt(i); valid_length = length >= 0 ? length : replace_size; @@ -449,9 +428,9 @@ private: { const auto * input_begin = &input_data[input_offset]; const auto * input_end = &input_data[input_offset + input_bytes]; - const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + const auto * prefix_end = GatherUtils::UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); size_t prefix_bytes = prefix_end > input_end ? input_bytes : prefix_end - input_begin; - const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + const auto * suffix_begin = GatherUtils::UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); size_t suffix_bytes = input_end - suffix_begin; size_t new_res_size = res_data.size() + prefix_bytes + replace.size + suffix_bytes + 1; /// +1 for zero terminator @@ -480,8 +459,9 @@ private: } } - template + template void constantVector( + size_t rows, const StringRef & input, const ColumnString::Chars & replace_data, const ColumnString::Offsets & replace_offsets, @@ -492,10 +472,10 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!three_args && length_is_const && const_length < 0) + if (!has_three_args && length_is_const && const_length < 0) { constantVector( - input, replace_data, replace_offsets, column_offset, column_length, const_offset, -1, res_data, res_offsets); + rows, input, replace_data, replace_offsets, column_offset, column_length, const_offset, -1, res_data, res_offsets); return; } @@ -506,13 +486,12 @@ private: Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative - if constexpr (!three_args && length_is_const) + if constexpr (!has_three_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; } - size_t rows = replace_offsets.size(); const auto * input_begin = reinterpret_cast(input.data); const auto * input_end = reinterpret_cast(input.data + input.size); Int64 offset = 0; // start from 1, maybe negative @@ -529,7 +508,7 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (three_args) + if constexpr (has_three_args) { valid_length = replace_size; } @@ -564,9 +543,9 @@ private: } else { - const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + const auto * prefix_end = GatherUtils::UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); size_t prefix_bytes = prefix_end > input_end ? input.size : prefix_end - input_begin; - const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + const auto * suffix_begin = GatherUtils::UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); size_t suffix_bytes = input_end - suffix_begin; size_t new_res_size = res_data.size() + prefix_bytes + replace_bytes + suffix_bytes + 1; /// +1 for zero terminator res_data.resize(new_res_size); @@ -594,8 +573,9 @@ private: } } - template + template void vectorVector( + size_t rows, const ColumnString::Chars & input_data, const ColumnString::Offsets & input_offsets, const ColumnString::Chars & replace_data, @@ -607,9 +587,10 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!three_args && length_is_const && const_length < 0) + if (!has_three_args && length_is_const && const_length < 0) { vectorVector( + rows, input_data, input_offsets, replace_data, @@ -625,13 +606,12 @@ private: Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative - if constexpr (!three_args && length_is_const) + if constexpr (!has_three_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; } - size_t rows = input_offsets.size(); Int64 offset = 0; // start from 1, maybe negative size_t valid_offset = 0; // start from 0, not negative size_t res_offset = 0; @@ -655,7 +635,7 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (three_args) + if constexpr (has_three_args) { valid_length = replace_size; } @@ -693,9 +673,9 @@ private: { const auto * input_begin = &input_data[input_offset]; const auto * input_end = &input_data[input_offset + input_bytes]; - const auto * prefix_end = UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); + const auto * prefix_end = GatherUtils::UTF8StringSource::skipCodePointsForward(input_begin, prefix_size, input_end); size_t prefix_bytes = prefix_end > input_end ? input_bytes : prefix_end - input_begin; - const auto * suffix_begin = UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); + const auto * suffix_begin = GatherUtils::UTF8StringSource::skipCodePointsBackward(input_end, suffix_size, input_begin); size_t suffix_bytes = input_end - suffix_begin; size_t new_res_size = res_data.size() + prefix_bytes + replace_bytes + suffix_bytes + 1; /// +1 for zero terminator res_data.resize(new_res_size); diff --git a/tests/queries/0_stateless/03205_overlay.reference b/tests/queries/0_stateless/03205_overlay.reference index 9e79db2e131..383a26986d6 100644 --- a/tests/queries/0_stateless/03205_overlay.reference +++ b/tests/queries/0_stateless/03205_overlay.reference @@ -1,168 +1,62 @@ -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark_SQL -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark CORE -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Spark ANSI SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL -Structured SQL +Negative test of overlay +Positive test 1 with various combinations of const/non-const columns +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Positive test 2 with various combinations of const/non-const columns +Spark_SQL Spark_SQL和CH +Spark_SQL Spark_SQL和CH +Spark_SQL Spark_SQL和CH +Spark_SQL Spark_SQL和CH +Spark_SQL Spark_SQL和CH +Spark_SQL Spark_SQL和CH +Spark_SQL Spark_SQL和CH +Spark_SQL Spark_SQL和CH +Positive test 3 with various combinations of const/non-const columns +Spark CORE Spark CORECH +Spark CORE Spark CORECH +Spark CORE Spark CORECH +Spark CORE Spark CORECH +Spark CORE Spark CORECH +Spark CORE Spark CORECH +Spark CORE Spark CORECH +Spark CORE Spark CORECH +Positive test 4 with various combinations of const/non-const columns +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Spark ANSI SQL Spark ANSI SQL和CH +Positive test 5 with various combinations of const/non-const columns +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH +Structured SQL Structured SQL和CH diff --git a/tests/queries/0_stateless/03205_overlay.sql b/tests/queries/0_stateless/03205_overlay.sql index b131312c934..4fd0791521d 100644 --- a/tests/queries/0_stateless/03205_overlay.sql +++ b/tests/queries/0_stateless/03205_overlay.sql @@ -1,60 +1,74 @@ -SELECT overlay('Spark SQL', 'ANSI ', 7, 0) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, 0) from numbers(3); -SELECT overlay('Spark SQL', materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlay('Spark SQL', 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlay('Spark SQL', 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), 0) from numbers(3); -SELECT overlay('Spark SQL', materialize('ANSI '), 7, materialize(0)) from numbers(3); -SELECT overlay('Spark SQL', 'ANSI ', materialize(7), materialize(0)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), 0) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, materialize(0)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), materialize(0)) from numbers(3); -SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); +SELECT 'Negative test of overlay'; +SELECT overlay('hello', 2); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT overlay('hello', 'world'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT overlay('hello', 'world', 2, 3, 'extra'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT overlay(123, 'world', 2, 3); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT overlay('hello', 456, 2, 3); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT overlay('hello', 'world', 'two', 3); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT overlay('hello', 'world', 2, 'three'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } -SELECT overlay('Spark SQL', '_', 6) from numbers(3); -SELECT overlay(materialize('Spark SQL'), '_', 6) from numbers(3); -SELECT overlay('Spark SQL', materialize('_'), 6) from numbers(3); -SELECT overlay('Spark SQL', '_', materialize(6)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('_'), 6) from numbers(3); -SELECT overlay(materialize('Spark SQL'), '_', materialize(6)) from numbers(3); -SELECT overlay('Spark SQL', materialize('_'), materialize(6)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('_'), materialize(6)) from numbers(3); - -SELECT overlay('Spark SQL', 'CORE', 7) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'CORE', 7) from numbers(3); -SELECT overlay('Spark SQL', materialize('CORE'), 7) from numbers(3); -SELECT overlay('Spark SQL', 'CORE', materialize(7)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('CORE'), 7) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'CORE', materialize(7)) from numbers(3); -SELECT overlay('Spark SQL', materialize('CORE'), materialize(7)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('CORE'), materialize(7)) from numbers(3); - -SELECT overlay('Spark SQL', 'ANSI ', 7, 0) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, 0) from numbers(3); -SELECT overlay('Spark SQL', materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlay('Spark SQL', 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlay('Spark SQL', 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), 0) from numbers(3); -SELECT overlay('Spark SQL', materialize('ANSI '), 7, materialize(0)) from numbers(3); -SELECT overlay('Spark SQL', 'ANSI ', materialize(7), materialize(0)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); - -SELECT overlay('Spark SQL', 'tructured', 2, 4) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'tructured', 2, 4) from numbers(3); -SELECT overlay('Spark SQL', materialize('tructured'), 2, 4) from numbers(3); -SELECT overlay('Spark SQL', 'tructured', materialize(2), 4) from numbers(3); -SELECT overlay('Spark SQL', 'tructured', 2, materialize(4)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('tructured'), 2, 4) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'tructured', materialize(2), 4) from numbers(3); -SELECT overlay(materialize('Spark SQL'), 'tructured', 2, materialize(4)) from numbers(3); -SELECT overlay('Spark SQL', materialize('tructured'), materialize(2), 4) from numbers(3); -SELECT overlay('Spark SQL', materialize('tructured'), 2, materialize(4)) from numbers(3); -SELECT overlay('Spark SQL', 'tructured', materialize(2), materialize(4)) from numbers(3); -SELECT overlay(materialize('Spark SQL'), materialize('tructured'), materialize(2), materialize(4)) from numbers(3); +SELECT 'Positive test 1 with various combinations of const/non-const columns'; +SELECT overlay('Spark SQL', 'ANSI ', 7, 0), overlayUTF8('Spark SQL和CH', 'ANSI ', 7, 0); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, 0), overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, 0); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, 0), overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, 0); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), 0), overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), 0); +SELECT overlay('Spark SQL', 'ANSI ', 7, materialize(0)), overlayUTF8('Spark SQL和CH', 'ANSI ', 7, materialize(0)); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, 0), overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, 0); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), 0), overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), 0); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, materialize(0)), overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, materialize(0)); +SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), 0), overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), 0); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, materialize(0)), overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, materialize(0)); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), materialize(0)), overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), materialize(0)); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), 0), overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), 0); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, materialize(0)), overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, materialize(0)); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), materialize(0)), overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), materialize(0)); +SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), materialize(0)), overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), materialize(0)); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), materialize(0)), overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), materialize(0)); + +SELECT 'Positive test 2 with various combinations of const/non-const columns'; +SELECT overlay('Spark SQL', '_', 6), overlayUTF8('Spark SQL和CH', '_', 6); +SELECT overlay(materialize('Spark SQL'), '_', 6), overlayUTF8(materialize('Spark SQL和CH'), '_', 6); +SELECT overlay('Spark SQL', materialize('_'), 6), overlayUTF8('Spark SQL和CH', materialize('_'), 6); +SELECT overlay('Spark SQL', '_', materialize(6)), overlayUTF8('Spark SQL和CH', '_', materialize(6)); +SELECT overlay(materialize('Spark SQL'), materialize('_'), 6), overlayUTF8(materialize('Spark SQL和CH'), materialize('_'), 6); +SELECT overlay(materialize('Spark SQL'), '_', materialize(6)), overlayUTF8(materialize('Spark SQL和CH'), '_', materialize(6)); +SELECT overlay('Spark SQL', materialize('_'), materialize(6)), overlayUTF8('Spark SQL和CH', materialize('_'), materialize(6)); +SELECT overlay(materialize('Spark SQL'), materialize('_'), materialize(6)), overlayUTF8(materialize('Spark SQL和CH'), materialize('_'), materialize(6)); + +SELECT 'Positive test 3 with various combinations of const/non-const columns'; +SELECT overlay('Spark SQL', 'CORE', 7), overlayUTF8('Spark SQL和CH', 'CORE', 7); +SELECT overlay(materialize('Spark SQL'), 'CORE', 7), overlayUTF8(materialize('Spark SQL和CH'), 'CORE', 7); +SELECT overlay('Spark SQL', materialize('CORE'), 7), overlayUTF8('Spark SQL和CH', materialize('CORE'), 7); +SELECT overlay('Spark SQL', 'CORE', materialize(7)), overlayUTF8('Spark SQL和CH', 'CORE', materialize(7)); +SELECT overlay(materialize('Spark SQL'), materialize('CORE'), 7), overlayUTF8(materialize('Spark SQL和CH'), materialize('CORE'), 7); +SELECT overlay(materialize('Spark SQL'), 'CORE', materialize(7)), overlayUTF8(materialize('Spark SQL和CH'), 'CORE', materialize(7)); +SELECT overlay('Spark SQL', materialize('CORE'), materialize(7)), overlayUTF8('Spark SQL和CH', materialize('CORE'), materialize(7)); +SELECT overlay(materialize('Spark SQL'), materialize('CORE'), materialize(7)), overlayUTF8(materialize('Spark SQL和CH'), materialize('CORE'), materialize(7)); + +SELECT 'Positive test 4 with various combinations of const/non-const columns'; +SELECT overlay('Spark SQL', 'ANSI ', 7, 0), overlayUTF8('Spark SQL和CH', 'ANSI ', 7, 0); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, 0), overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, 0); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, 0), overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, 0); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), 0), overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), 0); +SELECT overlay('Spark SQL', 'ANSI ', 7, materialize(0)), overlayUTF8('Spark SQL和CH', 'ANSI ', 7, materialize(0)); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), 7, 0), overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, 0); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), 0), overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), 0); +SELECT overlay(materialize('Spark SQL'), 'ANSI ', 7, materialize(0)), overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, materialize(0)); +SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), 0), overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), 0); +SELECT overlay('Spark SQL', materialize('ANSI '), 7, materialize(0)), overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, materialize(0)); +SELECT overlay('Spark SQL', 'ANSI ', materialize(7), materialize(0)), overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), materialize(0)); +SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), materialize(0)), overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), materialize(0)); + +SELECT 'Positive test 5 with various combinations of const/non-const columns'; +SELECT overlay('Spark SQL', 'tructured', 2, 4), overlayUTF8('Spark SQL和CH', 'tructured', 2, 4); +SELECT overlay(materialize('Spark SQL'), 'tructured', 2, 4), overlayUTF8(materialize('Spark SQL和CH'), 'tructured', 2, 4); +SELECT overlay('Spark SQL', materialize('tructured'), 2, 4), overlayUTF8('Spark SQL和CH', materialize('tructured'), 2, 4); +SELECT overlay('Spark SQL', 'tructured', materialize(2), 4), overlayUTF8('Spark SQL和CH', 'tructured', materialize(2), 4); +SELECT overlay('Spark SQL', 'tructured', 2, materialize(4)), overlayUTF8('Spark SQL和CH', 'tructured', 2, materialize(4)); +SELECT overlay(materialize('Spark SQL'), materialize('tructured'), 2, 4), overlayUTF8(materialize('Spark SQL和CH'), materialize('tructured'), 2, 4); +SELECT overlay(materialize('Spark SQL'), 'tructured', materialize(2), 4), overlayUTF8(materialize('Spark SQL和CH'), 'tructured', materialize(2), 4); +SELECT overlay(materialize('Spark SQL'), 'tructured', 2, materialize(4)), overlayUTF8(materialize('Spark SQL和CH'), 'tructured', 2, materialize(4)); +SELECT overlay('Spark SQL', materialize('tructured'), materialize(2), 4), overlayUTF8('Spark SQL和CH', materialize('tructured'), materialize(2), 4); +SELECT overlay('Spark SQL', materialize('tructured'), 2, materialize(4)), overlayUTF8('Spark SQL和CH', materialize('tructured'), 2, materialize(4)); +SELECT overlay('Spark SQL', 'tructured', materialize(2), materialize(4)), overlayUTF8('Spark SQL和CH', 'tructured', materialize(2), materialize(4)); +SELECT overlay(materialize('Spark SQL'), materialize('tructured'), materialize(2), materialize(4)), overlayUTF8(materialize('Spark SQL和CH'), materialize('tructured'), materialize(2), materialize(4)); diff --git a/tests/queries/0_stateless/03206_overlay_utf8.reference b/tests/queries/0_stateless/03206_overlay_utf8.reference deleted file mode 100644 index 19878c97184..00000000000 --- a/tests/queries/0_stateless/03206_overlay_utf8.reference +++ /dev/null @@ -1,168 +0,0 @@ -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark_SQL和CH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark CORECH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Spark ANSI SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH -Structured SQL和CH diff --git a/tests/queries/0_stateless/03206_overlay_utf8.sql b/tests/queries/0_stateless/03206_overlay_utf8.sql deleted file mode 100644 index 00b756c8b5b..00000000000 --- a/tests/queries/0_stateless/03206_overlay_utf8.sql +++ /dev/null @@ -1,60 +0,0 @@ -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, 0) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, materialize(0)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), materialize(0)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), 0) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, materialize(0)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), materialize(0)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); - -SELECT overlayUTF8('Spark SQL和CH', '_', 6) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), '_', 6) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('_'), 6) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', '_', materialize(6)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('_'), 6) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), '_', materialize(6)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('_'), materialize(6)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('_'), materialize(6)) from numbers(3); - -SELECT overlayUTF8('Spark SQL和CH', 'CORE', 7) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'CORE', 7) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('CORE'), 7) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'CORE', materialize(7)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('CORE'), 7) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'CORE', materialize(7)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('CORE'), materialize(7)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('CORE'), materialize(7)) from numbers(3); - -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, 0) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), 7, 0) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', materialize(7), 0) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'ANSI ', 7, materialize(0)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), 0) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('ANSI '), 7, materialize(0)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'ANSI ', materialize(7), materialize(0)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), materialize(0)) from numbers(3); - -SELECT overlayUTF8('Spark SQL和CH', 'tructured', 2, 4) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'tructured', 2, 4) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('tructured'), 2, 4) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'tructured', materialize(2), 4) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'tructured', 2, materialize(4)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('tructured'), 2, 4) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'tructured', materialize(2), 4) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), 'tructured', 2, materialize(4)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('tructured'), materialize(2), 4) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', materialize('tructured'), 2, materialize(4)) from numbers(3); -SELECT overlayUTF8('Spark SQL和CH', 'tructured', materialize(2), materialize(4)) from numbers(3); -SELECT overlayUTF8(materialize('Spark SQL和CH'), materialize('tructured'), materialize(2), materialize(4)) from numbers(3); From e623ad041f4937b0e7ed22f3159acfee6c0147b3 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Thu, 25 Jul 2024 16:44:17 +0200 Subject: [PATCH 077/260] Make C-z ignorance configurable (ignore_shell_suspend) in clickhouse-client C-z is extermelly useful for some users (like myself), so provide a way to configure it in client and avoid it's ignorance in clickhouse-disks (I hope it is OK since it is not that known utility and it does not have it's own configuration, while cli option is useless, one should remeber about it). Honestly I've never seen any interactive client that forbids C-z, so ignoring it my default looks strange to me. Signed-off-by: Azat Khuzhin --- programs/client/clickhouse-client.xml | 3 +++ programs/disks/DisksApp.cpp | 1 + programs/disks/DisksApp.h | 2 +- programs/disks/DisksClient.cpp | 1 - programs/disks/DisksClient.h | 3 +-- programs/keeper-client/KeeperClient.cpp | 1 + src/Client/ClientBase.cpp | 1 + src/Client/ReplxxLineReader.cpp | 4 +++- src/Client/ReplxxLineReader.h | 1 + 9 files changed, 12 insertions(+), 5 deletions(-) diff --git a/programs/client/clickhouse-client.xml b/programs/client/clickhouse-client.xml index 9ce7d1cb223..6eb8976a6ef 100644 --- a/programs/client/clickhouse-client.xml +++ b/programs/client/clickhouse-client.xml @@ -53,6 +53,9 @@ --> + + + If --- src/Functions/LowerUpperUTF8Impl.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Functions/LowerUpperUTF8Impl.h b/src/Functions/LowerUpperUTF8Impl.h index 8469bedde0c..36ee1723269 100644 --- a/src/Functions/LowerUpperUTF8Impl.h +++ b/src/Functions/LowerUpperUTF8Impl.h @@ -57,7 +57,7 @@ struct LowerUpperUTF8Impl input.toUTF8String(output); /// For valid UTF-8 input strings, ICU sometimes produces output with an extra '\0 at the end. Only the data before that - /// '\0' is valid. It the input is not valid UTF-8, then the behavior of lower/upperUTF8 is undefined by definition. In this + /// '\0' is valid. If the input is not valid UTF-8, then the behavior of lower/upperUTF8 is undefined by definition. In this /// case, the behavior is also reasonable. size_t valid_size = output.size(); if (!output.empty() && output.back() == '\0') From 2f6ad1271cfbd9aa62ad2365e70314aba4da21b9 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 00:27:02 +0200 Subject: [PATCH 181/260] fix tests + exception --- src/Storages/VirtualColumnUtils.cpp | 2 +- .../test_storage_azure_blob_storage/test.py | 10 +-- tests/integration/test_storage_hdfs/test.py | 9 +-- .../03203_hive_style_partitioning.reference | 2 - .../03203_hive_style_partitioning.sh | 61 +++---------------- 5 files changed, 15 insertions(+), 69 deletions(-) diff --git a/src/Storages/VirtualColumnUtils.cpp b/src/Storages/VirtualColumnUtils.cpp index ca82a1ce67a..f0d276e4e56 100644 --- a/src/Storages/VirtualColumnUtils.cpp +++ b/src/Storages/VirtualColumnUtils.cpp @@ -162,7 +162,7 @@ VirtualColumnsDescription getVirtualsForFileLikeStorage(ColumnsDescription & sto return; if (storage_columns.size() == 1) - throw Exception(ErrorCodes::INCORRECT_DATA, "Cannot implement partition by all columns in a file"); + throw Exception(ErrorCodes::INCORRECT_DATA, "Cannot use hive partitioning for file {}: it contains only partition columns. Disable use_hive_partitioning setting to read this file", path); auto local_type = storage_columns.get(name).type; storage_columns.remove(name); desc.addEphemeral(name, local_type, ""); diff --git a/tests/integration/test_storage_azure_blob_storage/test.py b/tests/integration/test_storage_azure_blob_storage/test.py index 637dbd38262..a3172329a99 100644 --- a/tests/integration/test_storage_azure_blob_storage/test.py +++ b/tests/integration/test_storage_azure_blob_storage/test.py @@ -1518,14 +1518,14 @@ def test_hive_partitioning_with_one_parameter(cluster): ) query = ( - f"SELECT column1, column2, _file, _path FROM azureBlobStorage(azure_conf2, " + f"SELECT column2, _file, _path, column1 FROM azureBlobStorage(azure_conf2, " f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " f"blob_path='{path}', format='CSV', structure='{table_format}')" ) assert azure_query( node, query, settings={"use_hive_partitioning": 1} ).splitlines() == [ - "Elizabeth\tGordon\tsample.csv\t{bucket}/{max_path}".format( + "Gordon\tsample.csv\t{bucket}/{max_path}\tElizabeth".format( bucket="cont", max_path=path ) ] @@ -1560,7 +1560,7 @@ def test_hive_partitioning_with_all_parameters(cluster): f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " f"blob_path='{path}', format='CSV', structure='{table_format}');" ) - pattern = r"DB::Exception: Cannot implement partition by all columns in a file" + pattern = r"DB::Exception: Cannot use hive partitioning for file" with pytest.raises(Exception, match=pattern): azure_query(node, query, settings={"use_hive_partitioning": 1}) @@ -1572,7 +1572,7 @@ def test_hive_partitioning_without_setting(cluster): table_format = "column1 String, column2 String" values_1 = f"('Elizabeth', 'Gordon')" values_2 = f"('Emilia', 'Gregor')" - path = "a/column1=Elizabeth/column2=Gordon/sample.csv" + path = "a/column1=Elizabeth/column2=Gordon/column3=Gordon/sample.csv" azure_query( node, @@ -1582,7 +1582,7 @@ def test_hive_partitioning_without_setting(cluster): ) query = ( - f"SELECT column1, column2, _file, _path FROM azureBlobStorage(azure_conf2, " + f"SELECT column1, column2, _file, _path, column3 FROM azureBlobStorage(azure_conf2, " f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " f"blob_path='{path}', format='CSV', structure='{table_format}');" ) diff --git a/tests/integration/test_storage_hdfs/test.py b/tests/integration/test_storage_hdfs/test.py index ad2e7084791..ea8c4efa745 100644 --- a/tests/integration/test_storage_hdfs/test.py +++ b/tests/integration/test_storage_hdfs/test.py @@ -1268,11 +1268,6 @@ def test_hive_partitioning_with_one_parameter(started_cluster): ) assert r == f"Elizabeth\n" - r = node1.query( - "SELECT column1 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/parquet_1', 'TSV')", - settings={"use_hive_partitioning": 1}, - ) - assert r == f"Gordon\n" def test_hive_partitioning_with_all_parameters(started_cluster): @@ -1285,11 +1280,11 @@ def test_hive_partitioning_with_all_parameters(started_cluster): == f"Elizabeth\tGordon\n" ) - pattern = r"DB::Exception: Cannot implement partition by all columns in a file" + pattern = r"DB::Exception: Cannot use hive partitioning for file" with pytest.raises(QueryRuntimeException, match=pattern): node1.query( - f"SELECT _column1 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/column1=Gordon/parquet_2', 'TSV');", + f"SELECT column1 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/column1=Gordon/parquet_2', 'TSV');", settings={"use_hive_partitioning": 1}, ) diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.reference b/tests/queries/0_stateless/03203_hive_style_partitioning.reference index b5eaef7f51e..af52dcd9b88 100644 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.reference +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.reference @@ -35,8 +35,6 @@ Cross Elizabeth Array(Int64) LowCardinality(Float64) 101 2070 -4081 -2070 2070 b 1 diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.sh b/tests/queries/0_stateless/03203_hive_style_partitioning.sh index 41b215578f0..4e165446c34 100755 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.sh +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.sh @@ -11,22 +11,10 @@ $CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE FILE HIVE PARTITIONING'" $CLICKHOUSE_LOCAL -n -q """ set use_hive_partitioning = 1; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; - -SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; - -SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; - -SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; - -SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; SELECT *, non_existing_column FROM file('$CURDIR/data_hive/partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=*/sample.parquet') WHERE column0 = _column0; +SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=*/sample.parquet') WHERE column0 = 'Elizabeth'; SELECT number, date FROM file('$CURDIR/data_hive/partitioning/number=42/date=2020-01-01/sample.parquet') LIMIT 1; SELECT array, float FROM file('$CURDIR/data_hive/partitioning/array=[1,2,3]/float=42.42/sample.parquet') LIMIT 1; @@ -37,7 +25,6 @@ SELECT count(*) FROM file('$CURDIR/data_hive/partitioning/number=42/date=2020-01 $CLICKHOUSE_LOCAL -n -q """ set use_hive_partitioning = 1; -SELECT _identifier FROM file('$CURDIR/data_hive/partitioning/identifier=*/email.csv') LIMIT 2; SELECT identifier FROM file('$CURDIR/data_hive/partitioning/identifier=*/email.csv') LIMIT 2; SELECT a FROM file('$CURDIR/data_hive/partitioning/a=b/a=b/sample.parquet') LIMIT 1; """ @@ -61,21 +48,7 @@ $CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE URL PARTITIONING'" $CLICKHOUSE_LOCAL -n -q """ set use_hive_partitioning = 1; -SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; - -SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = _column0; - -SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; -SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; - -SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; -SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; - -SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; -SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; - -SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; -SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; SELECT *, non_existing_column FROM url('http://localhost:11111/test/hive_partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10;""" @@ -92,24 +65,10 @@ $CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE S3 PARTITIONING'" $CLICKHOUSE_CLIENT -n -q """ set use_hive_partitioning = 1; -SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; - -SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = _column0; - -SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; -SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; - -SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; -SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; - -SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; -SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; - -SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; -SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; SELECT *, non_existing_column FROM s3('http://localhost:11111/test/hive_partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10; -SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=*/sample.parquet') WHERE column0 = _column0; +SELECT *, column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=*/sample.parquet') WHERE column0 = 'Elizabeth'; """ $CLICKHOUSE_CLIENT -n -q """ @@ -123,13 +82,7 @@ $CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE S3CLUSTER PARTITIONING'" $CLICKHOUSE_CLIENT -n -q """ set use_hive_partitioning = 1; -SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; +SELECT *, column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; -SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = _column0; - -SELECT *, _column0, _column1 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; -SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; - -SELECT *, _column0, _column1 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; -SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = 'Elizabeth'; """ From 8a89cd31a1e7770479af6eaf1b4211ef4ece1795 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Thu, 22 Aug 2024 00:29:32 +0200 Subject: [PATCH 182/260] Fix Upgrade Check: move some settings to 24.9 section --- src/Core/SettingsChangesHistory.cpp | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index fb59577b0f0..5e831c6301c 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -72,11 +72,13 @@ static std::initializer_list Date: Thu, 22 Aug 2024 00:48:29 +0200 Subject: [PATCH 183/260] fix black --- tests/integration/test_storage_hdfs/test.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/integration/test_storage_hdfs/test.py b/tests/integration/test_storage_hdfs/test.py index ea8c4efa745..a75c13b9ea6 100644 --- a/tests/integration/test_storage_hdfs/test.py +++ b/tests/integration/test_storage_hdfs/test.py @@ -1269,7 +1269,6 @@ def test_hive_partitioning_with_one_parameter(started_cluster): assert r == f"Elizabeth\n" - def test_hive_partitioning_with_all_parameters(started_cluster): hdfs_api = started_cluster.hdfs_api hdfs_api.write_data( From 0f3c7ae8c202f475fe55f33f45e9bca92155d52c Mon Sep 17 00:00:00 2001 From: Konstantin Bogdanov Date: Thu, 22 Aug 2024 01:15:16 +0200 Subject: [PATCH 184/260] feat: add docs --- utils/check-style/aspell-ignore/en/aspell-dict.txt | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index ffd9fae7f45..308e285c4bd 100644 --- a/utils/check-style/aspell-ignore/en/aspell-dict.txt +++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt @@ -1,4 +1,4 @@ -personal_ws-1.1 en 2942 +personal_ws-1.1 en 2983 AArch ACLs ALTERs @@ -957,6 +957,7 @@ ThreadPoolRemoteFSReaderThreads ThreadPoolRemoteFSReaderThreadsActive ThreadsActive ThreadsInOvercommitTracker +TimeSeries Timeunit TinyLog Tkachenko @@ -1098,12 +1099,12 @@ addressToLineWithInlines addressToSymbol adviced agg +aggThrow aggregatefunction aggregatingmergetree aggregatio aggretate aggthrow -aggThrow aiochclient allocator alphaTokens @@ -1875,8 +1876,8 @@ joinGet joinGetOrNull json jsonMergePatch -jsonasstring jsonasobject +jsonasstring jsoncolumns jsoncolumnsmonoblock jsoncompact @@ -1917,8 +1918,8 @@ kurtSamp kurtosis kurtpop kurtsamp -laion lagInFrame +laion lang laravel largestTriangleThreeBuckets @@ -2020,7 +2021,6 @@ maxMap maxintersections maxintersectionsposition maxmap -minMappedArrays maxmind mdadm meanZTest @@ -2213,8 +2213,8 @@ parseReadableSizeOrZero parseTimeDelta parseable parsers -partitionId partitionID +partitionId pathFull pclmulqdq pcre @@ -2443,6 +2443,7 @@ rewritable rightPad rightPadUTF rightUTF +ripeMD risc riscv ro @@ -2694,7 +2695,6 @@ themself threadpool throwIf timeDiff -TimeSeries timeSeriesData timeSeriesMetrics timeSeriesTags From 54caf1f84e3c3b5076adf29b49f4ee548f243091 Mon Sep 17 00:00:00 2001 From: Konstantin Bogdanov Date: Thu, 22 Aug 2024 01:20:46 +0200 Subject: [PATCH 185/260] fix: wrap in conditional preprocessor directives --- src/Functions/FunctionsHashing.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/Functions/FunctionsHashing.h b/src/Functions/FunctionsHashing.h index 5111ee2bd90..ec39cf1e2cf 100644 --- a/src/Functions/FunctionsHashing.h +++ b/src/Functions/FunctionsHashing.h @@ -192,6 +192,7 @@ T combineHashesFunc(T t1, T t2) return HashFunction::apply(reinterpret_cast(hashes), sizeof(hashes)); } +#if USE_SSL struct RipeMD160Impl { static constexpr auto name = "ripeMD160"; @@ -218,7 +219,7 @@ struct RipeMD160Impl static constexpr bool use_int_hash_for_pods = false; }; - +#endif struct SipHash64Impl { @@ -1647,6 +1648,7 @@ using FunctionIntHash32 = FunctionIntHash; using FunctionIntHash64 = FunctionIntHash; #if USE_SSL using FunctionHalfMD5 = FunctionAnyHash; +using FunctionRipeMD160Hash = FunctionAnyHash; #endif using FunctionSipHash128 = FunctionAnyHash; using FunctionSipHash128Keyed = FunctionAnyHash; @@ -1676,7 +1678,6 @@ using FunctionXXH3 = FunctionAnyHash; using FunctionWyHash64 = FunctionAnyHash; -using FunctionRipeMD160Hash = FunctionAnyHash; } #pragma clang diagnostic pop From 92e153585ded4f15e1292613584ff35a55c735f3 Mon Sep 17 00:00:00 2001 From: Tanya Bragin Date: Wed, 21 Aug 2024 19:19:07 -0700 Subject: [PATCH 186/260] Update README.md Add latest meetups from Alexey tour. --- README.md | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/README.md b/README.md index 17b6dcd2ac1..5e66b9da73e 100644 --- a/README.md +++ b/README.md @@ -40,7 +40,13 @@ Every month we get together with the community (users, contributors, customers, Keep an eye out for upcoming meetups and events around the world. Somewhere else you want us to be? Please feel free to reach out to tyler `` clickhouse `` com. You can also peruse [ClickHouse Events](https://clickhouse.com/company/news-events) for a list of all upcoming trainings, meetups, speaking engagements, etc. +The following upcoming meetups are featuring creator of ClickHouse & CTO, Alexey Milovidov: + * [ClickHouse Guangzhou User Group Meetup](https://mp.weixin.qq.com/s/GSvo-7xUoVzCsuUvlLTpCw) - August 25 +* [San Francisco Meetup (Cloudflare)](https://www.meetup.com/clickhouse-silicon-valley-meetup-group/events/302540575) - September 5 +* [Raleigh Meetup (Deutsche Bank)](https://www.meetup.com/clickhouse-nc-meetup-group/events/302557230) - September 9 +* [New York Meetup (Ramp)](https://www.meetup.com/clickhouse-new-york-user-group/events/302575342) - September 10 +* [Chicago Meetup (Jump Capital)](https://lu.ma/43tvmrfw) - September 12 ## Recent Recordings * **Recent Meetup Videos**: [Meetup Playlist](https://www.youtube.com/playlist?list=PL0Z2YDlm0b3iNDUzpY1S3L_iV4nARda_U) Whenever possible recordings of the ClickHouse Community Meetups are edited and presented as individual talks. Current featuring "Modern SQL in 2023", "Fast, Concurrent, and Consistent Asynchronous INSERTS in ClickHouse", and "Full-Text Indices: Design and Experiments" From 9d0b3e3937cca32bc8bc922876fb8e6ac53a3de9 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 22 Aug 2024 11:32:59 +0800 Subject: [PATCH 187/260] change as request --- .../functions/string-replace-functions.md | 145 +++++++++--------- src/Functions/overlay.cpp | 8 +- ...new_functions_must_be_documented.reference | 2 - 3 files changed, 76 insertions(+), 79 deletions(-) diff --git a/docs/en/sql-reference/functions/string-replace-functions.md b/docs/en/sql-reference/functions/string-replace-functions.md index 55e97688b18..0cc6b0b27d5 100644 --- a/docs/en/sql-reference/functions/string-replace-functions.md +++ b/docs/en/sql-reference/functions/string-replace-functions.md @@ -8,6 +8,78 @@ sidebar_label: Replacing in Strings [General strings functions](string-functions.md) and [functions for searching in strings](string-search-functions.md) are described separately. +## overlay + +Replace part of the string `input` with another string `replace`, starting at the 1-based index `offset`. + +**Syntax** + +```sql +overlay(s, replace, offset[, length]) +``` + +**Parameters** + +- `input`: A string type [String](../data-types/string.md). +- `replace`: A string type [String](../data-types/string.md). +- `offset`: An integer type [Int](../data-types/int-uint.md). If `offset` is negative, it is counted from the end of the `input` string. +- `length`: Optional. An integer type [Int](../data-types/int-uint.md). `length` specifies the length of the snippet within input to be replaced. If `length` is not specified, the number of bytes removed from `input` equals the length of `replace`; otherwise `length` bytes are removed. + +**Returned value** + +- A [String](../data-types/string.md) data type value. + +**Example** + +```sql +SELECT overlay('ClickHouse SQL', 'CORE', 12) AS res; +``` + +Result: + +```text +┌─res─────────────┐ +│ ClickHouse CORE │ +└─────────────────┘ +``` + +## overlayUTF8 + +Replace part of the string `input` with another string `replace`, starting at the 1-based index `offset`. + +Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. + +**Syntax** + +```sql +overlayUTF8(s, replace, offset[, length]) +``` + +**Parameters** + +- `s`: A string type [String](../data-types/string.md). +- `replace`: A string type [String](../data-types/string.md). +- `offset`: An integer type [Int](../data-types/int-uint.md). If `offset` is negative, it is counted from the end of the `input` string. +- `length`: Optional. An integer type [Int](../data-types/int-uint.md). `length` specifies the length of the snippet within input to be replaced. If `length` is not specified, the number of characters removed from `input` equals the length of `replace`; otherwise `length` characters are removed. + +**Returned value** + +- A [String](../data-types/string.md) data type value. + +**Example** + +```sql +SELECT overlayUTF8('ClickHouse是一款OLAP数据库', '开源', 12, 2) AS res; +``` + +Result: + +```text +┌─res────────────────────────┐ +│ ClickHouse是开源OLAP数据库 │ +└────────────────────────────┘ +``` + ## replaceOne Replaces the first occurrence of the substring `pattern` in `haystack` by the `replacement` string. @@ -248,76 +320,3 @@ select printf('%%%s %s %d', 'Hello', 'World', 2024); │ %Hello World 2024 │ └──────────────────────────────────────────────┘ ``` - -## overlay - -Replace a part of a string `s` with another string `replace`, starting at 1-based index `offset`. By default, the number of bytes removed from `s` equals the length of `replace`. If `length` (the optional fourth argument) is specified, a different number of bytes is removed. - - -**Syntax** - -```sql -overlay(s, replace, offset[, length]) -``` - -**Parameters** - -- `s`: A string type [String](../data-types/string.md). -- `replace`: A string type [String](../data-types/string.md). -- `offset`: An integer type [Int](../data-types/int-uint.md). -- `length`: Optional. An integer type [Int](../data-types/int-uint.md). - -**Returned value** - -- A [String](../data-types/string.md) data type value. If `offset` is negative the offset is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. - -**Example** - -```sql -SELECT overlay('Spark SQL', 'CORE', 7) AS res; -``` - -Result: - -```text - ┌─res────────┐ - │ Spark CORE │ - └────────────┘ -``` - -## overlayUTF8 - -Replace a part of a string `s` with another string `replace`, starting at 1-based index `offset`. By default, the number of characters removed from `s` equals the length of `replace`. If `length` (the optional fourth argument) is specified, a different number of characters is removed. - -Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. - -**Syntax** - -```sql -overlayUTF8(s, replace, offset[, length]) -``` - -**Parameters** - -- `s`: A string type [String](../data-types/string.md). -- `replace`: A string type [String](../data-types/string.md). -- `offset`: An integer type [Int](../data-types/int-uint.md). -- `length`: Optional. An integer type [Int](../data-types/int-uint.md). - -**Returned value** - -- A [String](../data-types/string.md) data type value. If `offset` is negative the offset is counted starting from the back. `length` specifies the length of the snippet within input to be replaced. - -**Example** - -```sql -SELECT overlayUTF8('ClickHouse是一款OLAP数据库', '开源', 12, 2) AS res; -``` - -Result: - -```text -┌─res────────────────────────┐ -│ ClickHouse是开源OLAP数据库 │ -└────────────────────────────┘ -``` diff --git a/src/Functions/overlay.cpp b/src/Functions/overlay.cpp index 20988c775a5..497ebb9c9cd 100644 --- a/src/Functions/overlay.cpp +++ b/src/Functions/overlay.cpp @@ -201,14 +201,14 @@ private: { if (offset > 0) { - if (static_cast(offset) > input_size + 1) [[unlikely]] + if (static_cast(offset) > input_size + 1) return input_size; else return offset - 1; } else { - if (input_size < -static_cast(offset)) [[unlikely]] + if (input_size < -static_cast(offset)) return 0; else return input_size + offset; @@ -704,14 +704,14 @@ REGISTER_FUNCTION(Overlay) { factory.registerFunction>( {.description = R"( -Replace a part of a string `s` with another string `replace`, starting at 1-based index `offset`. By default, the number of bytes removed from `s` equals the length of `replace`. If `length` (the optional fourth argument) is specified, a different number of bytes is removed. +Replace a part of a string `input` with another string `replace`, starting at 1-based index `offset`. By default, the number of bytes removed from `input` equals the length of `replace`. If `length` (the optional fourth argument) is specified, a different number of bytes is removed. )", .categories{"String"}}, FunctionFactory::Case::Insensitive); factory.registerFunction>( {.description = R"( -Replace a part of a string `s` with another string `replace`, starting at 1-based index `offset`. By default, the number of bytes removed from `s` equals the length of `replace`. If `length` (the optional fourth argument) is specified, a different number of bytes is removed. +Replace a part of a string `input` with another string `replace`, starting at 1-based index `offset`. By default, the number of characters removed from `input` equals the length of `replace`. If `length` (the optional fourth argument) is specified, a different number of characters is removed. Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. )", diff --git a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference index 6495b6619f9..c39f1fb1ce9 100644 --- a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference +++ b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference @@ -512,8 +512,6 @@ nullIf nullIn nullInIgnoreSet or -overlay -overlayUTF8 parseDateTime parseDateTime32BestEffort parseDateTime32BestEffortOrNull From 3ff9522b69ec7e51119f445152ffb9678a0f124f Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Thu, 22 Aug 2024 12:49:10 +0800 Subject: [PATCH 188/260] change as request --- src/Functions/overlay.cpp | 165 +++++++++--------- .../0_stateless/03205_overlay.reference | 58 +++--- tests/queries/0_stateless/03205_overlay.sql | 11 +- 3 files changed, 115 insertions(+), 119 deletions(-) diff --git a/src/Functions/overlay.cpp b/src/Functions/overlay.cpp index 497ebb9c9cd..df8b825eabe 100644 --- a/src/Functions/overlay.cpp +++ b/src/Functions/overlay.cpp @@ -53,139 +53,132 @@ public: if (input_rows_count == 0) return ColumnString::create(); - const size_t number_of_arguments = arguments.size(); - bool has_three_args = number_of_arguments == 3; + bool has_four_args = (arguments.size() == 4); - ColumnPtr column_offset = arguments[2].column; - ColumnPtr column_length; - if (!has_three_args) - column_length = arguments[3].column; + ColumnPtr col_input = arguments[0].column; + const auto * col_input_const = checkAndGetColumn(col_input.get()); + const auto * col_input_string = checkAndGetColumn(col_input.get()); + bool input_is_const = (col_input_const != nullptr); - const ColumnConst * column_offset_const = checkAndGetColumn(column_offset.get()); - const ColumnConst * column_length_const = nullptr; - if (!has_three_args) - column_length_const = checkAndGetColumn(column_length.get()); + ColumnPtr col_replace = arguments[1].column; + const auto * col_replace_const = checkAndGetColumn(col_replace.get()); + const auto * col_replace_string = checkAndGetColumn(col_replace.get()); + bool replace_is_const = (col_replace_const != nullptr); + ColumnPtr col_offset = arguments[2].column; + const ColumnConst * col_offset_const = checkAndGetColumn(col_offset.get()); bool offset_is_const = false; - bool length_is_const = false; Int64 offset = -1; - Int64 length = -1; - if (column_offset_const) + if (col_offset_const) { - offset = column_offset_const->getInt(0); + offset = col_offset_const->getInt(0); offset_is_const = true; } - if (column_length_const) + ColumnPtr col_length = has_four_args ? arguments[3].column : nullptr; + const ColumnConst * col_length_const = has_four_args ? checkAndGetColumn(col_length.get()) : nullptr; + bool length_is_const = false; + Int64 length = -1; + if (col_length_const) { - length = column_length_const->getInt(0); + length = col_length_const->getInt(0); length_is_const = true; } - auto res_col = ColumnString::create(); auto & res_data = res_col->getChars(); auto & res_offsets = res_col->getOffsets(); + res_offsets.resize_exact(input_rows_count); - - ColumnPtr column_input = arguments[0].column; - ColumnPtr column_replace = arguments[1].column; - - const auto * column_input_const = checkAndGetColumn(column_input.get()); - const auto * column_input_string = checkAndGetColumn(column_input.get()); - if (column_input_const) + if (col_input_const) { - StringRef input = column_input_const->getDataAt(0); + StringRef input = col_input_const->getDataAt(0); res_data.reserve((input.size + 1) * input_rows_count); } else { - res_data.reserve(column_input_string->getChars().size()); + res_data.reserve(col_input_string->getChars().size()); } - const auto * column_replace_const = checkAndGetColumn(column_replace.get()); - const auto * column_replace_string = checkAndGetColumn(column_replace.get()); - bool input_is_const = (column_input_const != nullptr); - bool replace_is_const = (column_replace_const != nullptr); -#define OVERLAY_EXECUTE_CASE(THREE_ARGS, OFFSET_IS_CONST, LENGTH_IS_CONST) \ +#define OVERLAY_EXECUTE_CASE(HAS_FOUR_ARGS, OFFSET_IS_CONST, LENGTH_IS_CONST) \ if (input_is_const && replace_is_const) \ - constantConstant( \ + constantConstant( \ input_rows_count, \ - column_input_const->getDataAt(0), \ - column_replace_const->getDataAt(0), \ - column_offset, \ - column_length, \ + col_input_const->getDataAt(0), \ + col_replace_const->getDataAt(0), \ + col_offset, \ + col_length, \ offset, \ length, \ res_data, \ res_offsets); \ else if (input_is_const && !replace_is_const) \ - constantVector( \ + constantVector( \ input_rows_count, \ - column_input_const->getDataAt(0), \ - column_replace_string->getChars(), \ - column_replace_string->getOffsets(), \ - column_offset, \ - column_length, \ + col_input_const->getDataAt(0), \ + col_replace_string->getChars(), \ + col_replace_string->getOffsets(), \ + col_offset, \ + col_length, \ offset, \ length, \ res_data, \ res_offsets); \ else if (!input_is_const && replace_is_const) \ - vectorConstant( \ + vectorConstant( \ input_rows_count, \ - column_input_string->getChars(), \ - column_input_string->getOffsets(), \ - column_replace_const->getDataAt(0), \ - column_offset, \ - column_length, \ + col_input_string->getChars(), \ + col_input_string->getOffsets(), \ + col_replace_const->getDataAt(0), \ + col_offset, \ + col_length, \ offset, \ length, \ res_data, \ res_offsets); \ else \ - vectorVector( \ + vectorVector( \ input_rows_count, \ - column_input_string->getChars(), \ - column_input_string->getOffsets(), \ - column_replace_string->getChars(), \ - column_replace_string->getOffsets(), \ - column_offset, \ - column_length, \ + col_input_string->getChars(), \ + col_input_string->getOffsets(), \ + col_replace_string->getChars(), \ + col_replace_string->getOffsets(), \ + col_offset, \ + col_length, \ offset, \ length, \ res_data, \ res_offsets); - if (has_three_args) + if (!has_four_args) { if (offset_is_const) { - OVERLAY_EXECUTE_CASE(true, true, false) + OVERLAY_EXECUTE_CASE(false, true, false) } else { - OVERLAY_EXECUTE_CASE(true, false, false) + OVERLAY_EXECUTE_CASE(false, false, false) } } else { if (offset_is_const && length_is_const) { - OVERLAY_EXECUTE_CASE(false, true, true) + OVERLAY_EXECUTE_CASE(true, true, true) } else if (offset_is_const && !length_is_const) { - OVERLAY_EXECUTE_CASE(false, true, false) + OVERLAY_EXECUTE_CASE(true, true, false) } else if (!offset_is_const && length_is_const) { - OVERLAY_EXECUTE_CASE(false, false, true) + OVERLAY_EXECUTE_CASE(true, false, true) } else { - OVERLAY_EXECUTE_CASE(false, false, false) + OVERLAY_EXECUTE_CASE(true, false, false) } } #undef OVERLAY_EXECUTE_CASE @@ -224,7 +217,7 @@ private: return bytes; } - template + template void constantConstant( size_t rows, const StringRef & input, @@ -236,7 +229,7 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!has_three_args && length_is_const && const_length < 0) + if (has_four_args && length_is_const && const_length < 0) { constantConstant( rows, input, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); @@ -250,12 +243,12 @@ private: size_t replace_size = getSliceSize(reinterpret_cast(replace.data), replace.size); size_t valid_length = 0; // not negative - if constexpr (!has_three_args && length_is_const) + if constexpr (has_four_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; } - else if constexpr (has_three_args) + else if constexpr (!has_four_args) { valid_length = replace_size; } @@ -273,14 +266,14 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (!has_three_args && !length_is_const) + if constexpr (has_four_args && !length_is_const) { length = column_length->getInt(i); valid_length = length >= 0 ? length : replace_size; } size_t prefix_size = valid_offset; - size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t suffix_size = (prefix_size + valid_length > input_size) ? 0 : (input_size - prefix_size - valid_length); if constexpr (!is_utf8) { @@ -332,11 +325,12 @@ private: /// add zero terminator res_data[res_offset] = 0; ++res_offset; + res_offsets[i] = res_offset; } } - template + template void vectorConstant( size_t rows, const ColumnString::Chars & input_data, @@ -349,7 +343,7 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!has_three_args && length_is_const && const_length < 0) + if (has_four_args && length_is_const && const_length < 0) { vectorConstant( rows, input_data, input_offsets, replace, column_offset, column_length, const_offset, -1, res_data, res_offsets); @@ -359,12 +353,12 @@ private: size_t replace_size = getSliceSize(reinterpret_cast(replace.data), replace.size); Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative - if constexpr (!has_three_args && length_is_const) + if constexpr (has_four_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; } - else if constexpr (has_three_args) + else if constexpr (!has_four_args) { valid_length = replace_size; } @@ -388,14 +382,14 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (!has_three_args && !length_is_const) + if constexpr (has_four_args && !length_is_const) { length = column_length->getInt(i); valid_length = length >= 0 ? length : replace_size; } size_t prefix_size = valid_offset; - size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t suffix_size = (prefix_size + valid_length > input_size) ? 0 : (input_size - prefix_size - valid_length); if constexpr (!is_utf8) { @@ -449,11 +443,12 @@ private: /// add zero terminator res_data[res_offset] = 0; ++res_offset; + res_offsets[i] = res_offset; } } - template + template void constantVector( size_t rows, const StringRef & input, @@ -466,7 +461,7 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!has_three_args && length_is_const && const_length < 0) + if (has_four_args && length_is_const && const_length < 0) { constantVector( rows, input, replace_data, replace_offsets, column_offset, column_length, const_offset, -1, res_data, res_offsets); @@ -480,7 +475,7 @@ private: Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative - if constexpr (!has_three_args && length_is_const) + if constexpr (has_four_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; @@ -502,7 +497,7 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (has_three_args) + if constexpr (!has_four_args) { valid_length = replace_size; } @@ -513,7 +508,7 @@ private: } size_t prefix_size = valid_offset; - size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t suffix_size = (prefix_size + valid_length > input_size) ? 0 : (input_size - prefix_size - valid_length); if constexpr (!is_utf8) { @@ -563,11 +558,12 @@ private: /// add zero terminator res_data[res_offset] = 0; ++res_offset; + res_offsets[i] = res_offset; } } - template + template void vectorVector( size_t rows, const ColumnString::Chars & input_data, @@ -581,7 +577,7 @@ private: ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) const { - if (!has_three_args && length_is_const && const_length < 0) + if (has_four_args && length_is_const && const_length < 0) { vectorVector( rows, @@ -600,7 +596,7 @@ private: Int64 length = 0; // maybe negative size_t valid_length = 0; // not negative - if constexpr (!has_three_args && length_is_const) + if constexpr (has_four_args && length_is_const) { assert(const_length >= 0); valid_length = const_length; @@ -629,7 +625,7 @@ private: valid_offset = getValidOffset(offset, input_size); } - if constexpr (has_three_args) + if constexpr (!has_four_args) { valid_length = replace_size; } @@ -640,7 +636,7 @@ private: } size_t prefix_size = valid_offset; - size_t suffix_size = prefix_size + valid_length > input_size ? 0 : input_size - prefix_size - valid_length; + size_t suffix_size = (prefix_size + valid_length > input_size) ? 0 : (input_size - prefix_size - valid_length); if constexpr (!is_utf8) { @@ -693,6 +689,7 @@ private: /// add zero terminator res_data[res_offset] = 0; ++res_offset; + res_offsets[i] = res_offset; } } diff --git a/tests/queries/0_stateless/03205_overlay.reference b/tests/queries/0_stateless/03205_overlay.reference index 67a699944e0..4be3baadaea 100644 --- a/tests/queries/0_stateless/03205_overlay.reference +++ b/tests/queries/0_stateless/03205_overlay.reference @@ -25,34 +25,34 @@ Spark ANSI SQL Spark ANSI SQL和CH Spark ANSI SQL Spark ANSI SQL和CH Spark ANSI SQL Spark ANSI SQL和CH Spark ANSI SQL Spark ANSI SQL和CH -Test with different offset values --12 _park SQL _park SQL和CH --11 _park SQL S_ark SQL和CH --10 _park SQL Sp_rk SQL和CH --9 _park SQL Spa_k SQL和CH --8 S_ark SQL Spar_ SQL和CH --7 Sp_rk SQL Spark_SQL和CH --6 Spa_k SQL Spark _QL和CH --5 Spar_ SQL Spark S_L和CH --4 Spark_SQL Spark SQ_和CH --3 Spark _QL Spark SQL_CH --2 Spark S_L Spark SQL和_H --1 Spark SQ_ Spark SQL和C_ -0 Spark SQL_ Spark SQL和CH_ -1 _park SQL _park SQL和CH -2 S_ark SQL S_ark SQL和CH -3 Sp_rk SQL Sp_rk SQL和CH -4 Spa_k SQL Spa_k SQL和CH -5 Spar_ SQL Spar_ SQL和CH -6 Spark_SQL Spark_SQL和CH -7 Spark _QL Spark _QL和CH -8 Spark S_L Spark S_L和CH -9 Spark SQ_ Spark SQ_和CH -10 Spark SQL_ Spark SQL_CH -11 Spark SQL_ Spark SQL和_H -12 Spark SQL_ Spark SQL和C_ -13 Spark SQL_ Spark SQL和CH_ -Test with different length values +Test with special offset values +-12 __ark SQL 之park SQL和CH +-11 __ark SQL S之ark SQL和CH +-10 __ark SQL Sp之rk SQL和CH +-9 __ark SQL Spa之k SQL和CH +-8 S__rk SQL Spar之 SQL和CH +-7 Sp__k SQL Spark之SQL和CH +-6 Spa__ SQL Spark 之QL和CH +-5 Spar__SQL Spark S之L和CH +-4 Spark__QL Spark SQ之和CH +-3 Spark __L Spark SQL之CH +-2 Spark S__ Spark SQL和之H +-1 Spark SQ__ Spark SQL和C之 +0 Spark SQL__ Spark SQL和CH之 +1 __ark SQL 之park SQL和CH +2 S__rk SQL S之ark SQL和CH +3 Sp__k SQL Sp之rk SQL和CH +4 Spa__ SQL Spa之k SQL和CH +5 Spar__SQL Spar之 SQL和CH +6 Spark__QL Spark之SQL和CH +7 Spark __L Spark 之QL和CH +8 Spark S__ Spark S之L和CH +9 Spark SQ__ Spark SQ之和CH +10 Spark SQL__ Spark SQL之CH +11 Spark SQL__ Spark SQL和之H +12 Spark SQL__ Spark SQL和C之 +13 Spark SQL__ Spark SQL和CH之 +Test with special length values -1 Spark ANSI Spark ANSI H 0 Spark ANSI SQL Spark ANSI SQL和CH 1 Spark ANSI QL Spark ANSI QL和CH @@ -61,7 +61,7 @@ Test with different length values 4 Spark ANSI Spark ANSI CH 5 Spark ANSI Spark ANSI H 6 Spark ANSI Spark ANSI -Test with different input and replace values +Test with special input and replace values _ _ Spark SQL Spark SQL和CH ANSI ANSI diff --git a/tests/queries/0_stateless/03205_overlay.sql b/tests/queries/0_stateless/03205_overlay.sql index 4d0b5ecbe03..765b29f93ec 100644 --- a/tests/queries/0_stateless/03205_overlay.sql +++ b/tests/queries/0_stateless/03205_overlay.sql @@ -1,5 +1,4 @@ SELECT 'Negative test of overlay'; -SELECT overlay('hello', 2); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } SELECT overlay('hello', 'world'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } SELECT overlay('hello', 'world', 2, 3, 'extra'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } SELECT overlay(123, 'world', 2, 3); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } @@ -35,13 +34,13 @@ SELECT overlay(materialize('Spark SQL'), 'ANSI ', materialize(7), materialize(0) SELECT overlay('Spark SQL', materialize('ANSI '), materialize(7), materialize(0)), overlayUTF8('Spark SQL和CH', materialize('ANSI '), materialize(7), materialize(0)); SELECT overlay(materialize('Spark SQL'), materialize('ANSI '), materialize(7), materialize(0)), overlayUTF8(materialize('Spark SQL和CH'), materialize('ANSI '), materialize(7), materialize(0)); -SELECT 'Test with different offset values'; -WITH number - 12 as offset SELECT offset, overlay('Spark SQL', '_', offset), overlayUTF8('Spark SQL和CH', '_', offset) from numbers(26); +SELECT 'Test with special offset values'; +WITH number - 12 AS offset SELECT offset, overlay('Spark SQL', '__', offset), overlayUTF8('Spark SQL和CH', '之', offset) FROM numbers(26); -SELECT 'Test with different length values'; -WITH number - 1 as length SELECT length, overlay('Spark SQL', 'ANSI ', 7, length), overlayUTF8('Spark SQL和CH', 'ANSI ', 7, length) from numbers(8); +SELECT 'Test with special length values'; +WITH number - 1 AS length SELECT length, overlay('Spark SQL', 'ANSI ', 7, length), overlayUTF8('Spark SQL和CH', 'ANSI ', 7, length) FROM numbers(8); -SELECT 'Test with different input and replace values'; +SELECT 'Test with special input and replace values'; SELECT overlay('', '_', 6), overlayUTF8('', '_', 6); SELECT overlay('Spark SQL', '', 6), overlayUTF8('Spark SQL和CH', '', 6); SELECT overlay('', 'ANSI ', 7, 0), overlayUTF8('', 'ANSI ', 7, 0); From be4439e3ec0a1491f4e333ac848844fd930a6e5b Mon Sep 17 00:00:00 2001 From: Alexey Date: Thu, 22 Aug 2024 10:30:48 +0300 Subject: [PATCH 189/260] Update install.md Added correct commands for russian vwersion of the installation from deb packets --- docs/ru/getting-started/install.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/ru/getting-started/install.md b/docs/ru/getting-started/install.md index aee445da843..4a0ec258c64 100644 --- a/docs/ru/getting-started/install.md +++ b/docs/ru/getting-started/install.md @@ -25,10 +25,10 @@ grep -q sse4_2 /proc/cpuinfo && echo "SSE 4.2 supported" || echo "SSE 4.2 not su Яндекс рекомендует использовать официальные скомпилированные `deb`-пакеты для Debian или Ubuntu. Для установки пакетов выполните: ``` bash -sudo apt-get install -y apt-transport-https ca-certificates dirmngr -sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 8919F6BD2B48D754 +sudo apt-get install -y apt-transport-https ca-certificates curl gnupg +curl -fsSL 'https://packages.clickhouse.com/rpm/lts/repodata/repomd.xml.key' | sudo gpg --dearmor -o /usr/share/keyrings/clickhouse-keyring.gpg -echo "deb https://packages.clickhouse.com/deb stable main" | sudo tee \ +echo "deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb stable main" | sudo tee \ /etc/apt/sources.list.d/clickhouse.list sudo apt-get update From 6466f374e0372b22a23d1193e534bd6c94f87b94 Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Thu, 22 Aug 2024 11:29:33 +0200 Subject: [PATCH 190/260] Update geohash.md --- .../en/sql-reference/functions/geo/geohash.md | 31 ++++++++++++++----- 1 file changed, 24 insertions(+), 7 deletions(-) diff --git a/docs/en/sql-reference/functions/geo/geohash.md b/docs/en/sql-reference/functions/geo/geohash.md index b6ac7a74092..c4f41fc53da 100644 --- a/docs/en/sql-reference/functions/geo/geohash.md +++ b/docs/en/sql-reference/functions/geo/geohash.md @@ -6,7 +6,7 @@ title: "Functions for Working with Geohash" ## Geohash -[Geohash](https://en.wikipedia.org/wiki/Geohash) is the geocode system, which subdivides Earth’s surface into buckets of grid shape and encodes each cell into a short string of letters and digits. It is a hierarchical data structure, so the longer is the geohash string, the more precise is the geographic location. +[Geohash](https://en.wikipedia.org/wiki/Geohash) is the geocode system, which subdivides Earth’s surface into buckets of grid shape and encodes each cell into a short string of letters and digits. It is a hierarchical data structure, so the longer the geohash string is, the more precise the geographic location will be. If you need to manually convert geographic coordinates to geohash strings, you can use [geohash.org](http://geohash.org/). @@ -14,26 +14,37 @@ If you need to manually convert geographic coordinates to geohash strings, you c Encodes latitude and longitude as a [geohash](#geohash)-string. +**Syntax** + ``` sql geohashEncode(longitude, latitude, [precision]) ``` **Input values** -- longitude - longitude part of the coordinate you want to encode. Floating in range`[-180°, 180°]` -- latitude - latitude part of the coordinate you want to encode. Floating in range `[-90°, 90°]` -- precision - Optional, length of the resulting encoded string, defaults to `12`. Integer in range `[1, 12]`. Any value less than `1` or greater than `12` is silently converted to `12`. +- `longitude` — Longitude part of the coordinate you want to encode. Floating in range`[-180°, 180°]`. [Float](../../data_types/float.md). +- `latitude` — Latitude part of the coordinate you want to encode. Floating in range `[-90°, 90°]`. [Float](../../data_types/float.md). +- `precision` (optional) — Length of the resulting encoded string. Defaults to `12`. Integer in the range `[1, 12]`. [Int8](../../data-types/int-uint.md). + +:::note +- All coordinate parameters must be of the same type: either `Float32` or `Float64`. +- For the `precision` parameter, any value less than `1` or greater than `12` is silently converted to `12`. +::: **Returned values** -- alphanumeric `String` of encoded coordinate (modified version of the base32-encoding alphabet is used). +- Alphanumeric string of the encoded coordinate (modified version of the base32-encoding alphabet is used). [String](../../data-types/string.md). **Example** +Query: + ``` sql SELECT geohashEncode(-5.60302734375, 42.593994140625, 0) AS res; ``` +Result: + ``` text ┌─res──────────┐ │ ezs42d000000 │ @@ -44,13 +55,19 @@ SELECT geohashEncode(-5.60302734375, 42.593994140625, 0) AS res; Decodes any [geohash](#geohash)-encoded string into longitude and latitude. +**Syntax** + +```sql +geohashDecode(hash_str) +``` + **Input values** -- encoded string - geohash-encoded string. +- `hash_str` — Geohash-encoded string. **Returned values** -- (longitude, latitude) - 2-tuple of `Float64` values of longitude and latitude. +- Tuple `(longitude, latitude)` of `Float64` values of longitude and latitude. [Tuple](../../data-types/tuple.md)([Float64](../../data-types/float.md)) **Example** From 95f45d2eaf39a9e8a6373c75749ec57f727be700 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 13:20:04 +0200 Subject: [PATCH 191/260] try to fix tests --- .../test_storage_azure_blob_storage/test.py | 14 +++++------ tests/integration/test_storage_hdfs/test.py | 25 +++---------------- .../03203_hive_style_partitioning.reference | 20 +++++++-------- .../03203_hive_style_partitioning.sh | 14 +++-------- 4 files changed, 23 insertions(+), 50 deletions(-) diff --git a/tests/integration/test_storage_azure_blob_storage/test.py b/tests/integration/test_storage_azure_blob_storage/test.py index a3172329a99..c1f518e45ce 100644 --- a/tests/integration/test_storage_azure_blob_storage/test.py +++ b/tests/integration/test_storage_azure_blob_storage/test.py @@ -1513,14 +1513,14 @@ def test_hive_partitioning_with_one_parameter(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," - f" container='cont', blob_path='{path}', format='CSV', compression='auto', structure='{table_format}') VALUES {values}", + f" container='cont', blob_path='{path}', format='CSVWithNames', compression='auto', structure='{table_format}') VALUES {values}", settings={"azure_truncate_on_insert": 1}, ) query = ( f"SELECT column2, _file, _path, column1 FROM azureBlobStorage(azure_conf2, " f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " - f"blob_path='{path}', format='CSV', structure='{table_format}')" + f"blob_path='{path}', format='CSVWithNames', structure='{table_format}')" ) assert azure_query( node, query, settings={"use_hive_partitioning": 1} @@ -1533,7 +1533,7 @@ def test_hive_partitioning_with_one_parameter(cluster): query = ( f"SELECT column2 FROM azureBlobStorage(azure_conf2, " f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " - f"blob_path='{path}', format='CSV', structure='{table_format}');" + f"blob_path='{path}', format='CSVWithNames', structure='{table_format}');" ) assert azure_query( node, query, settings={"use_hive_partitioning": 1} @@ -1551,14 +1551,14 @@ def test_hive_partitioning_with_all_parameters(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," - f" container='cont', blob_path='{path}', format='CSV', compression='auto', structure='{table_format}') VALUES {values_1}, {values_2}", + f" container='cont', blob_path='{path}', format='CSVWithNames', compression='auto', structure='{table_format}') VALUES {values_1}, {values_2}", settings={"azure_truncate_on_insert": 1}, ) query = ( f"SELECT column1, column2, _file, _path FROM azureBlobStorage(azure_conf2, " f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " - f"blob_path='{path}', format='CSV', structure='{table_format}');" + f"blob_path='{path}', format='CSVWithNames', structure='{table_format}');" ) pattern = r"DB::Exception: Cannot use hive partitioning for file" @@ -1577,14 +1577,14 @@ def test_hive_partitioning_without_setting(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," - f" container='cont', blob_path='{path}', format='CSV', compression='auto', structure='{table_format}') VALUES {values_1}, {values_2}", + f" container='cont', blob_path='{path}', format='CSVWithNames', compression='auto', structure='{table_format}') VALUES {values_1}, {values_2}", settings={"azure_truncate_on_insert": 1}, ) query = ( f"SELECT column1, column2, _file, _path, column3 FROM azureBlobStorage(azure_conf2, " f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " - f"blob_path='{path}', format='CSV', structure='{table_format}');" + f"blob_path='{path}', format='CSVWithNames', structure='{table_format}');" ) pattern = re.compile( r"DB::Exception: Unknown expression identifier '.*' in scope.*", re.DOTALL diff --git a/tests/integration/test_storage_hdfs/test.py b/tests/integration/test_storage_hdfs/test.py index a75c13b9ea6..31cc8609eb4 100644 --- a/tests/integration/test_storage_hdfs/test.py +++ b/tests/integration/test_storage_hdfs/test.py @@ -1259,35 +1259,16 @@ def test_respect_object_existence_on_partitioned_write(started_cluster): def test_hive_partitioning_with_one_parameter(started_cluster): hdfs_api = started_cluster.hdfs_api - hdfs_api.write_data(f"/column0=Elizabeth/parquet_1", f"Elizabeth\tGordon\n") - assert hdfs_api.read_data(f"/column0=Elizabeth/parquet_1") == f"Elizabeth\tGordon\n" + hdfs_api.write_data(f"/column0=Elizabeth/file_1", f"column0,column1\nElizabeth,Gordon\n") + assert hdfs_api.read_data(f"/column0=Elizabeth/file_1") == f"column0,column1\nElizabeth,Gordon\n" r = node1.query( - "SELECT column0 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/parquet_1', 'TSV')", + "SELECT column0 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/file_1', 'CSVWithNames')", settings={"use_hive_partitioning": 1}, ) assert r == f"Elizabeth\n" -def test_hive_partitioning_with_all_parameters(started_cluster): - hdfs_api = started_cluster.hdfs_api - hdfs_api.write_data( - f"/column0=Elizabeth/column1=Gordon/parquet_2", f"Elizabeth\tGordon\n" - ) - assert ( - hdfs_api.read_data(f"/column0=Elizabeth/column1=Gordon/parquet_2") - == f"Elizabeth\tGordon\n" - ) - - pattern = r"DB::Exception: Cannot use hive partitioning for file" - - with pytest.raises(QueryRuntimeException, match=pattern): - node1.query( - f"SELECT column1 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/column1=Gordon/parquet_2', 'TSV');", - settings={"use_hive_partitioning": 1}, - ) - - def test_hive_partitioning_without_setting(started_cluster): hdfs_api = started_cluster.hdfs_api hdfs_api.write_data( diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.reference b/tests/queries/0_stateless/03203_hive_style_partitioning.reference index af52dcd9b88..acdadc2510b 100644 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.reference +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.reference @@ -1,5 +1,5 @@ TESTING THE FILE HIVE PARTITIONING -last Elizabeth + last Elizabeth Frank Elizabeth Moreno Elizabeth Guzman Elizabeth @@ -19,8 +19,7 @@ Stanley Gibson Elizabeth Eugenia Greer Elizabeth Jeffery Delgado Elizabeth Clara Cross Elizabeth -Elizabeth Gordon Elizabeth -last Elizabeth + last Elizabeth Frank Elizabeth Moreno Elizabeth Guzman Elizabeth @@ -35,12 +34,13 @@ Cross Elizabeth Array(Int64) LowCardinality(Float64) 101 2070 +4081 +2070 2070 b 1 -1 TESTING THE URL PARTITIONING -last Elizabeth + last Elizabeth Frank Elizabeth Moreno Elizabeth Guzman Elizabeth @@ -60,10 +60,9 @@ Stanley Gibson Elizabeth Eugenia Greer Elizabeth Jeffery Delgado Elizabeth Clara Cross Elizabeth -Elizabeth Gordon Elizabeth 1 TESTING THE S3 PARTITIONING -last Elizabeth + last Elizabeth Frank Elizabeth Moreno Elizabeth Guzman Elizabeth @@ -83,8 +82,7 @@ Stanley Gibson Elizabeth Eugenia Greer Elizabeth Jeffery Delgado Elizabeth Clara Cross Elizabeth -Elizabeth Gordon Elizabeth -last Elizabeth + last Elizabeth Frank Elizabeth Moreno Elizabeth Guzman Elizabeth @@ -96,7 +94,7 @@ Delgado Elizabeth Cross Elizabeth OK TESTING THE S3CLUSTER PARTITIONING -last Elizabeth + last Elizabeth Frank Elizabeth Moreno Elizabeth Guzman Elizabeth @@ -106,7 +104,7 @@ Gibson Elizabeth Greer Elizabeth Delgado Elizabeth Cross Elizabeth -last Elizabeth + last Elizabeth Frank Elizabeth Moreno Elizabeth Guzman Elizabeth diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.sh b/tests/queries/0_stateless/03203_hive_style_partitioning.sh index 4e165446c34..b3d196924af 100755 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.sh +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.sh @@ -14,7 +14,7 @@ set use_hive_partitioning = 1; SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; SELECT *, non_existing_column FROM file('$CURDIR/data_hive/partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10; -SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=*/sample.parquet') WHERE column0 = 'Elizabeth'; +SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=*/sample.parquet') WHERE column0 = 'Elizabeth' LIMIT 10; SELECT number, date FROM file('$CURDIR/data_hive/partitioning/number=42/date=2020-01-01/sample.parquet') LIMIT 1; SELECT array, float FROM file('$CURDIR/data_hive/partitioning/array=[1,2,3]/float=42.42/sample.parquet') LIMIT 1; @@ -29,16 +29,10 @@ SELECT identifier FROM file('$CURDIR/data_hive/partitioning/identifier=*/email.c SELECT a FROM file('$CURDIR/data_hive/partitioning/a=b/a=b/sample.parquet') LIMIT 1; """ -$CLICKHOUSE_LOCAL -n -q """ -set use_hive_partitioning = 1; - -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth1/sample.parquet') LIMIT 10; -""" 2>&1 | grep -c "INCORRECT_DATA" - $CLICKHOUSE_LOCAL -n -q """ set use_hive_partitioning = 0; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; +SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; """ 2>&1 | grep -c "UNKNOWN_IDENTIFIER" @@ -68,7 +62,7 @@ set use_hive_partitioning = 1; SELECT *, column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; SELECT *, non_existing_column FROM s3('http://localhost:11111/test/hive_partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10; -SELECT *, column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=*/sample.parquet') WHERE column0 = 'Elizabeth'; +SELECT *, column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=*/sample.parquet') WHERE column0 = 'Elizabeth' LIMIT 10; """ $CLICKHOUSE_CLIENT -n -q """ @@ -84,5 +78,5 @@ set use_hive_partitioning = 1; SELECT *, column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; -SELECT *, column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = 'Elizabeth'; +SELECT *, column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = 'Elizabeth' LIMIT 10; """ From 62054cae666244fd072a56f70a6df73e68249cb0 Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Thu, 22 Aug 2024 13:49:16 +0200 Subject: [PATCH 192/260] Update geohash.md --- docs/en/sql-reference/functions/geo/geohash.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/en/sql-reference/functions/geo/geohash.md b/docs/en/sql-reference/functions/geo/geohash.md index c4f41fc53da..ce2e3c43b3e 100644 --- a/docs/en/sql-reference/functions/geo/geohash.md +++ b/docs/en/sql-reference/functions/geo/geohash.md @@ -22,8 +22,8 @@ geohashEncode(longitude, latitude, [precision]) **Input values** -- `longitude` — Longitude part of the coordinate you want to encode. Floating in range`[-180°, 180°]`. [Float](../../data_types/float.md). -- `latitude` — Latitude part of the coordinate you want to encode. Floating in range `[-90°, 90°]`. [Float](../../data_types/float.md). +- `longitude` — Longitude part of the coordinate you want to encode. Floating in range`[-180°, 180°]`. [Float](../../data-types/float.md). +- `latitude` — Latitude part of the coordinate you want to encode. Floating in range `[-90°, 90°]`. [Float](../../data-types/float.md). - `precision` (optional) — Length of the resulting encoded string. Defaults to `12`. Integer in the range `[1, 12]`. [Int8](../../data-types/int-uint.md). :::note From 84467077b886cd48c9cd33c69c1935b3f7863dd7 Mon Sep 17 00:00:00 2001 From: Vitaly Baranov Date: Thu, 22 Aug 2024 13:45:13 +0200 Subject: [PATCH 193/260] Fix test for role expiration in RoleCache. --- tests/integration/test_role/test.py | 81 +++++++++-------------------- 1 file changed, 26 insertions(+), 55 deletions(-) diff --git a/tests/integration/test_role/test.py b/tests/integration/test_role/test.py index b3b18dc8271..9d15f0f81db 100644 --- a/tests/integration/test_role/test.py +++ b/tests/integration/test_role/test.py @@ -418,72 +418,43 @@ def test_function_current_roles(): ) -def test_role_expiration(): - instance.query("CREATE USER ure") +@pytest.mark.parametrize("with_extra_role", [False, True]) +def test_role_expiration(with_extra_role): instance.query("CREATE ROLE rre") - instance.query("GRANT rre TO ure") + instance.query("CREATE USER ure DEFAULT ROLE rre") - instance.query("CREATE TABLE IF NOT EXISTS tre (id Int) Engine=Log") - instance.query("INSERT INTO tre VALUES (0)") + instance.query("CREATE TABLE table1 (id Int) Engine=Log") + instance.query("CREATE TABLE table2 (id Int) Engine=Log") + instance.query("INSERT INTO table1 VALUES (1)") + instance.query("INSERT INTO table2 VALUES (2)") + instance.query("GRANT SELECT ON table1 TO rre") + + assert instance.query("SELECT * FROM table1", user="ure") == "1\n" assert "Not enough privileges" in instance.query_and_get_error( - "SELECT * FROM tre", user="ure" + "SELECT * FROM table2", user="ure" ) - instance.query("GRANT SELECT ON tre TO rre") - - assert instance.query("SELECT * FROM tre", user="ure") == "0\n" - # access_control_improvements/role_cache_expiration_time_seconds value is 2 for the test # so we wait >2 seconds until the role is expired time.sleep(5) - instance.query("CREATE TABLE IF NOT EXISTS tre1 (id Int) Engine=Log") - instance.query("INSERT INTO tre1 VALUES (0)") - instance.query("GRANT SELECT ON tre1 TO rre") + if with_extra_role: + # Expiration of role "rre" from the role cache can be caused by another role being used. + instance.query("CREATE ROLE extra_role") + instance.query("CREATE USER extra_user DEFAULT ROLE extra_role") + instance.query("GRANT SELECT ON table1 TO extra_role") + assert instance.query("SELECT * FROM table1", user="extra_user") == "1\n" - assert instance.query("SELECT * from tre1", user="ure") == "0\n" + instance.query("GRANT SELECT ON table2 TO rre") + assert instance.query("SELECT * FROM table1", user="ure") == "1\n" + assert instance.query("SELECT * FROM table2", user="ure") == "2\n" - instance.query("DROP USER ure") instance.query("DROP ROLE rre") - instance.query("DROP TABLE tre") - instance.query("DROP TABLE tre1") - - -def test_two_roles_expiration(): - instance.query("CREATE USER ure") - instance.query("CREATE ROLE rre") - instance.query("GRANT rre TO ure") - - instance.query("CREATE ROLE rre_second") - - instance.query("CREATE TABLE IF NOT EXISTS tre (id Int) Engine=Log") - instance.query("INSERT INTO tre VALUES (0)") - - assert "Not enough privileges" in instance.query_and_get_error( - "SELECT * FROM tre", user="ure" - ) - - instance.query("GRANT SELECT ON tre TO rre") - - assert instance.query("SELECT * FROM tre", user="ure") == "0\n" - - # access_control_improvements/role_cache_expiration_time_seconds value is 2 for the test - # so we wait >2 seconds until the roles are expired - time.sleep(5) - - instance.query( - "GRANT SELECT ON tre1 TO rre_second" - ) # we expect that both rre and rre_second are gone from cache upon this operation - - instance.query("CREATE TABLE IF NOT EXISTS tre1 (id Int) Engine=Log") - instance.query("INSERT INTO tre1 VALUES (0)") - instance.query("GRANT SELECT ON tre1 TO rre") - - assert instance.query("SELECT * from tre1", user="ure") == "0\n" - instance.query("DROP USER ure") - instance.query("DROP ROLE rre") - instance.query("DROP ROLE rre_second") - instance.query("DROP TABLE tre") - instance.query("DROP TABLE tre1") + instance.query("DROP TABLE table1") + instance.query("DROP TABLE table2") + + if with_extra_role: + instance.query("DROP ROLE extra_role") + instance.query("DROP USER extra_user") From 664e9b3db9d47e45c642ad21e3a5273ab423199a Mon Sep 17 00:00:00 2001 From: Vitaly Baranov Date: Sun, 4 Aug 2024 13:30:41 +0200 Subject: [PATCH 194/260] Add one more test. --- tests/integration/test_role/test.py | 173 ++++++++++++++++++++++++++++ 1 file changed, 173 insertions(+) diff --git a/tests/integration/test_role/test.py b/tests/integration/test_role/test.py index 9d15f0f81db..225cab975ff 100644 --- a/tests/integration/test_role/test.py +++ b/tests/integration/test_role/test.py @@ -1,5 +1,6 @@ import time import pytest +import random from helpers.client import QueryRuntimeException from helpers.cluster import ClickHouseCluster from helpers.test_tools import TSV @@ -458,3 +459,175 @@ def test_role_expiration(with_extra_role): if with_extra_role: instance.query("DROP ROLE extra_role") instance.query("DROP USER extra_user") + + +def test_roles_cache(): + # This test takes 20 seconds. + test_time = 20 + + # Three users A, B, C. + users = ["A", "B", "C"] + instance.query("CREATE USER " + ", ".join(users)) + + # Table "tbl" has 10 columns. Each of the users has access to a different set of columns. + num_columns = 10 + columns = [f"x{i}" for i in range(1, num_columns + 1)] + columns_with_types = [column + " Int64" for column in columns] + columns_with_types_comma_separated = ", ".join(columns_with_types) + values = list(range(1, num_columns + 1)) + values_comma_separated = ", ".join([str(value) for value in values]) + instance.query( + f"CREATE TABLE tbl ({columns_with_types_comma_separated}) ENGINE=MergeTree ORDER BY tuple()" + ) + instance.query(f"INSERT INTO tbl VALUES ({values_comma_separated})") + columns_to_values = dict([(f"x{i}", i) for i in range(1, num_columns + 1)]) + + # In this test we create and modify roles multiple times along with updating the following variables. + # Then we check that each of the users has access to the expected set of columns. + roles = [] + users_to_roles = dict([(user, []) for user in users]) + roles_to_columns = {} + + # Checks that each of the users can access the expected set of columns and can't access other columns. + def check(): + for user in random.sample(users, len(users)): + expected_roles = users_to_roles[user] + expected_columns = list( + set(sum([roles_to_columns[role] for role in expected_roles], [])) + ) + expected_result = sorted( + [columns_to_values[column] for column in expected_columns] + ) + query = " UNION ALL ".join( + [ + f"SELECT * FROM viewIfPermitted(SELECT {column} AS c FROM tbl ELSE null('c Int64'))" + for column in columns + ] + ) + result = instance.query(query, user=user).splitlines() + result = sorted([int(value) for value in result]) + ok = result == expected_result + if not ok: + print(f"Show grants for {user}:") + print( + instance.query( + "SHOW GRANTS FOR " + ", ".join([user] + expected_roles) + ) + ) + print(f"Expected result: {expected_result}") + print(f"Got unexpected result: {result}") + assert ok + + # Grants one of our roles a permission to access one of the columns. + def grant_column(): + columns_used_in_roles = sum(roles_to_columns.values(), []) + columns_to_choose = [ + column for column in columns if column not in columns_used_in_roles + ] + if not columns_to_choose or not roles: + return False + column = random.choice(columns_to_choose) + role = random.choice(roles) + instance.query(f"GRANT SELECT({column}) ON tbl TO {role}") + roles_to_columns[role].append(column) + return True + + # Revokes a permission to access one of the granted column from all our roles. + def revoke_column(): + columns_used_in_roles = sum(roles_to_columns.values(), []) + columns_to_choose = list(set(columns_used_in_roles)) + if not columns_to_choose or not roles: + return False + column = random.choice(columns_to_choose) + roles_str = ", ".join(roles) + instance.query(f"REVOKE SELECT({column}) ON tbl FROM {roles_str}") + for role in roles_to_columns: + if column in roles_to_columns[role]: + roles_to_columns[role].remove(column) + return True + + # Creates a role and grants it to one of the users. + def create_role(): + for role in ["R1", "R2", "R3"]: + if role not in roles: + instance.query(f"CREATE ROLE {role}") + roles.append(role) + if role not in roles_to_columns: + roles_to_columns[role] = [] + if "R1" not in users_to_roles["A"]: + instance.query("GRANT R1 TO A") + users_to_roles["A"].append("R1") + elif "R2" not in users_to_roles["B"]: + instance.query("GRANT R2 TO B") + users_to_roles["B"].append("R2") + elif "R3" not in users_to_roles["B"]: + instance.query("GRANT R3 TO R2") + users_to_roles["B"].append("R3") + elif "R3" not in users_to_roles["C"]: + instance.query("GRANT R3 TO C") + users_to_roles["C"].append("R3") + else: + return False + return True + + # Drops one of our roles. + def drop_role(): + if not roles: + return False + role = random.choice(roles) + instance.query(f"DROP ROLE {role}") + roles.remove(role) + for u in users_to_roles: + if role in users_to_roles[u]: + users_to_roles[u].remove(role) + del roles_to_columns[role] + if (role == "R2") and ("R3" in users_to_roles["B"]): + users_to_roles["B"].remove("R3") + return True + + # Modifies some grants or roles randomly. + def modify(): + while True: + rnd = random.random() + if rnd < 0.4: + if grant_column(): + break + elif rnd < 0.5: + if revoke_column(): + break + elif rnd < 0.9: + if create_role(): + break + else: + if drop_role(): + break + + def maybe_modify(): + if random.random() < 0.9: + modify() + modify() + + # Sleeping is necessary in this test because the role cache in ClickHouse has expiration timeout. + def maybe_sleep(): + if random.random() < 0.1: + # "role_cache_expiration_time_seconds" is set to 2 seconds in the test configuration. + # We need a sleep longer than that in this test sometimes. + seconds = random.random() * 5 + print(f"Sleeping {seconds} seconds") + time.sleep(seconds) + + # Main part of the test. + start_time = time.time() + end_time = start_time + test_time + + while time.time() < end_time: + check() + maybe_sleep() + maybe_modify() + maybe_sleep() + + check() + + instance.query("DROP USER " + ", ".join(users)) + instance.query("DROP ROLE " + ", ".join(roles)) + instance.query("DROP TABLE tbl") From 7ef5c366e873c4fd99f257eefbb3a350848e308c Mon Sep 17 00:00:00 2001 From: Vitaly Baranov Date: Sun, 4 Aug 2024 13:33:50 +0200 Subject: [PATCH 195/260] Fix expiration in RoleCache. --- src/Access/RoleCache.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Access/RoleCache.cpp b/src/Access/RoleCache.cpp index 2d94df2eea5..cc1f1520b67 100644 --- a/src/Access/RoleCache.cpp +++ b/src/Access/RoleCache.cpp @@ -120,7 +120,7 @@ void RoleCache::collectEnabledRoles(EnabledRoles & enabled_roles, SubscriptionsO SubscriptionsOnRoles new_subscriptions_on_roles; new_subscriptions_on_roles.reserve(subscriptions_on_roles.size()); - auto get_role_function = [this, &subscriptions_on_roles](const UUID & id) TSA_NO_THREAD_SAFETY_ANALYSIS { return getRole(id, subscriptions_on_roles); }; + auto get_role_function = [this, &new_subscriptions_on_roles](const UUID & id) TSA_NO_THREAD_SAFETY_ANALYSIS { return getRole(id, new_subscriptions_on_roles); }; for (const auto & current_role : enabled_roles.params.current_roles) collectRoles(*new_info, skip_ids, get_role_function, current_role, true, false); From 54dd3afd49df9c92cd3621a5cec4c7464c341a71 Mon Sep 17 00:00:00 2001 From: alesapin Date: Thu, 22 Aug 2024 14:52:17 +0200 Subject: [PATCH 196/260] Turn off fault injection for insert in 01396_inactive_replica_cleanup_nodes_zookeeper --- .../01396_inactive_replica_cleanup_nodes_zookeeper.sh | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh b/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh index bff85b3e29f..9ea15071856 100755 --- a/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh +++ b/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh @@ -23,11 +23,10 @@ $CLICKHOUSE_CLIENT -n --query " DETACH TABLE r2; " -$CLICKHOUSE_CLIENT --max_block_size 1 --min_insert_block_size_rows 1 --min_insert_block_size_bytes 1 --max_insert_threads 16 --query "INSERT INTO r1 SELECT * FROM numbers_mt(${SCALE})" +# insert_keeper_fault_injection_probability=0 -- can slowdown insert a lot (produce a lot of parts) +$CLICKHOUSE_CLIENT --insert_keeper_fault_injection_probability=0 --max_block_size 1 --min_insert_block_size_rows 1 --min_insert_block_size_bytes 1 --max_insert_threads 16 --query "INSERT INTO r1 SELECT * FROM numbers_mt(${SCALE})" -# Now wait for cleanup thread - for _ in {1..60}; do $CLICKHOUSE_CLIENT --query "SYSTEM FLUSH LOGS" [[ $($CLICKHOUSE_CLIENT --query "SELECT sum(toUInt32(extract(message, 'Removed (\d+) old log entries'))) FROM system.text_log WHERE event_date >= yesterday() AND logger_name LIKE '%' || '$CLICKHOUSE_DATABASE' || '%r1%(ReplicatedMergeTreeCleanupThread)%' AND message LIKE '%Removed % old log entries%'") -gt $((SCALE - 10)) ]] && break; From 7a740819b9551a291827b9d37b8b724612587a20 Mon Sep 17 00:00:00 2001 From: alesapin Date: Thu, 22 Aug 2024 14:53:15 +0200 Subject: [PATCH 197/260] Accidentally deleted comment --- .../01396_inactive_replica_cleanup_nodes_zookeeper.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh b/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh index 9ea15071856..80e9253af2c 100755 --- a/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh +++ b/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh @@ -27,6 +27,7 @@ $CLICKHOUSE_CLIENT -n --query " $CLICKHOUSE_CLIENT --insert_keeper_fault_injection_probability=0 --max_block_size 1 --min_insert_block_size_rows 1 --min_insert_block_size_bytes 1 --max_insert_threads 16 --query "INSERT INTO r1 SELECT * FROM numbers_mt(${SCALE})" +# Now wait for cleanup thread for _ in {1..60}; do $CLICKHOUSE_CLIENT --query "SYSTEM FLUSH LOGS" [[ $($CLICKHOUSE_CLIENT --query "SELECT sum(toUInt32(extract(message, 'Removed (\d+) old log entries'))) FROM system.text_log WHERE event_date >= yesterday() AND logger_name LIKE '%' || '$CLICKHOUSE_DATABASE' || '%r1%(ReplicatedMergeTreeCleanupThread)%' AND message LIKE '%Removed % old log entries%'") -gt $((SCALE - 10)) ]] && break; From b3f084459f60b1e31c32736573af0810dee99230 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 14:53:53 +0200 Subject: [PATCH 198/260] fix black --- tests/integration/test_storage_hdfs/test.py | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_storage_hdfs/test.py b/tests/integration/test_storage_hdfs/test.py index 31cc8609eb4..b18940b7290 100644 --- a/tests/integration/test_storage_hdfs/test.py +++ b/tests/integration/test_storage_hdfs/test.py @@ -1259,8 +1259,13 @@ def test_respect_object_existence_on_partitioned_write(started_cluster): def test_hive_partitioning_with_one_parameter(started_cluster): hdfs_api = started_cluster.hdfs_api - hdfs_api.write_data(f"/column0=Elizabeth/file_1", f"column0,column1\nElizabeth,Gordon\n") - assert hdfs_api.read_data(f"/column0=Elizabeth/file_1") == f"column0,column1\nElizabeth,Gordon\n" + hdfs_api.write_data( + f"/column0=Elizabeth/file_1", f"column0,column1\nElizabeth,Gordon\n" + ) + assert ( + hdfs_api.read_data(f"/column0=Elizabeth/file_1") + == f"column0,column1\nElizabeth,Gordon\n" + ) r = node1.query( "SELECT column0 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/file_1', 'CSVWithNames')", @@ -1269,6 +1274,7 @@ def test_hive_partitioning_with_one_parameter(started_cluster): assert r == f"Elizabeth\n" + def test_hive_partitioning_without_setting(started_cluster): hdfs_api = started_cluster.hdfs_api hdfs_api.write_data( From 8d14d8523098a42cd778ef50a9b066508da8919c Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 15:24:33 +0200 Subject: [PATCH 199/260] fix black --- tests/integration/test_storage_hdfs/test.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/integration/test_storage_hdfs/test.py b/tests/integration/test_storage_hdfs/test.py index b18940b7290..7a92e8adb0d 100644 --- a/tests/integration/test_storage_hdfs/test.py +++ b/tests/integration/test_storage_hdfs/test.py @@ -1274,7 +1274,6 @@ def test_hive_partitioning_with_one_parameter(started_cluster): assert r == f"Elizabeth\n" - def test_hive_partitioning_without_setting(started_cluster): hdfs_api = started_cluster.hdfs_api hdfs_api.write_data( From add4718634317304f652579a9f201c3b81c96a7d Mon Sep 17 00:00:00 2001 From: Tanya Bragin Date: Thu, 22 Aug 2024 06:37:27 -0700 Subject: [PATCH 200/260] Update README.md - Meetups update Fixed one meetup location; Added more meetups --- README.md | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 5e66b9da73e..c9474ef0fc0 100644 --- a/README.md +++ b/README.md @@ -45,9 +45,17 @@ The following upcoming meetups are featuring creator of ClickHouse & CTO, Alexey * [ClickHouse Guangzhou User Group Meetup](https://mp.weixin.qq.com/s/GSvo-7xUoVzCsuUvlLTpCw) - August 25 * [San Francisco Meetup (Cloudflare)](https://www.meetup.com/clickhouse-silicon-valley-meetup-group/events/302540575) - September 5 * [Raleigh Meetup (Deutsche Bank)](https://www.meetup.com/clickhouse-nc-meetup-group/events/302557230) - September 9 -* [New York Meetup (Ramp)](https://www.meetup.com/clickhouse-new-york-user-group/events/302575342) - September 10 +* [New York Meetup (Rokt)](https://www.meetup.com/clickhouse-new-york-user-group/events/302575342) - September 10 * [Chicago Meetup (Jump Capital)](https://lu.ma/43tvmrfw) - September 12 +Other upcoming meetups +* [Seattle Meetup (Statsig)](https://www.meetup.com/clickhouse-seattle-user-group/events/302518075/) - August 27 +* [Melbourne Meetup](https://www.meetup.com/clickhouse-australia-user-group/events/302732666/) - August 27 +* [Sydney Meetup](https://www.meetup.com/clickhouse-australia-user-group/events/302862966/) - September 5 +* [Zurich Meetup](https://www.meetup.com/clickhouse-switzerland-meetup-group/events/302267429/) - September 5 +* [Toronto Meetup (Shopify)](https://www.meetup.com/clickhouse-toronto-user-group/events/301490855/) - September 10 +* [London Meetup](https://www.meetup.com/clickhouse-london-user-group/events/302977267) - September 17 + ## Recent Recordings * **Recent Meetup Videos**: [Meetup Playlist](https://www.youtube.com/playlist?list=PL0Z2YDlm0b3iNDUzpY1S3L_iV4nARda_U) Whenever possible recordings of the ClickHouse Community Meetups are edited and presented as individual talks. Current featuring "Modern SQL in 2023", "Fast, Concurrent, and Consistent Asynchronous INSERTS in ClickHouse", and "Full-Text Indices: Design and Experiments" * **Recording available**: [**v24.4 Release Call**](https://www.youtube.com/watch?v=dtUqgcfOGmE) All the features of 24.4, one convenient video! Watch it now! From 91e65feaaedd4806875aed3d4be4f07edeefdb71 Mon Sep 17 00:00:00 2001 From: Anton Popov Date: Thu, 22 Aug 2024 13:42:30 +0000 Subject: [PATCH 201/260] fix virtual columns in Merge engine --- src/Storages/StorageDistributed.cpp | 2 +- src/Storages/StorageMerge.cpp | 14 +++++++------- .../02890_describe_table_options.reference | 8 ++++++++ 3 files changed, 16 insertions(+), 8 deletions(-) diff --git a/src/Storages/StorageDistributed.cpp b/src/Storages/StorageDistributed.cpp index c4668159759..0b80858800b 100644 --- a/src/Storages/StorageDistributed.cpp +++ b/src/Storages/StorageDistributed.cpp @@ -290,7 +290,7 @@ VirtualColumnsDescription StorageDistributed::createVirtuals() desc.addEphemeral("_shard_num", std::make_shared(), "Deprecated. Use function shardNum instead"); - /// Add virtual columns from table of storage Merges. + /// Add virtual columns from table with Merge engine. desc.addEphemeral("_database", std::make_shared(std::make_shared()), "The name of database which the row comes from"); desc.addEphemeral("_table", std::make_shared(std::make_shared()), "The name of table which the row comes from"); diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp index e88844e2d31..0827321e296 100644 --- a/src/Storages/StorageMerge.cpp +++ b/src/Storages/StorageMerge.cpp @@ -642,10 +642,6 @@ std::vector ReadFromMerge::createChildrenPlans(SelectQ column_names_as_aliases.push_back(ExpressionActions::getSmallestColumn(storage_metadata_snapshot->getColumns().getAllPhysical()).name); } } - else - { - - } auto child = createPlanForTable( nested_storage_snaphsot, @@ -657,6 +653,7 @@ std::vector ReadFromMerge::createChildrenPlans(SelectQ row_policy_data_opt, modified_context, current_streams); + child.plan.addInterpreterContext(modified_context); if (child.plan.isInitialized()) @@ -914,12 +911,14 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextMutablePtr & mo modified_query_info.table_expression = replacement_table_expression; modified_query_info.planner_context->getOrCreateTableExpressionData(replacement_table_expression); - auto get_column_options = GetColumnsOptions(GetColumnsOptions::All).withExtendedObjects().withVirtuals(); - if (storage_snapshot_->storage.supportsSubcolumns()) - get_column_options.withSubcolumns(); + auto get_column_options = GetColumnsOptions(GetColumnsOptions::All) + .withExtendedObjects() + .withSubcolumns(storage_snapshot_->storage.supportsSubcolumns()); std::unordered_map column_name_to_node; + /// Consider only non-virtual columns of storage while checking for _table and _database columns. + /// I.e. always override virtual columns with these names from underlying table (if any). if (!storage_snapshot_->tryGetColumn(get_column_options, "_table")) { auto table_name_node = std::make_shared(current_storage_id.table_name); @@ -946,6 +945,7 @@ SelectQueryInfo ReadFromMerge::getModifiedQueryInfo(const ContextMutablePtr & mo column_name_to_node.emplace("_database", function_node); } + get_column_options.withVirtuals(); auto storage_columns = storage_snapshot_->metadata->getColumns(); bool with_aliases = /* common_processed_stage == QueryProcessingStage::FetchColumns && */ !storage_columns.getAliases().empty(); diff --git a/tests/queries/0_stateless/02890_describe_table_options.reference b/tests/queries/0_stateless/02890_describe_table_options.reference index 9181cb27cb0..b77ef4a0fdf 100644 --- a/tests/queries/0_stateless/02890_describe_table_options.reference +++ b/tests/queries/0_stateless/02890_describe_table_options.reference @@ -54,6 +54,8 @@ _row_exists UInt8 Persisted mask created by lightweight delete that show wheth _block_number UInt64 Persisted original number of block that was assigned at insert Delta, LZ4 1 _block_offset UInt64 Persisted original number of row in block that was assigned at insert Delta, LZ4 1 _shard_num UInt32 Deprecated. Use function shardNum instead 1 +_database LowCardinality(String) The name of database which the row comes from 1 +_table LowCardinality(String) The name of table which the row comes from 1 SET describe_compact_output = 0, describe_include_virtual_columns = 1, describe_include_subcolumns = 1; DESCRIBE TABLE t_describe_options; id UInt64 index column 0 0 @@ -87,6 +89,8 @@ _row_exists UInt8 Persisted mask created by lightweight delete that show wheth _block_number UInt64 Persisted original number of block that was assigned at insert Delta, LZ4 0 1 _block_offset UInt64 Persisted original number of row in block that was assigned at insert Delta, LZ4 0 1 _shard_num UInt32 Deprecated. Use function shardNum instead 0 1 +_database LowCardinality(String) The name of database which the row comes from 0 1 +_table LowCardinality(String) The name of table which the row comes from 0 1 arr.size0 UInt64 1 0 t.a String ZSTD(1) 1 0 t.b UInt64 ZSTD(1) 1 0 @@ -144,6 +148,8 @@ _row_exists UInt8 1 _block_number UInt64 1 _block_offset UInt64 1 _shard_num UInt32 1 +_database LowCardinality(String) 1 +_table LowCardinality(String) 1 SET describe_compact_output = 1, describe_include_virtual_columns = 1, describe_include_subcolumns = 1; DESCRIBE TABLE t_describe_options; id UInt64 0 0 @@ -177,6 +183,8 @@ _row_exists UInt8 0 1 _block_number UInt64 0 1 _block_offset UInt64 0 1 _shard_num UInt32 0 1 +_database LowCardinality(String) 0 1 +_table LowCardinality(String) 0 1 arr.size0 UInt64 1 0 t.a String 1 0 t.b UInt64 1 0 From ce33943b430a9ad512f4942083889dea4decb778 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Thu, 22 Aug 2024 15:50:59 +0200 Subject: [PATCH 202/260] Fix flaky check --- tests/docker_scripts/stateless_runner.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/docker_scripts/stateless_runner.sh b/tests/docker_scripts/stateless_runner.sh index 40a63f74a6b..d8921a04458 100755 --- a/tests/docker_scripts/stateless_runner.sh +++ b/tests/docker_scripts/stateless_runner.sh @@ -339,7 +339,7 @@ export -f run_tests if [ "$NUM_TRIES" -gt "1" ]; then # We don't run tests with Ordinary database in PRs, only in master. # So run new/changed tests with Ordinary at least once in flaky check. - NUM_TRIES=1; USE_DATABASE_ORDINARY=1; run_tests \ + NUM_TRIES=1 USE_DATABASE_ORDINARY=1 run_tests \ | sed 's/All tests have finished/Redacted: a message about tests finish is deleted/' | sed 's/No tests were run/Redacted: a message about no tests run is deleted/' ||: fi From a9e793532ae308767da3bc4da74d9631cd85eb90 Mon Sep 17 00:00:00 2001 From: Sema Checherinda Date: Thu, 22 Aug 2024 16:34:14 +0200 Subject: [PATCH 203/260] fix shutdown for PeriodicLog --- src/Interpreters/PeriodicLog.cpp | 3 ++- src/Interpreters/PeriodicLog.h | 7 ++++--- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/src/Interpreters/PeriodicLog.cpp b/src/Interpreters/PeriodicLog.cpp index 9d2891e11eb..15970ca5b81 100644 --- a/src/Interpreters/PeriodicLog.cpp +++ b/src/Interpreters/PeriodicLog.cpp @@ -1,6 +1,7 @@ #include #include #include +#include "Functions/DateTimeTransforms.h" namespace DB { @@ -27,7 +28,7 @@ template void PeriodicLog::shutdown() { stopCollect(); - this->stopFlushThread(); + Base::shutdown(); } template diff --git a/src/Interpreters/PeriodicLog.h b/src/Interpreters/PeriodicLog.h index 08c3f7eb23f..ceac8088d40 100644 --- a/src/Interpreters/PeriodicLog.h +++ b/src/Interpreters/PeriodicLog.h @@ -17,6 +17,7 @@ template class PeriodicLog : public SystemLog { using SystemLog::SystemLog; + using Base = SystemLog; public: using TimePoint = std::chrono::system_clock::time_point; @@ -24,12 +25,12 @@ public: /// Launches a background thread to collect metrics with interval void startCollect(size_t collect_interval_milliseconds_); - /// Stop background thread - void stopCollect(); - void shutdown() final; protected: + /// Stop background thread + void stopCollect(); + virtual void stepFunction(TimePoint current_time) = 0; private: From 5340ac5fbc7fba75d6a743d345c0f79dc466df0b Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 14:39:19 +0000 Subject: [PATCH 204/260] Update version_date.tsv and changelogs after v24.5.5.41-stable --- docs/changelogs/v24.5.5.41-stable.md | 71 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 2 + 2 files changed, 73 insertions(+) create mode 100644 docs/changelogs/v24.5.5.41-stable.md diff --git a/docs/changelogs/v24.5.5.41-stable.md b/docs/changelogs/v24.5.5.41-stable.md new file mode 100644 index 00000000000..8ba160e31d7 --- /dev/null +++ b/docs/changelogs/v24.5.5.41-stable.md @@ -0,0 +1,71 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.5.5.41-stable (441d4a6ebe3) FIXME as compared to v24.5.4.49-stable (63b760955a0) + +#### Improvement +* Backported in [#66768](https://github.com/ClickHouse/ClickHouse/issues/66768): Make allow_experimental_analyzer be controlled by the initiator for distributed queries. This ensures compatibility and correctness during operations in mixed version clusters. [#65777](https://github.com/ClickHouse/ClickHouse/pull/65777) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#65350](https://github.com/ClickHouse/ClickHouse/issues/65350): Fix possible abort on uncaught exception in ~WriteBufferFromFileDescriptor in StatusFile. [#64206](https://github.com/ClickHouse/ClickHouse/pull/64206) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#65621](https://github.com/ClickHouse/ClickHouse/issues/65621): Fix `Cannot find column` in distributed query with `ARRAY JOIN` by `Nested` column. Fixes [#64755](https://github.com/ClickHouse/ClickHouse/issues/64755). [#64801](https://github.com/ClickHouse/ClickHouse/pull/64801) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#67902](https://github.com/ClickHouse/ClickHouse/issues/67902): Fixing the `Not-ready Set` error after the `PREWHERE` optimization for StorageMerge. [#65057](https://github.com/ClickHouse/ClickHouse/pull/65057) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66884](https://github.com/ClickHouse/ClickHouse/issues/66884): Fix unexpeced size of low cardinality column in function calls. [#65298](https://github.com/ClickHouse/ClickHouse/pull/65298) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#65933](https://github.com/ClickHouse/ClickHouse/issues/65933): For queries that read from `PostgreSQL`, cancel the internal `PostgreSQL` query if the ClickHouse query is finished. Otherwise, `ClickHouse` query cannot be canceled until the internal `PostgreSQL` query is finished. [#65771](https://github.com/ClickHouse/ClickHouse/pull/65771) ([Maksim Kita](https://github.com/kitaisreal)). +* Backported in [#66301](https://github.com/ClickHouse/ClickHouse/issues/66301): Better handling of join conditions involving `IS NULL` checks (for example `ON (a = b AND (a IS NOT NULL) AND (b IS NOT NULL) ) OR ( (a IS NULL) AND (b IS NULL) )` is rewritten to `ON a <=> b`), fix incorrect optimization when condition other then `IS NULL` are present. [#65835](https://github.com/ClickHouse/ClickHouse/pull/65835) ([vdimir](https://github.com/vdimir)). +* Backported in [#66328](https://github.com/ClickHouse/ClickHouse/issues/66328): Add missing settings `input_format_csv_skip_first_lines/input_format_tsv_skip_first_lines/input_format_csv_try_infer_numbers_from_strings/input_format_csv_try_infer_strings_from_quoted_tuples` in schema inference cache because they can change the resulting schema. It prevents from incorrect result of schema inference with these settings changed. [#65980](https://github.com/ClickHouse/ClickHouse/pull/65980) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68252](https://github.com/ClickHouse/ClickHouse/issues/68252): Fixed `Not-ready Set` in some system tables when filtering using subqueries. [#66018](https://github.com/ClickHouse/ClickHouse/pull/66018) ([Michael Kolupaev](https://github.com/al13n321)). +* Backported in [#66155](https://github.com/ClickHouse/ClickHouse/issues/66155): Fixed buffer overflow bug in `unbin`/`unhex` implementation. [#66106](https://github.com/ClickHouse/ClickHouse/pull/66106) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#66454](https://github.com/ClickHouse/ClickHouse/issues/66454): Fixed a bug in ZooKeeper client: a session could get stuck in unusable state after receiving a hardware error from ZooKeeper. For example, this might happen due to "soft memory limit" in ClickHouse Keeper. [#66140](https://github.com/ClickHouse/ClickHouse/pull/66140) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#66226](https://github.com/ClickHouse/ClickHouse/issues/66226): Fix issue in SumIfToCountIfVisitor and signed integers. [#66146](https://github.com/ClickHouse/ClickHouse/pull/66146) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66680](https://github.com/ClickHouse/ClickHouse/issues/66680): Fix handling limit for `system.numbers_mt` when no index can be used. [#66231](https://github.com/ClickHouse/ClickHouse/pull/66231) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66604](https://github.com/ClickHouse/ClickHouse/issues/66604): Fixed how the ClickHouse server detects the maximum number of usable CPU cores as specified by cgroups v2 if the server runs in a container such as Docker. In more detail, containers often run their process in the root cgroup which has an empty name. In that case, ClickHouse ignored the CPU limits set by cgroups v2. [#66237](https://github.com/ClickHouse/ClickHouse/pull/66237) ([filimonov](https://github.com/filimonov)). +* Backported in [#66360](https://github.com/ClickHouse/ClickHouse/issues/66360): Fix the `Not-ready set` error when a subquery with `IN` is used in the constraint. [#66261](https://github.com/ClickHouse/ClickHouse/pull/66261) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#68064](https://github.com/ClickHouse/ClickHouse/issues/68064): Fix boolean literals in query sent to external database (for engines like `PostgreSQL`). [#66282](https://github.com/ClickHouse/ClickHouse/pull/66282) ([vdimir](https://github.com/vdimir)). +* Backported in [#68158](https://github.com/ClickHouse/ClickHouse/issues/68158): Fix cluster() for inter-server secret (preserve initial user as before). [#66364](https://github.com/ClickHouse/ClickHouse/pull/66364) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#66972](https://github.com/ClickHouse/ClickHouse/issues/66972): Fix `Column identifier is already registered` error with `group_by_use_nulls=true` and new analyzer. [#66400](https://github.com/ClickHouse/ClickHouse/pull/66400) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66691](https://github.com/ClickHouse/ClickHouse/issues/66691): Fix the VALID UNTIL clause in the user definition resetting after a restart. Closes [#66405](https://github.com/ClickHouse/ClickHouse/issues/66405). [#66409](https://github.com/ClickHouse/ClickHouse/pull/66409) ([Nikolay Degterinsky](https://github.com/evillique)). +* Backported in [#66969](https://github.com/ClickHouse/ClickHouse/issues/66969): Fix `Cannot find column` error for queries with constant expression in `GROUP BY` key and new analyzer enabled. [#66433](https://github.com/ClickHouse/ClickHouse/pull/66433) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66720](https://github.com/ClickHouse/ClickHouse/issues/66720): Correctly track memory for `Allocator::realloc`. [#66548](https://github.com/ClickHouse/ClickHouse/pull/66548) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#66951](https://github.com/ClickHouse/ClickHouse/issues/66951): Fix an invalid result for queries with `WINDOW`. This could happen when `PARTITION` columns have sparse serialization and window functions are executed in parallel. [#66579](https://github.com/ClickHouse/ClickHouse/pull/66579) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66757](https://github.com/ClickHouse/ClickHouse/issues/66757): Fix `Unknown identifier` and `Column is not under aggregate function` errors for queries with the expression `(column IS NULL).` The bug was triggered by [#65088](https://github.com/ClickHouse/ClickHouse/issues/65088), with the disabled analyzer only. [#66654](https://github.com/ClickHouse/ClickHouse/pull/66654) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66948](https://github.com/ClickHouse/ClickHouse/issues/66948): Fix `Method getResultType is not supported for QUERY query node` error when scalar subquery was used as the first argument of IN (with new analyzer). [#66655](https://github.com/ClickHouse/ClickHouse/pull/66655) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#68115](https://github.com/ClickHouse/ClickHouse/issues/68115): Fix possible PARAMETER_OUT_OF_BOUND error during reading variant subcolumn. [#66659](https://github.com/ClickHouse/ClickHouse/pull/66659) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67633](https://github.com/ClickHouse/ClickHouse/issues/67633): Fix for occasional deadlock in Context::getDDLWorker. [#66843](https://github.com/ClickHouse/ClickHouse/pull/66843) ([Alexander Gololobov](https://github.com/davenger)). +* Backported in [#67481](https://github.com/ClickHouse/ClickHouse/issues/67481): In rare cases ClickHouse could consider parts as broken because of some unexpected projections on disk. Now it's fixed. [#66898](https://github.com/ClickHouse/ClickHouse/pull/66898) ([alesapin](https://github.com/alesapin)). +* Backported in [#67814](https://github.com/ClickHouse/ClickHouse/issues/67814): Only relevant to the experimental Variant data type. Fix crash with Variant + AggregateFunction type. [#67122](https://github.com/ClickHouse/ClickHouse/pull/67122) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67197](https://github.com/ClickHouse/ClickHouse/issues/67197): TRUNCATE DATABASE used to stop replication as if it was a DROP DATABASE query, it's fixed. [#67129](https://github.com/ClickHouse/ClickHouse/pull/67129) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#67379](https://github.com/ClickHouse/ClickHouse/issues/67379): Fix error `Cannot convert column because it is non constant in source stream but must be constant in result.` for a query that reads from the `Merge` table over the `Distriburted` table with one shard. [#67146](https://github.com/ClickHouse/ClickHouse/pull/67146) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#67501](https://github.com/ClickHouse/ClickHouse/issues/67501): Fix crash in DistributedAsyncInsert when connection is empty. [#67219](https://github.com/ClickHouse/ClickHouse/pull/67219) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#67886](https://github.com/ClickHouse/ClickHouse/issues/67886): Correctly parse file name/URI containing `::` if it's not an archive. [#67433](https://github.com/ClickHouse/ClickHouse/pull/67433) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67576](https://github.com/ClickHouse/ClickHouse/issues/67576): Fix execution of nested short-circuit functions. [#67520](https://github.com/ClickHouse/ClickHouse/pull/67520) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67850](https://github.com/ClickHouse/ClickHouse/issues/67850): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#68272](https://github.com/ClickHouse/ClickHouse/issues/68272): Fix inserting into stream like engines (Kafka, RabbitMQ, NATS) through HTTP interface. [#67554](https://github.com/ClickHouse/ClickHouse/pull/67554) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#67807](https://github.com/ClickHouse/ClickHouse/issues/67807): Fix reloading SQL UDFs with UNION. Previously, restarting the server could make UDF invalid. [#67665](https://github.com/ClickHouse/ClickHouse/pull/67665) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67836](https://github.com/ClickHouse/ClickHouse/issues/67836): Fix potential stack overflow in `JSONMergePatch` function. Renamed this function from `jsonMergePatch` to `JSONMergePatch` because the previous name was wrong. The previous name is still kept for compatibility. Improved diagnostic of errors in the function. This closes [#67304](https://github.com/ClickHouse/ClickHouse/issues/67304). [#67756](https://github.com/ClickHouse/ClickHouse/pull/67756) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#67991](https://github.com/ClickHouse/ClickHouse/issues/67991): Validate experimental/suspicious data types in ALTER ADD/MODIFY COLUMN. [#67911](https://github.com/ClickHouse/ClickHouse/pull/67911) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68207](https://github.com/ClickHouse/ClickHouse/issues/68207): Fix wrong `count()` result when there is non-deterministic function in predicate. [#67922](https://github.com/ClickHouse/ClickHouse/pull/67922) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68091](https://github.com/ClickHouse/ClickHouse/issues/68091): Fixed the calculation of the maximum thread soft limit in containerized environments where the usable CPU count is limited. [#67963](https://github.com/ClickHouse/ClickHouse/pull/67963) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#68122](https://github.com/ClickHouse/ClickHouse/issues/68122): Fixed skipping of untouched parts in mutations with new analyzer. Previously with enabled analyzer data in part could be rewritten by mutation even if mutation doesn't affect this part according to predicate. [#68052](https://github.com/ClickHouse/ClickHouse/pull/68052) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68171](https://github.com/ClickHouse/ClickHouse/issues/68171): Removes an incorrect optimization to remove sorting in subqueries that use `OFFSET`. Fixes [#67906](https://github.com/ClickHouse/ClickHouse/issues/67906). [#68099](https://github.com/ClickHouse/ClickHouse/pull/68099) ([Graham Campbell](https://github.com/GrahamCampbell)). +* Backported in [#68337](https://github.com/ClickHouse/ClickHouse/issues/68337): Try fix postgres crash when query is cancelled. [#68288](https://github.com/ClickHouse/ClickHouse/pull/68288) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#68667](https://github.com/ClickHouse/ClickHouse/issues/68667): Fix `LOGICAL_ERROR`s when functions `sipHash64Keyed`, `sipHash128Keyed`, or `sipHash128ReferenceKeyed` are applied to empty arrays or tuples. [#68630](https://github.com/ClickHouse/ClickHouse/pull/68630) ([Robert Schulze](https://github.com/rschu1ze)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#66387](https://github.com/ClickHouse/ClickHouse/issues/66387): Disable broken cases from 02911_join_on_nullsafe_optimization. [#66310](https://github.com/ClickHouse/ClickHouse/pull/66310) ([vdimir](https://github.com/vdimir)). +* Backported in [#66426](https://github.com/ClickHouse/ClickHouse/issues/66426): Ignore subquery for IN in DDLLoadingDependencyVisitor. [#66395](https://github.com/ClickHouse/ClickHouse/pull/66395) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66544](https://github.com/ClickHouse/ClickHouse/issues/66544): Add additional log masking in CI. [#66523](https://github.com/ClickHouse/ClickHouse/pull/66523) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66859](https://github.com/ClickHouse/ClickHouse/issues/66859): Fix data race in S3::ClientCache. [#66644](https://github.com/ClickHouse/ClickHouse/pull/66644) ([Konstantin Morozov](https://github.com/k-morozov)). +* Backported in [#66875](https://github.com/ClickHouse/ClickHouse/issues/66875): Support one more case in JOIN ON ... IS NULL. [#66725](https://github.com/ClickHouse/ClickHouse/pull/66725) ([vdimir](https://github.com/vdimir)). +* Backported in [#67059](https://github.com/ClickHouse/ClickHouse/issues/67059): Increase asio pool size in case the server is tiny. [#66761](https://github.com/ClickHouse/ClickHouse/pull/66761) ([alesapin](https://github.com/alesapin)). +* Backported in [#66945](https://github.com/ClickHouse/ClickHouse/issues/66945): Small fix in realloc memory tracking. [#66820](https://github.com/ClickHouse/ClickHouse/pull/66820) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67252](https://github.com/ClickHouse/ClickHouse/issues/67252): Followup [#66725](https://github.com/ClickHouse/ClickHouse/issues/66725). [#66869](https://github.com/ClickHouse/ClickHouse/pull/66869) ([vdimir](https://github.com/vdimir)). +* Backported in [#67412](https://github.com/ClickHouse/ClickHouse/issues/67412): CI: Fix build results for release branches. [#67402](https://github.com/ClickHouse/ClickHouse/pull/67402) ([Max K.](https://github.com/maxknv)). +* Update version after release. [#67862](https://github.com/ClickHouse/ClickHouse/pull/67862) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Backported in [#68077](https://github.com/ClickHouse/ClickHouse/issues/68077): Add an explicit error for `ALTER MODIFY SQL SECURITY` on non-view tables. [#67953](https://github.com/ClickHouse/ClickHouse/pull/67953) ([pufit](https://github.com/pufit)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 8556375d543..9063d3ef971 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -6,6 +6,7 @@ v24.6.3.95-stable 2024-08-06 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 v24.5.5.78-stable 2024-08-05 +v24.5.5.41-stable 2024-08-22 v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13 @@ -14,6 +15,7 @@ v24.4.4.113-stable 2024-08-02 v24.4.3.25-stable 2024-06-14 v24.4.2.141-stable 2024-06-07 v24.4.1.2088-stable 2024-05-01 +v24.3.9.5-lts 2024-08-22 v24.3.8.13-lts 2024-08-20 v24.3.7.30-lts 2024-08-14 v24.3.6.48-lts 2024-08-02 From 1ea0163dfe6b3278d8a5e8d86c31b3d63d7a3780 Mon Sep 17 00:00:00 2001 From: Salvatore Mesoraca Date: Thu, 22 Aug 2024 16:42:14 +0200 Subject: [PATCH 205/260] Fix issue with maps with arrays as keys --- src/Functions/FunctionsHashing.h | 4 ++-- tests/queries/0_stateless/02534_keyed_siphash.reference | 7 ++++++- tests/queries/0_stateless/02534_keyed_siphash.sql | 5 ++++- 3 files changed, 12 insertions(+), 4 deletions(-) diff --git a/src/Functions/FunctionsHashing.h b/src/Functions/FunctionsHashing.h index 0cf4246fd66..3da0b2cd9be 100644 --- a/src/Functions/FunctionsHashing.h +++ b/src/Functions/FunctionsHashing.h @@ -93,9 +93,9 @@ namespace impl if (is_const) i = 0; assert(key0->size() == key1->size()); - if (offsets != nullptr) + if (offsets != nullptr && i > 0) { - const auto * const begin = offsets->begin(); + const auto * const begin = std::upper_bound(offsets->begin(), offsets->end(), i - 1); const auto * upper = std::upper_bound(begin, offsets->end(), i); if (upper != offsets->end()) i = upper - begin; diff --git a/tests/queries/0_stateless/02534_keyed_siphash.reference b/tests/queries/0_stateless/02534_keyed_siphash.reference index 31c0cae8981..8b147025a05 100644 --- a/tests/queries/0_stateless/02534_keyed_siphash.reference +++ b/tests/queries/0_stateless/02534_keyed_siphash.reference @@ -244,5 +244,10 @@ Test emtpy arrays and maps 0AD04BFD000000000000000000000000 4761183170873013810 0AD04BFD000000000000000000000000 +Test maps with arrays as keys 16734549324845627102 -D675BB3D687973A238AB891DD99C7047 +1D03941D808D04810D2363A6C107D622 +16734549324845627102 +16734549324845627102 +1D03941D808D04810D2363A6C107D622 +1D03941D808D04810D2363A6C107D622 diff --git a/tests/queries/0_stateless/02534_keyed_siphash.sql b/tests/queries/0_stateless/02534_keyed_siphash.sql index b499d8ef02b..ba3c4a9156d 100644 --- a/tests/queries/0_stateless/02534_keyed_siphash.sql +++ b/tests/queries/0_stateless/02534_keyed_siphash.sql @@ -351,5 +351,8 @@ SELECT sipHash64Keyed((1::UInt64, 2::UInt64), []); SELECT hex(sipHash128Keyed((1::UInt64, 2::UInt64), [])); SELECT sipHash64Keyed((1::UInt64, 2::UInt64), mapFromArrays([], [])); SELECT hex(sipHash128Keyed((1::UInt64, 2::UInt64), mapFromArrays([], []))); +SELECT 'Test maps with arrays as keys'; SELECT sipHash64Keyed((1::UInt64, 2::UInt64), map([0], 1, [2], 3)); -SELECT hex(sipHash128Keyed((0::UInt64, 0::UInt64), map([0], 1, [2], 3))); +SELECT hex(sipHash128Keyed((1::UInt64, 2::UInt64), map([0], 1, [2], 3))); +SELECT sipHash64Keyed((materialize(1::UInt64), 2::UInt64), map([0], 1, [2], 3)) FROM numbers(2); +SELECT hex(sipHash128Keyed((materialize(1::UInt64), 2::UInt64), map([0], 1, [2], 3))) FROM numbers(2); From a93d1919804d1c8dc7760f20084ade9a09710a47 Mon Sep 17 00:00:00 2001 From: Salvatore Mesoraca Date: Thu, 22 Aug 2024 16:43:38 +0200 Subject: [PATCH 206/260] Fix typo in test case --- tests/queries/0_stateless/02534_keyed_siphash.reference | 2 +- tests/queries/0_stateless/02534_keyed_siphash.sql | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02534_keyed_siphash.reference b/tests/queries/0_stateless/02534_keyed_siphash.reference index 8b147025a05..a05446a494e 100644 --- a/tests/queries/0_stateless/02534_keyed_siphash.reference +++ b/tests/queries/0_stateless/02534_keyed_siphash.reference @@ -239,7 +239,7 @@ Check bug found fuzzing Test arrays and maps 608E1FF030C9E206185B112C2A25F1A7 ABB65AE97711A2E053E324ED88B1D08B -Test emtpy arrays and maps +Test empty arrays and maps 4761183170873013810 0AD04BFD000000000000000000000000 4761183170873013810 diff --git a/tests/queries/0_stateless/02534_keyed_siphash.sql b/tests/queries/0_stateless/02534_keyed_siphash.sql index ba3c4a9156d..7cfc82512bd 100644 --- a/tests/queries/0_stateless/02534_keyed_siphash.sql +++ b/tests/queries/0_stateless/02534_keyed_siphash.sql @@ -346,7 +346,7 @@ INSERT INTO sipHashKeyed_keys FORMAT VALUES ({'a':'b', 'c':'d'}), ({'e':'f', 'g' SELECT hex(sipHash128ReferenceKeyed((0::UInt64, materialize(0::UInt64)), a)) FROM sipHashKeyed_keys ORDER BY a; DROP TABLE sipHashKeyed_keys; -SELECT 'Test emtpy arrays and maps'; +SELECT 'Test empty arrays and maps'; SELECT sipHash64Keyed((1::UInt64, 2::UInt64), []); SELECT hex(sipHash128Keyed((1::UInt64, 2::UInt64), [])); SELECT sipHash64Keyed((1::UInt64, 2::UInt64), mapFromArrays([], [])); From 0dc18247df3a290b4fb312325ff3b2a44a3f8357 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 15:10:24 +0000 Subject: [PATCH 207/260] Update version_date.tsv and changelogs after v24.6.3.38-stable --- docs/changelogs/v24.6.3.38-stable.md | 83 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 3 + 2 files changed, 86 insertions(+) create mode 100644 docs/changelogs/v24.6.3.38-stable.md diff --git a/docs/changelogs/v24.6.3.38-stable.md b/docs/changelogs/v24.6.3.38-stable.md new file mode 100644 index 00000000000..01d7e26e31f --- /dev/null +++ b/docs/changelogs/v24.6.3.38-stable.md @@ -0,0 +1,83 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.6.3.38-stable (4e33c831589) FIXME as compared to v24.6.2.17-stable (5710a8b5c0c) + +#### Improvement +* Backported in [#66770](https://github.com/ClickHouse/ClickHouse/issues/66770): Make allow_experimental_analyzer be controlled by the initiator for distributed queries. This ensures compatibility and correctness during operations in mixed version clusters. [#65777](https://github.com/ClickHouse/ClickHouse/pull/65777) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#66885](https://github.com/ClickHouse/ClickHouse/issues/66885): Fix unexpeced size of low cardinality column in function calls. [#65298](https://github.com/ClickHouse/ClickHouse/pull/65298) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66303](https://github.com/ClickHouse/ClickHouse/issues/66303): Better handling of join conditions involving `IS NULL` checks (for example `ON (a = b AND (a IS NOT NULL) AND (b IS NOT NULL) ) OR ( (a IS NULL) AND (b IS NULL) )` is rewritten to `ON a <=> b`), fix incorrect optimization when condition other then `IS NULL` are present. [#65835](https://github.com/ClickHouse/ClickHouse/pull/65835) ([vdimir](https://github.com/vdimir)). +* Backported in [#66330](https://github.com/ClickHouse/ClickHouse/issues/66330): Add missing settings `input_format_csv_skip_first_lines/input_format_tsv_skip_first_lines/input_format_csv_try_infer_numbers_from_strings/input_format_csv_try_infer_strings_from_quoted_tuples` in schema inference cache because they can change the resulting schema. It prevents from incorrect result of schema inference with these settings changed. [#65980](https://github.com/ClickHouse/ClickHouse/pull/65980) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#66157](https://github.com/ClickHouse/ClickHouse/issues/66157): Fixed buffer overflow bug in `unbin`/`unhex` implementation. [#66106](https://github.com/ClickHouse/ClickHouse/pull/66106) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#66210](https://github.com/ClickHouse/ClickHouse/issues/66210): Disable the `merge-filters` optimization introduced in [#64760](https://github.com/ClickHouse/ClickHouse/issues/64760). It may cause an exception if optimization merges two filter expressions and does not apply a short-circuit evaluation. [#66126](https://github.com/ClickHouse/ClickHouse/pull/66126) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66456](https://github.com/ClickHouse/ClickHouse/issues/66456): Fixed a bug in ZooKeeper client: a session could get stuck in unusable state after receiving a hardware error from ZooKeeper. For example, this might happen due to "soft memory limit" in ClickHouse Keeper. [#66140](https://github.com/ClickHouse/ClickHouse/pull/66140) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#66228](https://github.com/ClickHouse/ClickHouse/issues/66228): Fix issue in SumIfToCountIfVisitor and signed integers. [#66146](https://github.com/ClickHouse/ClickHouse/pull/66146) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66183](https://github.com/ClickHouse/ClickHouse/issues/66183): Fix rare case with missing data in the result of distributed query, close [#61432](https://github.com/ClickHouse/ClickHouse/issues/61432). [#66174](https://github.com/ClickHouse/ClickHouse/pull/66174) ([vdimir](https://github.com/vdimir)). +* Backported in [#66271](https://github.com/ClickHouse/ClickHouse/issues/66271): Don't throw `TIMEOUT_EXCEEDED` for `none_only_active` mode of `distributed_ddl_output_mode`. [#66218](https://github.com/ClickHouse/ClickHouse/pull/66218) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#66682](https://github.com/ClickHouse/ClickHouse/issues/66682): Fix handling limit for `system.numbers_mt` when no index can be used. [#66231](https://github.com/ClickHouse/ClickHouse/pull/66231) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66587](https://github.com/ClickHouse/ClickHouse/issues/66587): Fixed how the ClickHouse server detects the maximum number of usable CPU cores as specified by cgroups v2 if the server runs in a container such as Docker. In more detail, containers often run their process in the root cgroup which has an empty name. In that case, ClickHouse ignored the CPU limits set by cgroups v2. [#66237](https://github.com/ClickHouse/ClickHouse/pull/66237) ([filimonov](https://github.com/filimonov)). +* Backported in [#66362](https://github.com/ClickHouse/ClickHouse/issues/66362): Fix the `Not-ready set` error when a subquery with `IN` is used in the constraint. [#66261](https://github.com/ClickHouse/ClickHouse/pull/66261) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#68066](https://github.com/ClickHouse/ClickHouse/issues/68066): Fix boolean literals in query sent to external database (for engines like `PostgreSQL`). [#66282](https://github.com/ClickHouse/ClickHouse/pull/66282) ([vdimir](https://github.com/vdimir)). +* Backported in [#68566](https://github.com/ClickHouse/ClickHouse/issues/68566): Fix indexHint function case found by fuzzer. [#66286](https://github.com/ClickHouse/ClickHouse/pull/66286) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68159](https://github.com/ClickHouse/ClickHouse/issues/68159): Fix cluster() for inter-server secret (preserve initial user as before). [#66364](https://github.com/ClickHouse/ClickHouse/pull/66364) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#66613](https://github.com/ClickHouse/ClickHouse/issues/66613): Fix `Column identifier is already registered` error with `group_by_use_nulls=true` and new analyzer. [#66400](https://github.com/ClickHouse/ClickHouse/pull/66400) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66693](https://github.com/ClickHouse/ClickHouse/issues/66693): Fix the VALID UNTIL clause in the user definition resetting after a restart. Closes [#66405](https://github.com/ClickHouse/ClickHouse/issues/66405). [#66409](https://github.com/ClickHouse/ClickHouse/pull/66409) ([Nikolay Degterinsky](https://github.com/evillique)). +* Backported in [#66577](https://github.com/ClickHouse/ClickHouse/issues/66577): Fix `Cannot find column` error for queries with constant expression in `GROUP BY` key and new analyzer enabled. [#66433](https://github.com/ClickHouse/ClickHouse/pull/66433) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66721](https://github.com/ClickHouse/ClickHouse/issues/66721): Correctly track memory for `Allocator::realloc`. [#66548](https://github.com/ClickHouse/ClickHouse/pull/66548) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#66670](https://github.com/ClickHouse/ClickHouse/issues/66670): Fix reading of uninitialized memory when hashing empty tuples. This closes [#66559](https://github.com/ClickHouse/ClickHouse/issues/66559). [#66562](https://github.com/ClickHouse/ClickHouse/pull/66562) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#66952](https://github.com/ClickHouse/ClickHouse/issues/66952): Fix an invalid result for queries with `WINDOW`. This could happen when `PARTITION` columns have sparse serialization and window functions are executed in parallel. [#66579](https://github.com/ClickHouse/ClickHouse/pull/66579) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66956](https://github.com/ClickHouse/ClickHouse/issues/66956): Fix removing named collections in local storage. [#66599](https://github.com/ClickHouse/ClickHouse/pull/66599) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66716](https://github.com/ClickHouse/ClickHouse/issues/66716): Fix removing named collections in local storage. [#66599](https://github.com/ClickHouse/ClickHouse/pull/66599) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66759](https://github.com/ClickHouse/ClickHouse/issues/66759): Fix `Unknown identifier` and `Column is not under aggregate function` errors for queries with the expression `(column IS NULL).` The bug was triggered by [#65088](https://github.com/ClickHouse/ClickHouse/issues/65088), with the disabled analyzer only. [#66654](https://github.com/ClickHouse/ClickHouse/pull/66654) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66751](https://github.com/ClickHouse/ClickHouse/issues/66751): Fix `Method getResultType is not supported for QUERY query node` error when scalar subquery was used as the first argument of IN (with new analyzer). [#66655](https://github.com/ClickHouse/ClickHouse/pull/66655) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#68116](https://github.com/ClickHouse/ClickHouse/issues/68116): Fix possible PARAMETER_OUT_OF_BOUND error during reading variant subcolumn. [#66659](https://github.com/ClickHouse/ClickHouse/pull/66659) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67635](https://github.com/ClickHouse/ClickHouse/issues/67635): Fix for occasional deadlock in Context::getDDLWorker. [#66843](https://github.com/ClickHouse/ClickHouse/pull/66843) ([Alexander Gololobov](https://github.com/davenger)). +* Backported in [#67482](https://github.com/ClickHouse/ClickHouse/issues/67482): In rare cases ClickHouse could consider parts as broken because of some unexpected projections on disk. Now it's fixed. [#66898](https://github.com/ClickHouse/ClickHouse/pull/66898) ([alesapin](https://github.com/alesapin)). +* Backported in [#67816](https://github.com/ClickHouse/ClickHouse/issues/67816): Only relevant to the experimental Variant data type. Fix crash with Variant + AggregateFunction type. [#67122](https://github.com/ClickHouse/ClickHouse/pull/67122) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67199](https://github.com/ClickHouse/ClickHouse/issues/67199): TRUNCATE DATABASE used to stop replication as if it was a DROP DATABASE query, it's fixed. [#67129](https://github.com/ClickHouse/ClickHouse/pull/67129) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#67381](https://github.com/ClickHouse/ClickHouse/issues/67381): Fix error `Cannot convert column because it is non constant in source stream but must be constant in result.` for a query that reads from the `Merge` table over the `Distriburted` table with one shard. [#67146](https://github.com/ClickHouse/ClickHouse/pull/67146) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#67244](https://github.com/ClickHouse/ClickHouse/issues/67244): This closes [#67156](https://github.com/ClickHouse/ClickHouse/issues/67156). This closes [#66447](https://github.com/ClickHouse/ClickHouse/issues/66447). The bug was introduced in https://github.com/ClickHouse/ClickHouse/pull/62907. [#67178](https://github.com/ClickHouse/ClickHouse/pull/67178) ([Maksim Kita](https://github.com/kitaisreal)). +* Backported in [#67503](https://github.com/ClickHouse/ClickHouse/issues/67503): Fix crash in DistributedAsyncInsert when connection is empty. [#67219](https://github.com/ClickHouse/ClickHouse/pull/67219) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#67887](https://github.com/ClickHouse/ClickHouse/issues/67887): Correctly parse file name/URI containing `::` if it's not an archive. [#67433](https://github.com/ClickHouse/ClickHouse/pull/67433) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67578](https://github.com/ClickHouse/ClickHouse/issues/67578): Fix execution of nested short-circuit functions. [#67520](https://github.com/ClickHouse/ClickHouse/pull/67520) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68611](https://github.com/ClickHouse/ClickHouse/issues/68611): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#67852](https://github.com/ClickHouse/ClickHouse/issues/67852): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#68275](https://github.com/ClickHouse/ClickHouse/issues/68275): Fix inserting into stream like engines (Kafka, RabbitMQ, NATS) through HTTP interface. [#67554](https://github.com/ClickHouse/ClickHouse/pull/67554) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#67808](https://github.com/ClickHouse/ClickHouse/issues/67808): Fix reloading SQL UDFs with UNION. Previously, restarting the server could make UDF invalid. [#67665](https://github.com/ClickHouse/ClickHouse/pull/67665) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67838](https://github.com/ClickHouse/ClickHouse/issues/67838): Fix potential stack overflow in `JSONMergePatch` function. Renamed this function from `jsonMergePatch` to `JSONMergePatch` because the previous name was wrong. The previous name is still kept for compatibility. Improved diagnostic of errors in the function. This closes [#67304](https://github.com/ClickHouse/ClickHouse/issues/67304). [#67756](https://github.com/ClickHouse/ClickHouse/pull/67756) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#67993](https://github.com/ClickHouse/ClickHouse/issues/67993): Validate experimental/suspicious data types in ALTER ADD/MODIFY COLUMN. [#67911](https://github.com/ClickHouse/ClickHouse/pull/67911) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68208](https://github.com/ClickHouse/ClickHouse/issues/68208): Fix wrong `count()` result when there is non-deterministic function in predicate. [#67922](https://github.com/ClickHouse/ClickHouse/pull/67922) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68093](https://github.com/ClickHouse/ClickHouse/issues/68093): Fixed the calculation of the maximum thread soft limit in containerized environments where the usable CPU count is limited. [#67963](https://github.com/ClickHouse/ClickHouse/pull/67963) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#68124](https://github.com/ClickHouse/ClickHouse/issues/68124): Fixed skipping of untouched parts in mutations with new analyzer. Previously with enabled analyzer data in part could be rewritten by mutation even if mutation doesn't affect this part according to predicate. [#68052](https://github.com/ClickHouse/ClickHouse/pull/68052) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68221](https://github.com/ClickHouse/ClickHouse/issues/68221): Fixed a NULL pointer dereference, triggered by a specially crafted query, that crashed the server via hopEnd, hopStart, tumbleEnd, and tumbleStart. [#68098](https://github.com/ClickHouse/ClickHouse/pull/68098) ([Salvatore Mesoraca](https://github.com/aiven-sal)). +* Backported in [#68173](https://github.com/ClickHouse/ClickHouse/issues/68173): Removes an incorrect optimization to remove sorting in subqueries that use `OFFSET`. Fixes [#67906](https://github.com/ClickHouse/ClickHouse/issues/67906). [#68099](https://github.com/ClickHouse/ClickHouse/pull/68099) ([Graham Campbell](https://github.com/GrahamCampbell)). +* Backported in [#68339](https://github.com/ClickHouse/ClickHouse/issues/68339): Try fix postgres crash when query is cancelled. [#68288](https://github.com/ClickHouse/ClickHouse/pull/68288) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#68396](https://github.com/ClickHouse/ClickHouse/issues/68396): Fix missing sync replica mode in query `SYSTEM SYNC REPLICA`. [#68326](https://github.com/ClickHouse/ClickHouse/pull/68326) ([Duc Canh Le](https://github.com/canhld94)). +* Backported in [#68668](https://github.com/ClickHouse/ClickHouse/issues/68668): Fix `LOGICAL_ERROR`s when functions `sipHash64Keyed`, `sipHash128Keyed`, or `sipHash128ReferenceKeyed` are applied to empty arrays or tuples. [#68630](https://github.com/ClickHouse/ClickHouse/pull/68630) ([Robert Schulze](https://github.com/rschu1ze)). + +#### NO CL ENTRY + +* NO CL ENTRY: 'Revert "Backport [#66599](https://github.com/ClickHouse/ClickHouse/issues/66599) to 24.6: Fix dropping named collection in local storage"'. [#66922](https://github.com/ClickHouse/ClickHouse/pull/66922) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#66332](https://github.com/ClickHouse/ClickHouse/issues/66332): Do not raise a NOT_IMPLEMENTED error when getting s3 metrics with a multiple disk configuration. [#65403](https://github.com/ClickHouse/ClickHouse/pull/65403) ([Elena Torró](https://github.com/elenatorro)). +* Backported in [#66142](https://github.com/ClickHouse/ClickHouse/issues/66142): Fix flaky test_storage_s3_queue tests. [#66009](https://github.com/ClickHouse/ClickHouse/pull/66009) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#66389](https://github.com/ClickHouse/ClickHouse/issues/66389): Disable broken cases from 02911_join_on_nullsafe_optimization. [#66310](https://github.com/ClickHouse/ClickHouse/pull/66310) ([vdimir](https://github.com/vdimir)). +* Backported in [#66428](https://github.com/ClickHouse/ClickHouse/issues/66428): Ignore subquery for IN in DDLLoadingDependencyVisitor. [#66395](https://github.com/ClickHouse/ClickHouse/pull/66395) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66546](https://github.com/ClickHouse/ClickHouse/issues/66546): Add additional log masking in CI. [#66523](https://github.com/ClickHouse/ClickHouse/pull/66523) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#66861](https://github.com/ClickHouse/ClickHouse/issues/66861): Fix data race in S3::ClientCache. [#66644](https://github.com/ClickHouse/ClickHouse/pull/66644) ([Konstantin Morozov](https://github.com/k-morozov)). +* Backported in [#66877](https://github.com/ClickHouse/ClickHouse/issues/66877): Support one more case in JOIN ON ... IS NULL. [#66725](https://github.com/ClickHouse/ClickHouse/pull/66725) ([vdimir](https://github.com/vdimir)). +* Backported in [#67061](https://github.com/ClickHouse/ClickHouse/issues/67061): Increase asio pool size in case the server is tiny. [#66761](https://github.com/ClickHouse/ClickHouse/pull/66761) ([alesapin](https://github.com/alesapin)). +* Backported in [#66940](https://github.com/ClickHouse/ClickHouse/issues/66940): Small fix in realloc memory tracking. [#66820](https://github.com/ClickHouse/ClickHouse/pull/66820) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67254](https://github.com/ClickHouse/ClickHouse/issues/67254): Followup [#66725](https://github.com/ClickHouse/ClickHouse/issues/66725). [#66869](https://github.com/ClickHouse/ClickHouse/pull/66869) ([vdimir](https://github.com/vdimir)). +* Backported in [#67414](https://github.com/ClickHouse/ClickHouse/issues/67414): CI: Fix build results for release branches. [#67402](https://github.com/ClickHouse/ClickHouse/pull/67402) ([Max K.](https://github.com/maxknv)). +* Update version after release. [#67909](https://github.com/ClickHouse/ClickHouse/pull/67909) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Backported in [#68079](https://github.com/ClickHouse/ClickHouse/issues/68079): Add an explicit error for `ALTER MODIFY SQL SECURITY` on non-view tables. [#67953](https://github.com/ClickHouse/ClickHouse/pull/67953) ([pufit](https://github.com/pufit)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 8556375d543..cc168f58862 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -3,9 +3,11 @@ v24.7.3.42-stable 2024-08-08 v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 v24.6.3.95-stable 2024-08-06 +v24.6.3.38-stable 2024-08-22 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 v24.5.5.78-stable 2024-08-05 +v24.5.5.41-stable 2024-08-22 v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13 @@ -14,6 +16,7 @@ v24.4.4.113-stable 2024-08-02 v24.4.3.25-stable 2024-06-14 v24.4.2.141-stable 2024-06-07 v24.4.1.2088-stable 2024-05-01 +v24.3.9.5-lts 2024-08-22 v24.3.8.13-lts 2024-08-20 v24.3.7.30-lts 2024-08-14 v24.3.6.48-lts 2024-08-02 From 0b9c24f31d548c87deca3334282c14fc78a295ba Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Thu, 15 Aug 2024 12:09:50 +0000 Subject: [PATCH 208/260] write metadata to disk and keeper in the same format --- src/Storages/ColumnsDescription.cpp | 30 +++++--- src/Storages/ColumnsDescription.h | 6 +- .../__init__.py | 0 .../config/enable_keeper.xml | 26 +++++++ .../config/users.xml | 8 +++ .../test.py | 71 +++++++++++++++++++ 6 files changed, 128 insertions(+), 13 deletions(-) create mode 100644 tests/integration/test_aliases_in_default_expr_not_break_table_structure/__init__.py create mode 100644 tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/enable_keeper.xml create mode 100644 tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/users.xml create mode 100644 tests/integration/test_aliases_in_default_expr_not_break_table_structure/test.py diff --git a/src/Storages/ColumnsDescription.cpp b/src/Storages/ColumnsDescription.cpp index 0d724245b49..0212bbd6fff 100644 --- a/src/Storages/ColumnsDescription.cpp +++ b/src/Storages/ColumnsDescription.cpp @@ -113,7 +113,15 @@ bool ColumnDescription::operator==(const ColumnDescription & other) const && ast_to_str(ttl) == ast_to_str(other.ttl); } -void ColumnDescription::writeText(WriteBuffer & buf) const +String formatASTStateAware(IAST & ast, IAST::FormatState & state) +{ + WriteBufferFromOwnString buf; + IAST::FormatSettings settings(buf, true, false); + ast.formatImpl(settings, state, IAST::FormatStateStacked()); + return buf.str(); +} + +void ColumnDescription::writeText(WriteBuffer & buf, IAST::FormatState & state, bool include_comment) const { /// NOTE: Serialization format is insane. @@ -126,20 +134,21 @@ void ColumnDescription::writeText(WriteBuffer & buf) const writeChar('\t', buf); DB::writeText(DB::toString(default_desc.kind), buf); writeChar('\t', buf); - writeEscapedString(queryToString(default_desc.expression), buf); + writeEscapedString(formatASTStateAware(*default_desc.expression, state), buf); } - if (!comment.empty()) + if (!comment.empty() && include_comment) { writeChar('\t', buf); DB::writeText("COMMENT ", buf); - writeEscapedString(queryToString(ASTLiteral(Field(comment))), buf); + auto ast = ASTLiteral(Field(comment)); + writeEscapedString(formatASTStateAware(ast, state), buf); } if (codec) { writeChar('\t', buf); - writeEscapedString(queryToString(codec), buf); + writeEscapedString(formatASTStateAware(*codec, state), buf); } if (!settings.empty()) @@ -150,21 +159,21 @@ void ColumnDescription::writeText(WriteBuffer & buf) const ASTSetQuery ast; ast.is_standalone = false; ast.changes = settings; - writeEscapedString(queryToString(ast), buf); + writeEscapedString(formatASTStateAware(ast, state), buf); DB::writeText(")", buf); } if (!statistics.empty()) { writeChar('\t', buf); - writeEscapedString(queryToString(statistics.getAST()), buf); + writeEscapedString(formatASTStateAware(*statistics.getAST(), state), buf); } if (ttl) { writeChar('\t', buf); DB::writeText("TTL ", buf); - writeEscapedString(queryToString(ttl), buf); + writeEscapedString(formatASTStateAware(*ttl, state), buf); } writeChar('\n', buf); @@ -895,16 +904,17 @@ void ColumnsDescription::resetColumnTTLs() } -String ColumnsDescription::toString() const +String ColumnsDescription::toString(bool include_comments) const { WriteBufferFromOwnString buf; + IAST::FormatState ast_format_state; writeCString("columns format version: 1\n", buf); DB::writeText(columns.size(), buf); writeCString(" columns:\n", buf); for (const ColumnDescription & column : columns) - column.writeText(buf); + column.writeText(buf, ast_format_state, include_comments); return buf.str(); } diff --git a/src/Storages/ColumnsDescription.h b/src/Storages/ColumnsDescription.h index f0760160f0a..c89c26501e8 100644 --- a/src/Storages/ColumnsDescription.h +++ b/src/Storages/ColumnsDescription.h @@ -104,7 +104,7 @@ struct ColumnDescription bool operator==(const ColumnDescription & other) const; bool operator!=(const ColumnDescription & other) const { return !(*this == other); } - void writeText(WriteBuffer & buf) const; + void writeText(WriteBuffer & buf, IAST::FormatState & state, bool include_comment) const; void readText(ReadBuffer & buf); }; @@ -137,7 +137,7 @@ public: /// NOTE Must correspond with Nested::flatten function. void flattenNested(); /// TODO: remove, insert already flattened Nested columns. - bool operator==(const ColumnsDescription & other) const { return columns == other.columns; } + bool operator==(const ColumnsDescription & other) const { return toString(false) == other.toString(false); } bool operator!=(const ColumnsDescription & other) const { return !(*this == other); } auto begin() const { return columns.begin(); } @@ -221,7 +221,7 @@ public: /// Does column has non default specified compression codec bool hasCompressionCodec(const String & column_name) const; - String toString() const; + String toString(bool include_comments = true) const; static ColumnsDescription parse(const String & str); size_t size() const diff --git a/tests/integration/test_aliases_in_default_expr_not_break_table_structure/__init__.py b/tests/integration/test_aliases_in_default_expr_not_break_table_structure/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/enable_keeper.xml b/tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/enable_keeper.xml new file mode 100644 index 00000000000..4ca4f604ec3 --- /dev/null +++ b/tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/enable_keeper.xml @@ -0,0 +1,26 @@ + + + 2181 + 1 + /var/lib/clickhouse/coordination/log + /var/lib/clickhouse/coordination/snapshots + + 20000 + + + + 1 + localhost + 9444 + + + + + + + localhost + 2181 + + 20000 + + \ No newline at end of file diff --git a/tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/users.xml b/tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/users.xml new file mode 100644 index 00000000000..c5de0b6819c --- /dev/null +++ b/tests/integration/test_aliases_in_default_expr_not_break_table_structure/config/users.xml @@ -0,0 +1,8 @@ + + + + default + + + + \ No newline at end of file diff --git a/tests/integration/test_aliases_in_default_expr_not_break_table_structure/test.py b/tests/integration/test_aliases_in_default_expr_not_break_table_structure/test.py new file mode 100644 index 00000000000..e0c15e18c23 --- /dev/null +++ b/tests/integration/test_aliases_in_default_expr_not_break_table_structure/test.py @@ -0,0 +1,71 @@ +import pytest +import random +import string + +from helpers.cluster import ClickHouseCluster + + +cluster = ClickHouseCluster(__file__) +node = cluster.add_instance( + "node", + main_configs=[ + "config/enable_keeper.xml", + "config/users.xml", + ], + stay_alive=True, + with_minio=True, + macros={"shard": 1, "replica": 1}, +) + + +@pytest.fixture(scope="module") +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def randomize_table_name(table_name, random_suffix_length=10): + letters = string.ascii_letters + string.digits + return f"{table_name}_{''.join(random.choice(letters) for _ in range(random_suffix_length))}" + + +@pytest.mark.parametrize("engine", ["ReplicatedMergeTree"]) +def test_aliases_in_default_expr_not_break_table_structure(start_cluster, engine): + """ + Making sure that using aliases in columns' default expressions does not lead to having different columns metadata in ZooKeeper and on disk. + Issue: https://github.com/ClickHouse/clickhouse-private/issues/5150 + """ + + data = '{"event": {"col1-key": "col1-val", "col2-key": "col2-val"}}' + + table_name = randomize_table_name("t") + + node.query( + f""" + DROP TABLE IF EXISTS {table_name}; + CREATE TABLE {table_name} + ( + `data` String, + `col1` String DEFAULT JSONExtractString(JSONExtractString(data, 'event') AS event, 'col1-key'), + `col2` String MATERIALIZED JSONExtractString(JSONExtractString(data, 'event') AS event, 'col2-key') + ) + ENGINE = {engine}('/test/{table_name}', '{{replica}}') + ORDER BY col1 + """ + ) + + node.restart_clickhouse() + + node.query( + f""" + INSERT INTO {table_name} (data) VALUES ('{data}'); + """ + ) + assert node.query(f"SELECT data FROM {table_name}").strip() == data + assert node.query(f"SELECT col1 FROM {table_name}").strip() == "col1-val" + assert node.query(f"SELECT col2 FROM {table_name}").strip() == "col2-val" + + node.query(f"DROP TABLE {table_name}") From 859d2bfe273f571458be6f007761bc8c743d589a Mon Sep 17 00:00:00 2001 From: Sema Checherinda Date: Thu, 22 Aug 2024 17:18:06 +0200 Subject: [PATCH 209/260] move stopFlushThread to SystemLogBase --- src/Common/SystemLogBase.cpp | 19 +++++++++++++++++++ src/Common/SystemLogBase.h | 2 ++ src/Interpreters/PeriodicLog.cpp | 6 +++--- src/Interpreters/PeriodicLog.h | 2 +- src/Interpreters/SystemLog.cpp | 21 +-------------------- src/Interpreters/SystemLog.h | 7 +------ 6 files changed, 27 insertions(+), 30 deletions(-) diff --git a/src/Common/SystemLogBase.cpp b/src/Common/SystemLogBase.cpp index 127c8862a35..45f4eb1c5a6 100644 --- a/src/Common/SystemLogBase.cpp +++ b/src/Common/SystemLogBase.cpp @@ -273,6 +273,25 @@ void SystemLogBase::startup() saving_thread = std::make_unique([this] { savingThreadFunction(); }); } +template +void SystemLogBase::stopFlushThread() +{ + { + std::lock_guard lock(thread_mutex); + + if (!saving_thread || !saving_thread->joinable()) + return; + + if (is_shutdown) + return; + + is_shutdown = true; + queue->shutdown(); + } + + saving_thread->join(); +} + template void SystemLogBase::add(LogElement element) { diff --git a/src/Common/SystemLogBase.h b/src/Common/SystemLogBase.h index 0d7b04d5c57..0942e920a42 100644 --- a/src/Common/SystemLogBase.h +++ b/src/Common/SystemLogBase.h @@ -216,6 +216,8 @@ public: static consteval bool shouldTurnOffLogger() { return false; } protected: + void stopFlushThread() final; + std::shared_ptr> queue; }; } diff --git a/src/Interpreters/PeriodicLog.cpp b/src/Interpreters/PeriodicLog.cpp index 15970ca5b81..1b285aad3ff 100644 --- a/src/Interpreters/PeriodicLog.cpp +++ b/src/Interpreters/PeriodicLog.cpp @@ -11,7 +11,7 @@ void PeriodicLog::startCollect(size_t collect_interval_milliseconds_ { collect_interval_milliseconds = collect_interval_milliseconds_; is_shutdown_metric_thread = false; - flush_thread = std::make_unique([this] { threadFunction(); }); + collecting_thread = std::make_unique([this] { threadFunction(); }); } template @@ -20,8 +20,8 @@ void PeriodicLog::stopCollect() bool old_val = false; if (!is_shutdown_metric_thread.compare_exchange_strong(old_val, true)) return; - if (flush_thread) - flush_thread->join(); + if (collecting_thread) + collecting_thread->join(); } template diff --git a/src/Interpreters/PeriodicLog.h b/src/Interpreters/PeriodicLog.h index ceac8088d40..8254a02434a 100644 --- a/src/Interpreters/PeriodicLog.h +++ b/src/Interpreters/PeriodicLog.h @@ -36,7 +36,7 @@ protected: private: void threadFunction(); - std::unique_ptr flush_thread; + std::unique_ptr collecting_thread; size_t collect_interval_milliseconds; std::atomic is_shutdown_metric_thread{false}; }; diff --git a/src/Interpreters/SystemLog.cpp b/src/Interpreters/SystemLog.cpp index 832c39bfaf8..6a3ec197c6e 100644 --- a/src/Interpreters/SystemLog.cpp +++ b/src/Interpreters/SystemLog.cpp @@ -402,32 +402,13 @@ SystemLog::SystemLog( template void SystemLog::shutdown() { - stopFlushThread(); + Base::stopFlushThread(); auto table = DatabaseCatalog::instance().tryGetTable(table_id, getContext()); if (table) table->flushAndShutdown(); } -template -void SystemLog::stopFlushThread() -{ - { - std::lock_guard lock(thread_mutex); - - if (!saving_thread || !saving_thread->joinable()) - return; - - if (is_shutdown) - return; - - is_shutdown = true; - queue->shutdown(); - } - - saving_thread->join(); -} - template void SystemLog::savingThreadFunction() diff --git a/src/Interpreters/SystemLog.h b/src/Interpreters/SystemLog.h index 9e1af3578bd..31652c1af67 100644 --- a/src/Interpreters/SystemLog.h +++ b/src/Interpreters/SystemLog.h @@ -125,8 +125,6 @@ public: void shutdown() override; - void stopFlushThread() override; - /** Creates new table if it does not exist. * Renames old table if its structure is not suitable. * This cannot be done in constructor to avoid deadlock while renaming a table under locked Context when SystemLog object is created. @@ -136,10 +134,7 @@ public: protected: LoggerPtr log; - using ISystemLog::is_shutdown; - using ISystemLog::saving_thread; - using ISystemLog::thread_mutex; - using Base::queue; + using Base::queue; StoragePtr getStorage() const; From 837f2bba8a136170b6aa8800b6b30849a9310e5f Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 17:23:45 +0200 Subject: [PATCH 210/260] init --- .../0_stateless/00080_show_tables_and_system_tables.sql | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/00080_show_tables_and_system_tables.sql b/tests/queries/0_stateless/00080_show_tables_and_system_tables.sql index a58f9ddb0ac..02e3645ece0 100644 --- a/tests/queries/0_stateless/00080_show_tables_and_system_tables.sql +++ b/tests/queries/0_stateless/00080_show_tables_and_system_tables.sql @@ -6,8 +6,8 @@ CREATE DATABASE {CLICKHOUSE_DATABASE:Identifier}; CREATE TABLE {CLICKHOUSE_DATABASE:Identifier}.A (A UInt8) ENGINE = TinyLog; CREATE TABLE {CLICKHOUSE_DATABASE:Identifier}.B (A UInt8) ENGINE = TinyLog; -SHOW TABLES from {CLICKHOUSE_DATABASE:Identifier}; -SHOW TABLES in system where engine like '%System%' and name in ('numbers', 'one'); +SHOW TABLES FROM {CLICKHOUSE_DATABASE:Identifier}; +SHOW TABLES IN system WHERE engine LIKE '%System%' AND name IN ('numbers', 'one') AND database = 'system'; SELECT name, toUInt32(metadata_modification_time) > 0, engine_full, create_table_query FROM system.tables WHERE database = currentDatabase() ORDER BY name FORMAT TSVRaw; @@ -16,7 +16,7 @@ SELECT name FROM system.tables WHERE is_temporary = 1 AND name = 'test_temporary CREATE TABLE {CLICKHOUSE_DATABASE:Identifier}.test_log(id UInt64) ENGINE = Log; CREATE MATERIALIZED VIEW {CLICKHOUSE_DATABASE:Identifier}.test_materialized ENGINE = Log AS SELECT * FROM {CLICKHOUSE_DATABASE:Identifier}.test_log; -SELECT dependencies_database, dependencies_table FROM system.tables WHERE name = 'test_log' and database=currentDatabase(); +SELECT dependencies_database, dependencies_table FROM system.tables WHERE name = 'test_log' AND database=currentDatabase(); DROP DATABASE {CLICKHOUSE_DATABASE:Identifier}; From 51fbc629c6dff4653e687228b0507947516072bb Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 15:42:17 +0000 Subject: [PATCH 211/260] Update version_date.tsv and changelogs after v24.7.3.47-stable --- docs/changelogs/v24.7.3.47-stable.md | 55 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 4 ++ 2 files changed, 59 insertions(+) create mode 100644 docs/changelogs/v24.7.3.47-stable.md diff --git a/docs/changelogs/v24.7.3.47-stable.md b/docs/changelogs/v24.7.3.47-stable.md new file mode 100644 index 00000000000..e5f23a70fe1 --- /dev/null +++ b/docs/changelogs/v24.7.3.47-stable.md @@ -0,0 +1,55 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.7.3.47-stable (2e50fe27a14) FIXME as compared to v24.7.2.13-stable (6e41f601b2f) + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#68232](https://github.com/ClickHouse/ClickHouse/issues/68232): Fixed `Not-ready Set` in some system tables when filtering using subqueries. [#66018](https://github.com/ClickHouse/ClickHouse/pull/66018) ([Michael Kolupaev](https://github.com/al13n321)). +* Backported in [#67969](https://github.com/ClickHouse/ClickHouse/issues/67969): Fixed reading of subcolumns after `ALTER ADD COLUMN` query. [#66243](https://github.com/ClickHouse/ClickHouse/pull/66243) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68068](https://github.com/ClickHouse/ClickHouse/issues/68068): Fix boolean literals in query sent to external database (for engines like `PostgreSQL`). [#66282](https://github.com/ClickHouse/ClickHouse/pull/66282) ([vdimir](https://github.com/vdimir)). +* Backported in [#67637](https://github.com/ClickHouse/ClickHouse/issues/67637): Fix for occasional deadlock in Context::getDDLWorker. [#66843](https://github.com/ClickHouse/ClickHouse/pull/66843) ([Alexander Gololobov](https://github.com/davenger)). +* Backported in [#67820](https://github.com/ClickHouse/ClickHouse/issues/67820): Fix possible deadlock on query cancel with parallel replicas. [#66905](https://github.com/ClickHouse/ClickHouse/pull/66905) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67818](https://github.com/ClickHouse/ClickHouse/issues/67818): Only relevant to the experimental Variant data type. Fix crash with Variant + AggregateFunction type. [#67122](https://github.com/ClickHouse/ClickHouse/pull/67122) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67766](https://github.com/ClickHouse/ClickHouse/issues/67766): Fix crash of `uniq` and `uniqTheta ` with `tuple()` argument. Closes [#67303](https://github.com/ClickHouse/ClickHouse/issues/67303). [#67306](https://github.com/ClickHouse/ClickHouse/pull/67306) ([flynn](https://github.com/ucasfl)). +* Backported in [#67881](https://github.com/ClickHouse/ClickHouse/issues/67881): Correctly parse file name/URI containing `::` if it's not an archive. [#67433](https://github.com/ClickHouse/ClickHouse/pull/67433) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#68613](https://github.com/ClickHouse/ClickHouse/issues/68613): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#67854](https://github.com/ClickHouse/ClickHouse/issues/67854): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#68278](https://github.com/ClickHouse/ClickHouse/issues/68278): Fix inserting into stream like engines (Kafka, RabbitMQ, NATS) through HTTP interface. [#67554](https://github.com/ClickHouse/ClickHouse/pull/67554) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68040](https://github.com/ClickHouse/ClickHouse/issues/68040): Fix creation of view with recursive CTE. [#67587](https://github.com/ClickHouse/ClickHouse/pull/67587) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)). +* Backported in [#68038](https://github.com/ClickHouse/ClickHouse/issues/68038): Fix crash on `percent_rank`. `percent_rank`'s default frame type is changed to `range unbounded preceding and unbounded following`. `IWindowFunction`'s default window frame is considered and now window functions without window frame definition in sql can be put into different `WindowTransfomer`s properly. [#67661](https://github.com/ClickHouse/ClickHouse/pull/67661) ([lgbo](https://github.com/lgbo-ustc)). +* Backported in [#67713](https://github.com/ClickHouse/ClickHouse/issues/67713): Fix reloading SQL UDFs with UNION. Previously, restarting the server could make UDF invalid. [#67665](https://github.com/ClickHouse/ClickHouse/pull/67665) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67840](https://github.com/ClickHouse/ClickHouse/issues/67840): Fix potential stack overflow in `JSONMergePatch` function. Renamed this function from `jsonMergePatch` to `JSONMergePatch` because the previous name was wrong. The previous name is still kept for compatibility. Improved diagnostic of errors in the function. This closes [#67304](https://github.com/ClickHouse/ClickHouse/issues/67304). [#67756](https://github.com/ClickHouse/ClickHouse/pull/67756) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#67995](https://github.com/ClickHouse/ClickHouse/issues/67995): Validate experimental/suspicious data types in ALTER ADD/MODIFY COLUMN. [#67911](https://github.com/ClickHouse/ClickHouse/pull/67911) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68224](https://github.com/ClickHouse/ClickHouse/issues/68224): Fix wrong `count()` result when there is non-deterministic function in predicate. [#67922](https://github.com/ClickHouse/ClickHouse/pull/67922) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68095](https://github.com/ClickHouse/ClickHouse/issues/68095): Fixed the calculation of the maximum thread soft limit in containerized environments where the usable CPU count is limited. [#67963](https://github.com/ClickHouse/ClickHouse/pull/67963) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#68126](https://github.com/ClickHouse/ClickHouse/issues/68126): Fixed skipping of untouched parts in mutations with new analyzer. Previously with enabled analyzer data in part could be rewritten by mutation even if mutation doesn't affect this part according to predicate. [#68052](https://github.com/ClickHouse/ClickHouse/pull/68052) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68223](https://github.com/ClickHouse/ClickHouse/issues/68223): Fixed a NULL pointer dereference, triggered by a specially crafted query, that crashed the server via hopEnd, hopStart, tumbleEnd, and tumbleStart. [#68098](https://github.com/ClickHouse/ClickHouse/pull/68098) ([Salvatore Mesoraca](https://github.com/aiven-sal)). +* Backported in [#68175](https://github.com/ClickHouse/ClickHouse/issues/68175): Removes an incorrect optimization to remove sorting in subqueries that use `OFFSET`. Fixes [#67906](https://github.com/ClickHouse/ClickHouse/issues/67906). [#68099](https://github.com/ClickHouse/ClickHouse/pull/68099) ([Graham Campbell](https://github.com/GrahamCampbell)). +* Backported in [#68341](https://github.com/ClickHouse/ClickHouse/issues/68341): Try fix postgres crash when query is cancelled. [#68288](https://github.com/ClickHouse/ClickHouse/pull/68288) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#68398](https://github.com/ClickHouse/ClickHouse/issues/68398): Fix missing sync replica mode in query `SYSTEM SYNC REPLICA`. [#68326](https://github.com/ClickHouse/ClickHouse/pull/68326) ([Duc Canh Le](https://github.com/canhld94)). +* Backported in [#68669](https://github.com/ClickHouse/ClickHouse/issues/68669): Fix `LOGICAL_ERROR`s when functions `sipHash64Keyed`, `sipHash128Keyed`, or `sipHash128ReferenceKeyed` are applied to empty arrays or tuples. [#68630](https://github.com/ClickHouse/ClickHouse/pull/68630) ([Robert Schulze](https://github.com/rschu1ze)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#67518](https://github.com/ClickHouse/ClickHouse/issues/67518): Split slow test 03036_dynamic_read_subcolumns. [#66954](https://github.com/ClickHouse/ClickHouse/pull/66954) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67516](https://github.com/ClickHouse/ClickHouse/issues/67516): Split 01508_partition_pruning_long. [#66983](https://github.com/ClickHouse/ClickHouse/pull/66983) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67529](https://github.com/ClickHouse/ClickHouse/issues/67529): Reduce max time of 00763_long_lock_buffer_alter_destination_table. [#67185](https://github.com/ClickHouse/ClickHouse/pull/67185) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#67803](https://github.com/ClickHouse/ClickHouse/issues/67803): Disable some Dynamic tests under sanitizers, rewrite 03202_dynamic_null_map_subcolumn to sql. [#67359](https://github.com/ClickHouse/ClickHouse/pull/67359) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67643](https://github.com/ClickHouse/ClickHouse/issues/67643): [Green CI] Fix potentially flaky test_mask_sensitive_info integration test. [#67506](https://github.com/ClickHouse/ClickHouse/pull/67506) ([Alexey Katsman](https://github.com/alexkats)). +* Backported in [#67609](https://github.com/ClickHouse/ClickHouse/issues/67609): Fix test_zookeeper_config_load_balancing after adding the xdist worker name to the instance. [#67590](https://github.com/ClickHouse/ClickHouse/pull/67590) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#67871](https://github.com/ClickHouse/ClickHouse/issues/67871): Fix 02434_cancel_insert_when_client_dies. [#67600](https://github.com/ClickHouse/ClickHouse/pull/67600) ([vdimir](https://github.com/vdimir)). +* Backported in [#67704](https://github.com/ClickHouse/ClickHouse/issues/67704): Fix 02910_bad_logs_level_in_local in fast tests. [#67603](https://github.com/ClickHouse/ClickHouse/pull/67603) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#67689](https://github.com/ClickHouse/ClickHouse/issues/67689): Fix 01605_adaptive_granularity_block_borders. [#67605](https://github.com/ClickHouse/ClickHouse/pull/67605) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67827](https://github.com/ClickHouse/ClickHouse/issues/67827): Try fix 03143_asof_join_ddb_long. [#67620](https://github.com/ClickHouse/ClickHouse/pull/67620) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67892](https://github.com/ClickHouse/ClickHouse/issues/67892): Revert "Merge pull request [#66510](https://github.com/ClickHouse/ClickHouse/issues/66510) from canhld94/fix_trivial_count_non_deterministic_func". [#67800](https://github.com/ClickHouse/ClickHouse/pull/67800) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68081](https://github.com/ClickHouse/ClickHouse/issues/68081): Add an explicit error for `ALTER MODIFY SQL SECURITY` on non-view tables. [#67953](https://github.com/ClickHouse/ClickHouse/pull/67953) ([pufit](https://github.com/pufit)). +* Update version after release. [#68044](https://github.com/ClickHouse/ClickHouse/pull/68044) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Backported in [#68269](https://github.com/ClickHouse/ClickHouse/issues/68269): [Green CI] Fix test 01903_correct_block_size_prediction_with_default. [#68203](https://github.com/ClickHouse/ClickHouse/pull/68203) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#68432](https://github.com/ClickHouse/ClickHouse/issues/68432): tests: make 01600_parts_states_metrics_long better. [#68265](https://github.com/ClickHouse/ClickHouse/pull/68265) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#68538](https://github.com/ClickHouse/ClickHouse/issues/68538): CI: Native build for package_aarch64. [#68457](https://github.com/ClickHouse/ClickHouse/pull/68457) ([Max K.](https://github.com/maxknv)). +* Backported in [#68555](https://github.com/ClickHouse/ClickHouse/issues/68555): CI: Minor release workflow fix. [#68536](https://github.com/ClickHouse/ClickHouse/pull/68536) ([Max K.](https://github.com/maxknv)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 8556375d543..6ef5ace4ba6 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,11 +1,14 @@ v24.8.1.2684-lts 2024-08-21 +v24.7.3.47-stable 2024-08-22 v24.7.3.42-stable 2024-08-08 v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 v24.6.3.95-stable 2024-08-06 +v24.6.3.38-stable 2024-08-22 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 v24.5.5.78-stable 2024-08-05 +v24.5.5.41-stable 2024-08-22 v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13 @@ -14,6 +17,7 @@ v24.4.4.113-stable 2024-08-02 v24.4.3.25-stable 2024-06-14 v24.4.2.141-stable 2024-06-07 v24.4.1.2088-stable 2024-05-01 +v24.3.9.5-lts 2024-08-22 v24.3.8.13-lts 2024-08-20 v24.3.7.30-lts 2024-08-14 v24.3.6.48-lts 2024-08-02 From 5f61e193401c5fa46db03542cb88ba4188ed00e9 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 17:46:47 +0200 Subject: [PATCH 212/260] small fixes --- docs/ru/getting-started/install.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/ru/getting-started/install.md b/docs/ru/getting-started/install.md index 4a0ec258c64..5bce41ec07a 100644 --- a/docs/ru/getting-started/install.md +++ b/docs/ru/getting-started/install.md @@ -22,7 +22,7 @@ grep -q sse4_2 /proc/cpuinfo && echo "SSE 4.2 supported" || echo "SSE 4.2 not su ### Из deb-пакетов {#install-from-deb-packages} -Яндекс рекомендует использовать официальные скомпилированные `deb`-пакеты для Debian или Ubuntu. Для установки пакетов выполните: +Рекомендуется использовать официальные скомпилированные `deb`-пакеты для Debian или Ubuntu. Для установки пакетов выполните: ``` bash sudo apt-get install -y apt-transport-https ca-certificates curl gnupg @@ -55,7 +55,7 @@ clickhouse-client # or "clickhouse-client --password" if you've set up a passwor ::: ### Из rpm-пакетов {#from-rpm-packages} -Команда ClickHouse в Яндексе рекомендует использовать официальные предкомпилированные `rpm`-пакеты для CentOS, RedHat и всех остальных дистрибутивов Linux, основанных на rpm. +Команда ClickHouse рекомендует использовать официальные предкомпилированные `rpm`-пакеты для CentOS, RedHat и всех остальных дистрибутивов Linux, основанных на rpm. #### Установка официального репозитория @@ -102,7 +102,7 @@ sudo yum install clickhouse-server clickhouse-client ### Из tgz-архивов {#from-tgz-archives} -Команда ClickHouse в Яндексе рекомендует использовать предкомпилированные бинарники из `tgz`-архивов для всех дистрибутивов, где невозможна установка `deb`- и `rpm`- пакетов. +Команда ClickHouse рекомендует использовать предкомпилированные бинарники из `tgz`-архивов для всех дистрибутивов, где невозможна установка `deb`- и `rpm`- пакетов. Интересующую версию архивов можно скачать вручную с помощью `curl` или `wget` из репозитория https://packages.clickhouse.com/tgz/. После этого архивы нужно распаковать и воспользоваться скриптами установки. Пример установки самой свежей версии: From 980b02bfd67defbbdf78165e8225fb754d722d7a Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 17:48:57 +0200 Subject: [PATCH 213/260] fix compatibility with en version --- docs/ru/getting-started/install.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/docs/ru/getting-started/install.md b/docs/ru/getting-started/install.md index 5bce41ec07a..f8a660fbec9 100644 --- a/docs/ru/getting-started/install.md +++ b/docs/ru/getting-started/install.md @@ -31,9 +31,17 @@ curl -fsSL 'https://packages.clickhouse.com/rpm/lts/repodata/repomd.xml.key' | s echo "deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb stable main" | sudo tee \ /etc/apt/sources.list.d/clickhouse.list sudo apt-get update +``` +#### Установка ClickHouse server и client + +```bash sudo apt-get install -y clickhouse-server clickhouse-client +``` +#### Запуск ClickHouse server + +```bash sudo service clickhouse-server start clickhouse-client # or "clickhouse-client --password" if you've set up a password. ``` From 7c3a013d56c1dbd5b72f04f6be61f007004aaefa Mon Sep 17 00:00:00 2001 From: Mark Needham Date: Thu, 22 Aug 2024 16:53:30 +0100 Subject: [PATCH 214/260] Update newjson.md --- docs/en/sql-reference/data-types/newjson.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/sql-reference/data-types/newjson.md b/docs/en/sql-reference/data-types/newjson.md index 9e43216df6c..f7fc7e1498e 100644 --- a/docs/en/sql-reference/data-types/newjson.md +++ b/docs/en/sql-reference/data-types/newjson.md @@ -70,7 +70,7 @@ SELECT '{"a" : {"b" : 42},"c" : [1, 2, 3], "d" : "Hello, World!"}'::JSON as json └────────────────────────────────────────────────┘ ``` -CAST from named `Tuple`, `Map` and `Object('json')` to `JSON` type will be supported later. +CAST from `JSON`, named `Tuple`, `Map` and `Object('json')` to `JSON` type will be supported later. ## Reading JSON paths as subcolumns From 28fbd8a4eff4eafa7db99eb37e38376ffda11763 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 17:56:16 +0200 Subject: [PATCH 215/260] fix stateless tests --- .../queries/0_stateless/03203_hive_style_partitioning.reference | 2 -- tests/queries/0_stateless/03203_hive_style_partitioning.sh | 2 +- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.reference b/tests/queries/0_stateless/03203_hive_style_partitioning.reference index acdadc2510b..a9d856babce 100644 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.reference +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.reference @@ -34,8 +34,6 @@ Cross Elizabeth Array(Int64) LowCardinality(Float64) 101 2070 -4081 -2070 2070 b 1 diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.sh b/tests/queries/0_stateless/03203_hive_style_partitioning.sh index b3d196924af..6734c5f14ad 100755 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.sh +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.sh @@ -32,7 +32,7 @@ SELECT a FROM file('$CURDIR/data_hive/partitioning/a=b/a=b/sample.parquet') LIMI $CLICKHOUSE_LOCAL -n -q """ set use_hive_partitioning = 0; -SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; +SELECT *, non_existing_column FROM file('$CURDIR/data_hive/partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10; """ 2>&1 | grep -c "UNKNOWN_IDENTIFIER" From 9c0e1df1663dd5c56066dd615fc3cafe6408d308 Mon Sep 17 00:00:00 2001 From: alesapin Date: Thu, 22 Aug 2024 17:58:15 +0200 Subject: [PATCH 216/260] Fix flaky test 00989_parallel_parts_loading --- tests/queries/0_stateless/00989_parallel_parts_loading.sql | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/queries/0_stateless/00989_parallel_parts_loading.sql b/tests/queries/0_stateless/00989_parallel_parts_loading.sql index 407e124f137..dc074241ff6 100644 --- a/tests/queries/0_stateless/00989_parallel_parts_loading.sql +++ b/tests/queries/0_stateless/00989_parallel_parts_loading.sql @@ -1,3 +1,5 @@ +-- Tags: no-random-settings, no-random-merge-tree-settings +-- small insert block size can make insert terribly slow, especially with some build like msan DROP TABLE IF EXISTS mt; CREATE TABLE mt (x UInt64) ENGINE = MergeTree ORDER BY x SETTINGS parts_to_delay_insert = 100000, parts_to_throw_insert = 100000; From 0bd8ebf62616ce882b0ebc46945c837a5a91ba44 Mon Sep 17 00:00:00 2001 From: Tyler Hannan Date: Thu, 22 Aug 2024 17:58:56 +0200 Subject: [PATCH 217/260] Update README.md adding community call. resolving recent recordings --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index c9474ef0fc0..9099fd48659 100644 --- a/README.md +++ b/README.md @@ -34,7 +34,7 @@ curl https://clickhouse.com/ | sh Every month we get together with the community (users, contributors, customers, those interested in learning more about ClickHouse) to discuss what is coming in the latest release. If you are interested in sharing what you've built on ClickHouse, let us know. -* [v24.8 Community Call](https://clickhouse.com/company/events/v24-8-community-release-call) - August 20 +* [v24.9 Community Call](https://clickhouse.com/company/events/v24-9-community-release-call) - September 266 ## Upcoming Events @@ -58,7 +58,7 @@ Other upcoming meetups ## Recent Recordings * **Recent Meetup Videos**: [Meetup Playlist](https://www.youtube.com/playlist?list=PL0Z2YDlm0b3iNDUzpY1S3L_iV4nARda_U) Whenever possible recordings of the ClickHouse Community Meetups are edited and presented as individual talks. Current featuring "Modern SQL in 2023", "Fast, Concurrent, and Consistent Asynchronous INSERTS in ClickHouse", and "Full-Text Indices: Design and Experiments" -* **Recording available**: [**v24.4 Release Call**](https://www.youtube.com/watch?v=dtUqgcfOGmE) All the features of 24.4, one convenient video! Watch it now! +* **Recording available**: [**v24.8 LTS Release Call**](https://www.youtube.com/watch?v=AeLmp2jc51k) All the features of 24.8 LTS, one convenient video! Watch it now! ## Interested in joining ClickHouse and making it your full-time job? From 52cdd88eb6d7bbb5d395dd80445655ad47c83c92 Mon Sep 17 00:00:00 2001 From: alesapin Date: Thu, 22 Aug 2024 17:59:10 +0200 Subject: [PATCH 218/260] Better comment --- tests/queries/0_stateless/00989_parallel_parts_loading.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/00989_parallel_parts_loading.sql b/tests/queries/0_stateless/00989_parallel_parts_loading.sql index dc074241ff6..3b73e6a0e3c 100644 --- a/tests/queries/0_stateless/00989_parallel_parts_loading.sql +++ b/tests/queries/0_stateless/00989_parallel_parts_loading.sql @@ -1,5 +1,5 @@ -- Tags: no-random-settings, no-random-merge-tree-settings --- small insert block size can make insert terribly slow, especially with some build like msan +-- small number of insert threads can make insert terribly slow, especially with some build like msan DROP TABLE IF EXISTS mt; CREATE TABLE mt (x UInt64) ENGINE = MergeTree ORDER BY x SETTINGS parts_to_delay_insert = 100000, parts_to_throw_insert = 100000; From e7b89537bf1bb760c6082f04de4668bd1c00f33a Mon Sep 17 00:00:00 2001 From: Sema Checherinda Date: Thu, 22 Aug 2024 18:02:42 +0200 Subject: [PATCH 219/260] fix style --- src/Interpreters/PeriodicLog.cpp | 1 - src/Interpreters/SystemLog.h | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/src/Interpreters/PeriodicLog.cpp b/src/Interpreters/PeriodicLog.cpp index 1b285aad3ff..22bc14856c4 100644 --- a/src/Interpreters/PeriodicLog.cpp +++ b/src/Interpreters/PeriodicLog.cpp @@ -1,7 +1,6 @@ #include #include #include -#include "Functions/DateTimeTransforms.h" namespace DB { diff --git a/src/Interpreters/SystemLog.h b/src/Interpreters/SystemLog.h index 31652c1af67..c03f9370068 100644 --- a/src/Interpreters/SystemLog.h +++ b/src/Interpreters/SystemLog.h @@ -134,7 +134,7 @@ public: protected: LoggerPtr log; - using Base::queue; + using Base::queue; StoragePtr getStorage() const; From 1692360233593e635c5a7797847bdfd8a0ffa33e Mon Sep 17 00:00:00 2001 From: Tyler Hannan Date: Thu, 22 Aug 2024 18:12:38 +0200 Subject: [PATCH 220/260] Update README.md 26 and 266 are different --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 9099fd48659..83a5c05c667 100644 --- a/README.md +++ b/README.md @@ -34,7 +34,7 @@ curl https://clickhouse.com/ | sh Every month we get together with the community (users, contributors, customers, those interested in learning more about ClickHouse) to discuss what is coming in the latest release. If you are interested in sharing what you've built on ClickHouse, let us know. -* [v24.9 Community Call](https://clickhouse.com/company/events/v24-9-community-release-call) - September 266 +* [v24.9 Community Call](https://clickhouse.com/company/events/v24-9-community-release-call) - September 26 ## Upcoming Events From 4264fbc037accedecebcd8122910e4406e92cd58 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 16:16:47 +0000 Subject: [PATCH 221/260] Update version_date.tsv and changelogs after v24.8.2.3-lts --- docker/keeper/Dockerfile | 2 +- docker/server/Dockerfile.alpine | 2 +- docker/server/Dockerfile.ubuntu | 2 +- docs/changelogs/v24.8.2.3-lts.md | 12 ++++++++++++ utils/list-versions/version_date.tsv | 5 +++++ 5 files changed, 20 insertions(+), 3 deletions(-) create mode 100644 docs/changelogs/v24.8.2.3-lts.md diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile index fc93cee5bbc..6ff7ea43374 100644 --- a/docker/keeper/Dockerfile +++ b/docker/keeper/Dockerfile @@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.8.1.2684" +ARG VERSION="24.8.2.3" ARG PACKAGES="clickhouse-keeper" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine index 3ceaf2a08b4..c87885d3b49 100644 --- a/docker/server/Dockerfile.alpine +++ b/docker/server/Dockerfile.alpine @@ -32,7 +32,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.8.1.2684" +ARG VERSION="24.8.2.3" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index 76db997821c..6ccf74823e2 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -28,7 +28,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list ARG REPO_CHANNEL="stable" ARG REPOSITORY="deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb ${REPO_CHANNEL} main" -ARG VERSION="24.8.1.2684" +ARG VERSION="24.8.2.3" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" #docker-official-library:off diff --git a/docs/changelogs/v24.8.2.3-lts.md b/docs/changelogs/v24.8.2.3-lts.md new file mode 100644 index 00000000000..69dfc9961a2 --- /dev/null +++ b/docs/changelogs/v24.8.2.3-lts.md @@ -0,0 +1,12 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.8.2.3-lts (b54f79ed323) FIXME as compared to v24.8.1.2684-lts (161c62fd295) + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#68670](https://github.com/ClickHouse/ClickHouse/issues/68670): Fix `LOGICAL_ERROR`s when functions `sipHash64Keyed`, `sipHash128Keyed`, or `sipHash128ReferenceKeyed` are applied to empty arrays or tuples. [#68630](https://github.com/ClickHouse/ClickHouse/pull/68630) ([Robert Schulze](https://github.com/rschu1ze)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 8556375d543..199c4f822f4 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,11 +1,15 @@ +v24.8.2.3-lts 2024-08-22 v24.8.1.2684-lts 2024-08-21 +v24.7.3.47-stable 2024-08-22 v24.7.3.42-stable 2024-08-08 v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 v24.6.3.95-stable 2024-08-06 +v24.6.3.38-stable 2024-08-22 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 v24.5.5.78-stable 2024-08-05 +v24.5.5.41-stable 2024-08-22 v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13 @@ -14,6 +18,7 @@ v24.4.4.113-stable 2024-08-02 v24.4.3.25-stable 2024-06-14 v24.4.2.141-stable 2024-06-07 v24.4.1.2088-stable 2024-05-01 +v24.3.9.5-lts 2024-08-22 v24.3.8.13-lts 2024-08-20 v24.3.7.30-lts 2024-08-14 v24.3.6.48-lts 2024-08-02 From fa453c3664b18da7a6945e662b881f80fedadc5b Mon Sep 17 00:00:00 2001 From: Max Kainov Date: Thu, 22 Aug 2024 18:13:45 +0200 Subject: [PATCH 222/260] Disable SqlLogic job --- tests/ci/ci_config.py | 7 ++++--- tests/ci/ci_definitions.py | 2 +- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py index 58de25f039f..0885f1d9ec2 100644 --- a/tests/ci/ci_config.py +++ b/tests/ci/ci_config.py @@ -498,9 +498,10 @@ class CI: JobNames.SQLANCER_DEBUG: CommonJobConfigs.SQLLANCER_TEST.with_properties( required_builds=[BuildNames.PACKAGE_DEBUG], ), - JobNames.SQL_LOGIC_TEST: CommonJobConfigs.SQLLOGIC_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_RELEASE], - ), + # TODO: job does not work at all, uncomment and fix + # JobNames.SQL_LOGIC_TEST: CommonJobConfigs.SQLLOGIC_TEST.with_properties( + # required_builds=[BuildNames.PACKAGE_RELEASE], + # ), JobNames.SQLTEST: CommonJobConfigs.SQL_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], ), diff --git a/tests/ci/ci_definitions.py b/tests/ci/ci_definitions.py index 1cdb3f1487e..9d95a19790f 100644 --- a/tests/ci/ci_definitions.py +++ b/tests/ci/ci_definitions.py @@ -204,7 +204,7 @@ class JobNames(metaclass=WithIter): PERFORMANCE_TEST_AMD64 = "Performance Comparison (release)" PERFORMANCE_TEST_ARM64 = "Performance Comparison (aarch64)" - SQL_LOGIC_TEST = "Sqllogic test (release)" + # SQL_LOGIC_TEST = "Sqllogic test (release)" SQLANCER = "SQLancer (release)" SQLANCER_DEBUG = "SQLancer (debug)" From 06c46ee75bcb94fe02ac68df6a4a044145804d76 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 18:56:50 +0200 Subject: [PATCH 223/260] add one more test --- .../0_stateless/03203_hive_style_partitioning.reference | 1 + tests/queries/0_stateless/03203_hive_style_partitioning.sh | 6 ++++++ 2 files changed, 7 insertions(+) diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.reference b/tests/queries/0_stateless/03203_hive_style_partitioning.reference index a9d856babce..0fbc1fb556e 100644 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.reference +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.reference @@ -37,6 +37,7 @@ Array(Int64) LowCardinality(Float64) 2070 b 1 +1 TESTING THE URL PARTITIONING last Elizabeth Frank Elizabeth diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.sh b/tests/queries/0_stateless/03203_hive_style_partitioning.sh index 6734c5f14ad..8ab18f5edfe 100755 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.sh +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.sh @@ -29,6 +29,12 @@ SELECT identifier FROM file('$CURDIR/data_hive/partitioning/identifier=*/email.c SELECT a FROM file('$CURDIR/data_hive/partitioning/a=b/a=b/sample.parquet') LIMIT 1; """ +$CLICKHOUSE_LOCAL -n -q """ +set use_hive_partitioning = 1; + +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth1/sample.parquet') LIMIT 10; +""" 2>&1 | grep -c "INCORRECT_DATA" + $CLICKHOUSE_LOCAL -n -q """ set use_hive_partitioning = 0; From 2a32207e9ee44d52d6fbca7313d847b4eef1c4fb Mon Sep 17 00:00:00 2001 From: Konstantin Bogdanov Date: Thu, 22 Aug 2024 01:20:46 +0200 Subject: [PATCH 224/260] fix: wrap in conditional preprocessor directives --- src/Functions/FunctionsHashingRipe.cpp | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/Functions/FunctionsHashingRipe.cpp b/src/Functions/FunctionsHashingRipe.cpp index 5b06b8ab924..315296b7690 100644 --- a/src/Functions/FunctionsHashingRipe.cpp +++ b/src/Functions/FunctionsHashingRipe.cpp @@ -7,6 +7,7 @@ /// due to excessive resource consumption. namespace DB { +#if USE_SSL REGISTER_FUNCTION(HashingRipe) { factory.registerFunction(FunctionDocumentation{ @@ -18,4 +19,5 @@ REGISTER_FUNCTION(HashingRipe) )"}}, .categories{"Hash"}}); } +#endif } From ef9fbe3006b3023bf47e3a0109490d166071c2aa Mon Sep 17 00:00:00 2001 From: Konstantin Bogdanov Date: Thu, 22 Aug 2024 01:20:46 +0200 Subject: [PATCH 225/260] fix: disable running test in fasttest due to missing OpenSSL --- tests/queries/0_stateless/03222_ripeMD160.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03222_ripeMD160.sql b/tests/queries/0_stateless/03222_ripeMD160.sql index 592f9f830dd..9d418376a20 100644 --- a/tests/queries/0_stateless/03222_ripeMD160.sql +++ b/tests/queries/0_stateless/03222_ripeMD160.sql @@ -1,3 +1,4 @@ +-- Tags: no-fasttest -- Ouput can be verified using: https://emn178.github.io/online-tools/ripemd-160/ SELECT hex(ripeMD160('The quick brown fox jumps over the lazy dog')); From a2ff8e4384f04b2e57d3de93a1ba63971f94794e Mon Sep 17 00:00:00 2001 From: Dergousov Date: Thu, 22 Aug 2024 20:44:52 +0300 Subject: [PATCH 226/260] fix: correct return type inconsistencies in docs --- docs/en/sql-reference/functions/hash-functions.md | 5 ++--- docs/ru/sql-reference/functions/hash-functions.md | 5 +++-- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/docs/en/sql-reference/functions/hash-functions.md b/docs/en/sql-reference/functions/hash-functions.md index 9b7ac8af0e3..cd1c85b5f4c 100644 --- a/docs/en/sql-reference/functions/hash-functions.md +++ b/docs/en/sql-reference/functions/hash-functions.md @@ -689,9 +689,8 @@ SELECT kostikConsistentHash(16045690984833335023, 2); ``` ## ripeMD160 -Produces [RIPEMD-160](https://en.wikipedia.org/wiki/RIPEMD) hash of a string and returns the resulting set of bytes as [FixedString](../data-types/fixedstring.md). - +Produces [RIPEMD-160](https://en.wikipedia.org/wiki/RIPEMD) hash value. **Syntax** @@ -705,7 +704,7 @@ ripeMD160('input') **Returned value** -- A [UInt256](../data-types/int-uint.md) hash value +- A [UInt256]((../data-types/int-uint.md)) hash value where the 160-bit RIPEMD-160 hash is stored in the first 20 bytes. The remaining 12 bytes are zero-padded. **Example** Use the [hex](../functions/encoding-functions.md/#hex) function to represent the result as a hex-encoded string. diff --git a/docs/ru/sql-reference/functions/hash-functions.md b/docs/ru/sql-reference/functions/hash-functions.md index 66d77e66972..b7adcfc1829 100644 --- a/docs/ru/sql-reference/functions/hash-functions.md +++ b/docs/ru/sql-reference/functions/hash-functions.md @@ -125,7 +125,8 @@ SELECT hex(sipHash128('foo', '\x01', 3)); ``` ## ripeMD160 -Генерирует [RIPEMD-160](https://en.wikipedia.org/wiki/RIPEMD) хеш строки и возвращает полученный набор байт в виде [FixedString](../data-types/fixedstring.md). + +Генерирует [RIPEMD-160](https://en.wikipedia.org/wiki/RIPEMD) хеш строки. **Синтаксис** @@ -139,7 +140,7 @@ ripeMD160('input') **Возвращаемое значение** -- [UInt256](../data-types/int-uint.md) хеш-значение +- [UInt256](../data-types/int-uint.md), где 160-битный хеш RIPEMD-160 хранится в первых 20 байтах. Оставшиеся 12 байт заполняются нулями. **Пример** Используйте функцию [hex](../functions/encoding-functions.md#hex) для представления результата в виде строки с шестнадцатеричной кодировкой From f89193fa416cc333f549d72bb8ba453907edc951 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 19:12:19 +0000 Subject: [PATCH 227/260] Update version_date.tsv and changelogs after v24.5.5.41-stable --- utils/list-versions/version_date.tsv | 2 -- 1 file changed, 2 deletions(-) diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 199c4f822f4..0e25f8d3b62 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,11 +1,9 @@ v24.8.2.3-lts 2024-08-22 v24.8.1.2684-lts 2024-08-21 -v24.7.3.47-stable 2024-08-22 v24.7.3.42-stable 2024-08-08 v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 v24.6.3.95-stable 2024-08-06 -v24.6.3.38-stable 2024-08-22 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 v24.5.5.78-stable 2024-08-05 From 4200b3d5cbbfe065073c40f1e122c44189f3554f Mon Sep 17 00:00:00 2001 From: Max Kainov Date: Thu, 22 Aug 2024 14:02:25 +0200 Subject: [PATCH 228/260] CI: Stress test fix --- tests/clickhouse-test | 2 +- tests/docker_scripts/stress_runner.sh | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 4f9380d6f20..ad6173065fe 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -3567,7 +3567,7 @@ if __name__ == "__main__": f"Cannot access the specified directory with queries ({args.queries})", file=sys.stderr, ) - sys.exit(1) + assert False, "No --queries provided" CAPTURE_CLIENT_STACKTRACE = args.capture_client_stacktrace diff --git a/tests/docker_scripts/stress_runner.sh b/tests/docker_scripts/stress_runner.sh index 7666398e10b..039c60c8e4e 100755 --- a/tests/docker_scripts/stress_runner.sh +++ b/tests/docker_scripts/stress_runner.sh @@ -10,8 +10,7 @@ dmesg --clear # shellcheck disable=SC1091 source /setup_export_logs.sh -ln -s /repo/tests/clickhouse-test/ci/stress.py /usr/bin/stress -ln -s /repo/tests/clickhouse-test/clickhouse-test /usr/bin/clickhouse-test +ln -s /repo/tests/clickhouse-test /usr/bin/clickhouse-test # Stress tests and upgrade check uses similar code that was placed # in a separate bash library. See tests/ci/stress_tests.lib @@ -266,6 +265,7 @@ fi start_server +cd /repo/tests/ || exit 1 # clickhouse-test can find queries dir from there python3 /repo/tests/ci/stress.py --hung-check --drop-databases --output-folder /test_output --skip-func-tests "$SKIP_TESTS_OPTION" --global-time-limit 1200 \ && echo -e "Test script exit code$OK" >> /test_output/test_results.tsv \ || echo -e "Test script failed$FAIL script exit code: $?" >> /test_output/test_results.tsv From 69f6ea5083f1686becce4ca9fcf47d1404f2d3ed Mon Sep 17 00:00:00 2001 From: Konstantin Bogdanov Date: Thu, 22 Aug 2024 22:07:02 +0200 Subject: [PATCH 229/260] Update docs/en/sql-reference/functions/hash-functions.md --- docs/en/sql-reference/functions/hash-functions.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/sql-reference/functions/hash-functions.md b/docs/en/sql-reference/functions/hash-functions.md index cd1c85b5f4c..55126640e34 100644 --- a/docs/en/sql-reference/functions/hash-functions.md +++ b/docs/en/sql-reference/functions/hash-functions.md @@ -704,7 +704,7 @@ ripeMD160('input') **Returned value** -- A [UInt256]((../data-types/int-uint.md)) hash value where the 160-bit RIPEMD-160 hash is stored in the first 20 bytes. The remaining 12 bytes are zero-padded. +- A [UInt256](../data-types/int-uint.md) hash value where the 160-bit RIPEMD-160 hash is stored in the first 20 bytes. The remaining 12 bytes are zero-padded. **Example** Use the [hex](../functions/encoding-functions.md/#hex) function to represent the result as a hex-encoded string. From dc862b1411884a462bba8dcf86a474ccbe57e380 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Thu, 22 Aug 2024 23:40:18 +0200 Subject: [PATCH 230/260] fix test --- tests/queries/0_stateless/03203_hive_style_partitioning.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.sh b/tests/queries/0_stateless/03203_hive_style_partitioning.sh index 8ab18f5edfe..60e8a6e9faa 100755 --- a/tests/queries/0_stateless/03203_hive_style_partitioning.sh +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.sh @@ -32,7 +32,7 @@ SELECT a FROM file('$CURDIR/data_hive/partitioning/a=b/a=b/sample.parquet') LIMI $CLICKHOUSE_LOCAL -n -q """ set use_hive_partitioning = 1; -SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth1/sample.parquet') LIMIT 10; +SELECT *, column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth1/sample.parquet') LIMIT 10; """ 2>&1 | grep -c "INCORRECT_DATA" $CLICKHOUSE_LOCAL -n -q """ From 4c790999eb6ad74e3a8f99c072dcc12c956a63d8 Mon Sep 17 00:00:00 2001 From: Max Kainov Date: Fri, 23 Aug 2024 02:18:26 +0200 Subject: [PATCH 231/260] CI: Force package_debug build on release branches --- .github/workflows/release_branches.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/.github/workflows/release_branches.yml b/.github/workflows/release_branches.yml index 82826794ea3..ec119b6ff95 100644 --- a/.github/workflows/release_branches.yml +++ b/.github/workflows/release_branches.yml @@ -130,6 +130,7 @@ jobs: with: build_name: package_debug data: ${{ needs.RunConfig.outputs.data }} + force: true BuilderBinDarwin: needs: [RunConfig, BuildDockers] if: ${{ !failure() && !cancelled() }} From f5739dfe06db8610818fafb5c3a2c33f59fd0a8d Mon Sep 17 00:00:00 2001 From: Max Kainov Date: Fri, 23 Aug 2024 02:51:27 +0200 Subject: [PATCH 232/260] CI: Make job rerun possible if triggered manually --- tests/ci/ci.py | 7 +++++-- tests/ci/ci_utils.py | 5 +++++ 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/tests/ci/ci.py b/tests/ci/ci.py index a9ae078b449..d201b6602f5 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -333,7 +333,10 @@ def _pre_action(s3, job_name, batch, indata, pr_info): CI.JobNames.BUILD_CHECK, ): # we might want to rerun build report job rerun_helper = RerunHelper(commit, _get_ext_check_name(job_name)) - if rerun_helper.is_already_finished_by_status(): + if ( + rerun_helper.is_already_finished_by_status() + and not Utils.is_job_triggered_manually() + ): print("WARNING: Rerunning job with GH status ") status = rerun_helper.get_finished_status() assert status @@ -344,7 +347,7 @@ def _pre_action(s3, job_name, batch, indata, pr_info): skip_status = status.state # ci cache check - if not to_be_skipped and not no_cache: + if not to_be_skipped and not no_cache and not Utils.is_job_triggered_manually(): ci_cache = CiCache(s3, indata["jobs_data"]["digests"]).update() job_config = CI.get_job_config(job_name) if ci_cache.is_successful( diff --git a/tests/ci/ci_utils.py b/tests/ci/ci_utils.py index a4c0977f47c..e8d9e7dc254 100644 --- a/tests/ci/ci_utils.py +++ b/tests/ci/ci_utils.py @@ -18,6 +18,7 @@ class Envs: ) S3_BUILDS_BUCKET = os.getenv("S3_BUILDS_BUCKET", "clickhouse-builds") GITHUB_WORKFLOW = os.getenv("GITHUB_WORKFLOW", "") + GITHUB_ACTOR = os.getenv("GITHUB_ACTOR", "") class WithIter(type): @@ -282,3 +283,7 @@ class Utils: ): res = res.replace(*r) return res + + @staticmethod + def is_job_triggered_manually(): + return "robot" not in Envs.GITHUB_ACTOR From 60e4bcbbf0b1991b42bcab4b83e55be344e8a659 Mon Sep 17 00:00:00 2001 From: Tanya Bragin Date: Thu, 22 Aug 2024 20:45:28 -0700 Subject: [PATCH 233/260] Update README.md Update Raleigh meetup link --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 83a5c05c667..546f08afd3d 100644 --- a/README.md +++ b/README.md @@ -44,7 +44,7 @@ The following upcoming meetups are featuring creator of ClickHouse & CTO, Alexey * [ClickHouse Guangzhou User Group Meetup](https://mp.weixin.qq.com/s/GSvo-7xUoVzCsuUvlLTpCw) - August 25 * [San Francisco Meetup (Cloudflare)](https://www.meetup.com/clickhouse-silicon-valley-meetup-group/events/302540575) - September 5 -* [Raleigh Meetup (Deutsche Bank)](https://www.meetup.com/clickhouse-nc-meetup-group/events/302557230) - September 9 +* [Raleigh Meetup (Deutsche Bank)](https://www.meetup.com/triangletechtalks/events/302723486/) - September 9 * [New York Meetup (Rokt)](https://www.meetup.com/clickhouse-new-york-user-group/events/302575342) - September 10 * [Chicago Meetup (Jump Capital)](https://lu.ma/43tvmrfw) - September 12 From e5380806653f8d391c6e88664b0096c3c51240f5 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 23 Aug 2024 07:09:03 +0000 Subject: [PATCH 234/260] Update version_date.tsv and changelogs after v24.5.6.45-stable --- docs/changelogs/v24.5.6.45-stable.md | 33 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 2 +- 2 files changed, 34 insertions(+), 1 deletion(-) create mode 100644 docs/changelogs/v24.5.6.45-stable.md diff --git a/docs/changelogs/v24.5.6.45-stable.md b/docs/changelogs/v24.5.6.45-stable.md new file mode 100644 index 00000000000..b329ebab27b --- /dev/null +++ b/docs/changelogs/v24.5.6.45-stable.md @@ -0,0 +1,33 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.5.6.45-stable (bdca8604c29) FIXME as compared to v24.5.5.78-stable (0138248cb62) + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#67902](https://github.com/ClickHouse/ClickHouse/issues/67902): Fixing the `Not-ready Set` error after the `PREWHERE` optimization for StorageMerge. [#65057](https://github.com/ClickHouse/ClickHouse/pull/65057) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#68252](https://github.com/ClickHouse/ClickHouse/issues/68252): Fixed `Not-ready Set` in some system tables when filtering using subqueries. [#66018](https://github.com/ClickHouse/ClickHouse/pull/66018) ([Michael Kolupaev](https://github.com/al13n321)). +* Backported in [#68064](https://github.com/ClickHouse/ClickHouse/issues/68064): Fix boolean literals in query sent to external database (for engines like `PostgreSQL`). [#66282](https://github.com/ClickHouse/ClickHouse/pull/66282) ([vdimir](https://github.com/vdimir)). +* Backported in [#68158](https://github.com/ClickHouse/ClickHouse/issues/68158): Fix cluster() for inter-server secret (preserve initial user as before). [#66364](https://github.com/ClickHouse/ClickHouse/pull/66364) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#68115](https://github.com/ClickHouse/ClickHouse/issues/68115): Fix possible PARAMETER_OUT_OF_BOUND error during reading variant subcolumn. [#66659](https://github.com/ClickHouse/ClickHouse/pull/66659) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67886](https://github.com/ClickHouse/ClickHouse/issues/67886): Correctly parse file name/URI containing `::` if it's not an archive. [#67433](https://github.com/ClickHouse/ClickHouse/pull/67433) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#68272](https://github.com/ClickHouse/ClickHouse/issues/68272): Fix inserting into stream like engines (Kafka, RabbitMQ, NATS) through HTTP interface. [#67554](https://github.com/ClickHouse/ClickHouse/pull/67554) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#67807](https://github.com/ClickHouse/ClickHouse/issues/67807): Fix reloading SQL UDFs with UNION. Previously, restarting the server could make UDF invalid. [#67665](https://github.com/ClickHouse/ClickHouse/pull/67665) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67836](https://github.com/ClickHouse/ClickHouse/issues/67836): Fix potential stack overflow in `JSONMergePatch` function. Renamed this function from `jsonMergePatch` to `JSONMergePatch` because the previous name was wrong. The previous name is still kept for compatibility. Improved diagnostic of errors in the function. This closes [#67304](https://github.com/ClickHouse/ClickHouse/issues/67304). [#67756](https://github.com/ClickHouse/ClickHouse/pull/67756) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#67991](https://github.com/ClickHouse/ClickHouse/issues/67991): Validate experimental/suspicious data types in ALTER ADD/MODIFY COLUMN. [#67911](https://github.com/ClickHouse/ClickHouse/pull/67911) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68207](https://github.com/ClickHouse/ClickHouse/issues/68207): Fix wrong `count()` result when there is non-deterministic function in predicate. [#67922](https://github.com/ClickHouse/ClickHouse/pull/67922) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68091](https://github.com/ClickHouse/ClickHouse/issues/68091): Fixed the calculation of the maximum thread soft limit in containerized environments where the usable CPU count is limited. [#67963](https://github.com/ClickHouse/ClickHouse/pull/67963) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#68122](https://github.com/ClickHouse/ClickHouse/issues/68122): Fixed skipping of untouched parts in mutations with new analyzer. Previously with enabled analyzer data in part could be rewritten by mutation even if mutation doesn't affect this part according to predicate. [#68052](https://github.com/ClickHouse/ClickHouse/pull/68052) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68171](https://github.com/ClickHouse/ClickHouse/issues/68171): Removes an incorrect optimization to remove sorting in subqueries that use `OFFSET`. Fixes [#67906](https://github.com/ClickHouse/ClickHouse/issues/67906). [#68099](https://github.com/ClickHouse/ClickHouse/pull/68099) ([Graham Campbell](https://github.com/GrahamCampbell)). +* Backported in [#68337](https://github.com/ClickHouse/ClickHouse/issues/68337): Try fix postgres crash when query is cancelled. [#68288](https://github.com/ClickHouse/ClickHouse/pull/68288) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#68667](https://github.com/ClickHouse/ClickHouse/issues/68667): Fix `LOGICAL_ERROR`s when functions `sipHash64Keyed`, `sipHash128Keyed`, or `sipHash128ReferenceKeyed` are applied to empty arrays or tuples. [#68630](https://github.com/ClickHouse/ClickHouse/pull/68630) ([Robert Schulze](https://github.com/rschu1ze)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Update version after release. [#67862](https://github.com/ClickHouse/ClickHouse/pull/67862) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Backported in [#68077](https://github.com/ClickHouse/ClickHouse/issues/68077): Add an explicit error for `ALTER MODIFY SQL SECURITY` on non-view tables. [#67953](https://github.com/ClickHouse/ClickHouse/pull/67953) ([pufit](https://github.com/pufit)). +* Backported in [#68756](https://github.com/ClickHouse/ClickHouse/issues/68756): To make patch release possible from every commit on release branch, package_debug build is required and must not be skipped. [#68750](https://github.com/ClickHouse/ClickHouse/pull/68750) ([Max K.](https://github.com/maxknv)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 0e25f8d3b62..57a59d7ac49 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -6,8 +6,8 @@ v24.7.1.2915-stable 2024-07-30 v24.6.3.95-stable 2024-08-06 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 +v24.5.6.45-stable 2024-08-23 v24.5.5.78-stable 2024-08-05 -v24.5.5.41-stable 2024-08-22 v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13 From e1a7bd9163bebf0aeab12d8dd46c729f73b068be Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 23 Aug 2024 07:37:32 +0000 Subject: [PATCH 235/260] Update version_date.tsv and changelogs after v24.6.4.42-stable --- docs/changelogs/v24.6.4.42-stable.md | 33 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 3 ++- 2 files changed, 35 insertions(+), 1 deletion(-) create mode 100644 docs/changelogs/v24.6.4.42-stable.md diff --git a/docs/changelogs/v24.6.4.42-stable.md b/docs/changelogs/v24.6.4.42-stable.md new file mode 100644 index 00000000000..29b6ba095af --- /dev/null +++ b/docs/changelogs/v24.6.4.42-stable.md @@ -0,0 +1,33 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.6.4.42-stable (c534bb4b4dd) FIXME as compared to v24.6.3.95-stable (8325c920d11) + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#68066](https://github.com/ClickHouse/ClickHouse/issues/68066): Fix boolean literals in query sent to external database (for engines like `PostgreSQL`). [#66282](https://github.com/ClickHouse/ClickHouse/pull/66282) ([vdimir](https://github.com/vdimir)). +* Backported in [#68566](https://github.com/ClickHouse/ClickHouse/issues/68566): Fix indexHint function case found by fuzzer. [#66286](https://github.com/ClickHouse/ClickHouse/pull/66286) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68159](https://github.com/ClickHouse/ClickHouse/issues/68159): Fix cluster() for inter-server secret (preserve initial user as before). [#66364](https://github.com/ClickHouse/ClickHouse/pull/66364) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#68116](https://github.com/ClickHouse/ClickHouse/issues/68116): Fix possible PARAMETER_OUT_OF_BOUND error during reading variant subcolumn. [#66659](https://github.com/ClickHouse/ClickHouse/pull/66659) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67887](https://github.com/ClickHouse/ClickHouse/issues/67887): Correctly parse file name/URI containing `::` if it's not an archive. [#67433](https://github.com/ClickHouse/ClickHouse/pull/67433) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#68611](https://github.com/ClickHouse/ClickHouse/issues/68611): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#68275](https://github.com/ClickHouse/ClickHouse/issues/68275): Fix inserting into stream like engines (Kafka, RabbitMQ, NATS) through HTTP interface. [#67554](https://github.com/ClickHouse/ClickHouse/pull/67554) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#67993](https://github.com/ClickHouse/ClickHouse/issues/67993): Validate experimental/suspicious data types in ALTER ADD/MODIFY COLUMN. [#67911](https://github.com/ClickHouse/ClickHouse/pull/67911) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68208](https://github.com/ClickHouse/ClickHouse/issues/68208): Fix wrong `count()` result when there is non-deterministic function in predicate. [#67922](https://github.com/ClickHouse/ClickHouse/pull/67922) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68093](https://github.com/ClickHouse/ClickHouse/issues/68093): Fixed the calculation of the maximum thread soft limit in containerized environments where the usable CPU count is limited. [#67963](https://github.com/ClickHouse/ClickHouse/pull/67963) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#68124](https://github.com/ClickHouse/ClickHouse/issues/68124): Fixed skipping of untouched parts in mutations with new analyzer. Previously with enabled analyzer data in part could be rewritten by mutation even if mutation doesn't affect this part according to predicate. [#68052](https://github.com/ClickHouse/ClickHouse/pull/68052) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68221](https://github.com/ClickHouse/ClickHouse/issues/68221): Fixed a NULL pointer dereference, triggered by a specially crafted query, that crashed the server via hopEnd, hopStart, tumbleEnd, and tumbleStart. [#68098](https://github.com/ClickHouse/ClickHouse/pull/68098) ([Salvatore Mesoraca](https://github.com/aiven-sal)). +* Backported in [#68173](https://github.com/ClickHouse/ClickHouse/issues/68173): Removes an incorrect optimization to remove sorting in subqueries that use `OFFSET`. Fixes [#67906](https://github.com/ClickHouse/ClickHouse/issues/67906). [#68099](https://github.com/ClickHouse/ClickHouse/pull/68099) ([Graham Campbell](https://github.com/GrahamCampbell)). +* Backported in [#68339](https://github.com/ClickHouse/ClickHouse/issues/68339): Try fix postgres crash when query is cancelled. [#68288](https://github.com/ClickHouse/ClickHouse/pull/68288) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#68396](https://github.com/ClickHouse/ClickHouse/issues/68396): Fix missing sync replica mode in query `SYSTEM SYNC REPLICA`. [#68326](https://github.com/ClickHouse/ClickHouse/pull/68326) ([Duc Canh Le](https://github.com/canhld94)). +* Backported in [#68668](https://github.com/ClickHouse/ClickHouse/issues/68668): Fix `LOGICAL_ERROR`s when functions `sipHash64Keyed`, `sipHash128Keyed`, or `sipHash128ReferenceKeyed` are applied to empty arrays or tuples. [#68630](https://github.com/ClickHouse/ClickHouse/pull/68630) ([Robert Schulze](https://github.com/rschu1ze)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Update version after release. [#67909](https://github.com/ClickHouse/ClickHouse/pull/67909) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Backported in [#68079](https://github.com/ClickHouse/ClickHouse/issues/68079): Add an explicit error for `ALTER MODIFY SQL SECURITY` on non-view tables. [#67953](https://github.com/ClickHouse/ClickHouse/pull/67953) ([pufit](https://github.com/pufit)). +* Backported in [#68758](https://github.com/ClickHouse/ClickHouse/issues/68758): To make patch release possible from every commit on release branch, package_debug build is required and must not be skipped. [#68750](https://github.com/ClickHouse/ClickHouse/pull/68750) ([Max K.](https://github.com/maxknv)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 0e25f8d3b62..8ce510f110d 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -3,11 +3,12 @@ v24.8.1.2684-lts 2024-08-21 v24.7.3.42-stable 2024-08-08 v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 +v24.6.4.42-stable 2024-08-23 v24.6.3.95-stable 2024-08-06 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 +v24.5.6.45-stable 2024-08-23 v24.5.5.78-stable 2024-08-05 -v24.5.5.41-stable 2024-08-22 v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13 From eec720dab60ea63b033919bbc4c1f6837920a42d Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 23 Aug 2024 08:05:27 +0000 Subject: [PATCH 236/260] Update version_date.tsv and changelogs after v24.7.4.51-stable --- docs/changelogs/v24.7.4.51-stable.md | 36 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 2 ++ 2 files changed, 38 insertions(+) create mode 100644 docs/changelogs/v24.7.4.51-stable.md diff --git a/docs/changelogs/v24.7.4.51-stable.md b/docs/changelogs/v24.7.4.51-stable.md new file mode 100644 index 00000000000..a7cf9790383 --- /dev/null +++ b/docs/changelogs/v24.7.4.51-stable.md @@ -0,0 +1,36 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.7.4.51-stable (70fe2f6fa52) FIXME as compared to v24.7.3.42-stable (63730bc4293) + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#68232](https://github.com/ClickHouse/ClickHouse/issues/68232): Fixed `Not-ready Set` in some system tables when filtering using subqueries. [#66018](https://github.com/ClickHouse/ClickHouse/pull/66018) ([Michael Kolupaev](https://github.com/al13n321)). +* Backported in [#68068](https://github.com/ClickHouse/ClickHouse/issues/68068): Fix boolean literals in query sent to external database (for engines like `PostgreSQL`). [#66282](https://github.com/ClickHouse/ClickHouse/pull/66282) ([vdimir](https://github.com/vdimir)). +* Backported in [#68613](https://github.com/ClickHouse/ClickHouse/issues/68613): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#68278](https://github.com/ClickHouse/ClickHouse/issues/68278): Fix inserting into stream like engines (Kafka, RabbitMQ, NATS) through HTTP interface. [#67554](https://github.com/ClickHouse/ClickHouse/pull/67554) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68040](https://github.com/ClickHouse/ClickHouse/issues/68040): Fix creation of view with recursive CTE. [#67587](https://github.com/ClickHouse/ClickHouse/pull/67587) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)). +* Backported in [#68038](https://github.com/ClickHouse/ClickHouse/issues/68038): Fix crash on `percent_rank`. `percent_rank`'s default frame type is changed to `range unbounded preceding and unbounded following`. `IWindowFunction`'s default window frame is considered and now window functions without window frame definition in sql can be put into different `WindowTransfomer`s properly. [#67661](https://github.com/ClickHouse/ClickHouse/pull/67661) ([lgbo](https://github.com/lgbo-ustc)). +* Backported in [#68224](https://github.com/ClickHouse/ClickHouse/issues/68224): Fix wrong `count()` result when there is non-deterministic function in predicate. [#67922](https://github.com/ClickHouse/ClickHouse/pull/67922) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68095](https://github.com/ClickHouse/ClickHouse/issues/68095): Fixed the calculation of the maximum thread soft limit in containerized environments where the usable CPU count is limited. [#67963](https://github.com/ClickHouse/ClickHouse/pull/67963) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#68126](https://github.com/ClickHouse/ClickHouse/issues/68126): Fixed skipping of untouched parts in mutations with new analyzer. Previously with enabled analyzer data in part could be rewritten by mutation even if mutation doesn't affect this part according to predicate. [#68052](https://github.com/ClickHouse/ClickHouse/pull/68052) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#68223](https://github.com/ClickHouse/ClickHouse/issues/68223): Fixed a NULL pointer dereference, triggered by a specially crafted query, that crashed the server via hopEnd, hopStart, tumbleEnd, and tumbleStart. [#68098](https://github.com/ClickHouse/ClickHouse/pull/68098) ([Salvatore Mesoraca](https://github.com/aiven-sal)). +* Backported in [#68175](https://github.com/ClickHouse/ClickHouse/issues/68175): Removes an incorrect optimization to remove sorting in subqueries that use `OFFSET`. Fixes [#67906](https://github.com/ClickHouse/ClickHouse/issues/67906). [#68099](https://github.com/ClickHouse/ClickHouse/pull/68099) ([Graham Campbell](https://github.com/GrahamCampbell)). +* Backported in [#68341](https://github.com/ClickHouse/ClickHouse/issues/68341): Try fix postgres crash when query is cancelled. [#68288](https://github.com/ClickHouse/ClickHouse/pull/68288) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#68398](https://github.com/ClickHouse/ClickHouse/issues/68398): Fix missing sync replica mode in query `SYSTEM SYNC REPLICA`. [#68326](https://github.com/ClickHouse/ClickHouse/pull/68326) ([Duc Canh Le](https://github.com/canhld94)). +* Backported in [#68669](https://github.com/ClickHouse/ClickHouse/issues/68669): Fix `LOGICAL_ERROR`s when functions `sipHash64Keyed`, `sipHash128Keyed`, or `sipHash128ReferenceKeyed` are applied to empty arrays or tuples. [#68630](https://github.com/ClickHouse/ClickHouse/pull/68630) ([Robert Schulze](https://github.com/rschu1ze)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#67803](https://github.com/ClickHouse/ClickHouse/issues/67803): Disable some Dynamic tests under sanitizers, rewrite 03202_dynamic_null_map_subcolumn to sql. [#67359](https://github.com/ClickHouse/ClickHouse/pull/67359) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#68081](https://github.com/ClickHouse/ClickHouse/issues/68081): Add an explicit error for `ALTER MODIFY SQL SECURITY` on non-view tables. [#67953](https://github.com/ClickHouse/ClickHouse/pull/67953) ([pufit](https://github.com/pufit)). +* Update version after release. [#68044](https://github.com/ClickHouse/ClickHouse/pull/68044) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Backported in [#68269](https://github.com/ClickHouse/ClickHouse/issues/68269): [Green CI] Fix test 01903_correct_block_size_prediction_with_default. [#68203](https://github.com/ClickHouse/ClickHouse/pull/68203) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#68432](https://github.com/ClickHouse/ClickHouse/issues/68432): tests: make 01600_parts_states_metrics_long better. [#68265](https://github.com/ClickHouse/ClickHouse/pull/68265) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#68538](https://github.com/ClickHouse/ClickHouse/issues/68538): CI: Native build for package_aarch64. [#68457](https://github.com/ClickHouse/ClickHouse/pull/68457) ([Max K.](https://github.com/maxknv)). +* Backported in [#68555](https://github.com/ClickHouse/ClickHouse/issues/68555): CI: Minor release workflow fix. [#68536](https://github.com/ClickHouse/ClickHouse/pull/68536) ([Max K.](https://github.com/maxknv)). +* Backported in [#68760](https://github.com/ClickHouse/ClickHouse/issues/68760): To make patch release possible from every commit on release branch, package_debug build is required and must not be skipped. [#68750](https://github.com/ClickHouse/ClickHouse/pull/68750) ([Max K.](https://github.com/maxknv)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 57a59d7ac49..d9674ed2366 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,8 +1,10 @@ v24.8.2.3-lts 2024-08-22 v24.8.1.2684-lts 2024-08-21 +v24.7.4.51-stable 2024-08-23 v24.7.3.42-stable 2024-08-08 v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 +v24.6.4.42-stable 2024-08-23 v24.6.3.95-stable 2024-08-06 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 From 6ba686d2510a2d95ab4332560163d0b4600533a9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?= Date: Fri, 23 Aug 2024 09:20:40 +0000 Subject: [PATCH 237/260] Split test case and reduce number of random runs to reduce time necessary to run the test --- .../01395_limit_more_cases.reference | 1 - .../0_stateless/01395_limit_more_cases.sh | 24 ++++--------------- .../01395_limit_more_cases_random.reference | 1 + .../01395_limit_more_cases_random.sh | 22 +++++++++++++++++ 4 files changed, 28 insertions(+), 20 deletions(-) create mode 100644 tests/queries/0_stateless/01395_limit_more_cases_random.reference create mode 100755 tests/queries/0_stateless/01395_limit_more_cases_random.sh diff --git a/tests/queries/0_stateless/01395_limit_more_cases.reference b/tests/queries/0_stateless/01395_limit_more_cases.reference index c9d0dd73ab8..d68b987ea19 100644 --- a/tests/queries/0_stateless/01395_limit_more_cases.reference +++ b/tests/queries/0_stateless/01395_limit_more_cases.reference @@ -254,4 +254,3 @@ 15 13 0 0 0 0 0 0 15 14 0 0 0 0 0 0 15 15 0 0 0 0 0 0 -0 0 0 diff --git a/tests/queries/0_stateless/01395_limit_more_cases.sh b/tests/queries/0_stateless/01395_limit_more_cases.sh index 177147d2142..9709bd74f26 100755 --- a/tests/queries/0_stateless/01395_limit_more_cases.sh +++ b/tests/queries/0_stateless/01395_limit_more_cases.sh @@ -9,8 +9,11 @@ SIZE=13 for OFFSET in {0..15}; do for LIMIT in {0..15}; do echo "SELECT - $OFFSET, $LIMIT, - count() AS c, min(number) AS first, max(number) AS last, + $OFFSET, + $LIMIT, + count() AS c, + min(number) AS first, + max(number) AS last, throwIf(first != ($OFFSET < $SIZE AND $LIMIT > 0 ? $OFFSET : 0)), throwIf(last != ($OFFSET < $SIZE AND $LIMIT > 0 ? least($SIZE - 1, $OFFSET + $LIMIT - 1) : 0)), throwIf((c != 0 OR first != 0 OR last != 0) AND (c != last - first + 1)) @@ -18,20 +21,3 @@ for OFFSET in {0..15}; do " done done | $CLICKHOUSE_CLIENT -n --max_block_size 5 - -# Randomized test - -ITERATIONS=1000 -for _ in $(seq $ITERATIONS); do - SIZE=$(($RANDOM % 100)) - OFFSET=$(($RANDOM % 111)) - LIMIT=$(($RANDOM % 111)) - - echo "WITH count() AS c, min(number) AS first, max(number) AS last - SELECT - throwIf(first != ($OFFSET < $SIZE AND $LIMIT > 0 ? $OFFSET : 0)), - throwIf(last != ($OFFSET < $SIZE AND $LIMIT > 0 ? least($SIZE - 1, $OFFSET + $LIMIT - 1) : 0)), - throwIf((c != 0 OR first != 0 OR last != 0) AND (c != last - first + 1)) - FROM (SELECT * FROM numbers($SIZE) LIMIT $OFFSET, $LIMIT); - " -done | $CLICKHOUSE_CLIENT -n --max_block_size $(($RANDOM % 20 + 1)) | uniq diff --git a/tests/queries/0_stateless/01395_limit_more_cases_random.reference b/tests/queries/0_stateless/01395_limit_more_cases_random.reference new file mode 100644 index 00000000000..06b63ea6c2f --- /dev/null +++ b/tests/queries/0_stateless/01395_limit_more_cases_random.reference @@ -0,0 +1 @@ +0 0 0 diff --git a/tests/queries/0_stateless/01395_limit_more_cases_random.sh b/tests/queries/0_stateless/01395_limit_more_cases_random.sh new file mode 100755 index 00000000000..c2f6b060aab --- /dev/null +++ b/tests/queries/0_stateless/01395_limit_more_cases_random.sh @@ -0,0 +1,22 @@ +#!/usr/bin/env bash +# Tags: no-fasttest + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +SIZE=13 +ITERATIONS=300 +for _ in $(seq $ITERATIONS); do + SIZE=$(($RANDOM % 100)) + OFFSET=$(($RANDOM % 111)) + LIMIT=$(($RANDOM % 111)) + + echo "WITH count() AS c, min(number) AS first, max(number) AS last + SELECT + throwIf(first != ($OFFSET < $SIZE AND $LIMIT > 0 ? $OFFSET : 0)), + throwIf(last != ($OFFSET < $SIZE AND $LIMIT > 0 ? least($SIZE - 1, $OFFSET + $LIMIT - 1) : 0)), + throwIf((c != 0 OR first != 0 OR last != 0) AND (c != last - first + 1)) + FROM (SELECT * FROM numbers($SIZE) LIMIT $OFFSET, $LIMIT); + " +done | $CLICKHOUSE_CLIENT -n --max_block_size $(($RANDOM % 20 + 1)) | uniq From 8c4329964f597b1eb8139990a41360243f9337f9 Mon Sep 17 00:00:00 2001 From: Maxim Dergousov Date: Fri, 23 Aug 2024 12:50:18 +0300 Subject: [PATCH 238/260] small cosmetic changes in docs --- docs/en/sql-reference/functions/hash-functions.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/docs/en/sql-reference/functions/hash-functions.md b/docs/en/sql-reference/functions/hash-functions.md index 55126640e34..908e288cf59 100644 --- a/docs/en/sql-reference/functions/hash-functions.md +++ b/docs/en/sql-reference/functions/hash-functions.md @@ -695,7 +695,7 @@ Produces [RIPEMD-160](https://en.wikipedia.org/wiki/RIPEMD) hash value. **Syntax** ```sql -ripeMD160('input') +ripeMD160(input) ``` **Parameters** @@ -707,6 +707,7 @@ ripeMD160('input') - A [UInt256](../data-types/int-uint.md) hash value where the 160-bit RIPEMD-160 hash is stored in the first 20 bytes. The remaining 12 bytes are zero-padded. **Example** + Use the [hex](../functions/encoding-functions.md/#hex) function to represent the result as a hex-encoded string. Query: From 8cc5d766b5d70f22646e6dc2832f806736c76311 Mon Sep 17 00:00:00 2001 From: Maxim Dergousov Date: Fri, 23 Aug 2024 12:52:55 +0300 Subject: [PATCH 239/260] small cosmetic changes in docs --- docs/ru/sql-reference/functions/hash-functions.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/docs/ru/sql-reference/functions/hash-functions.md b/docs/ru/sql-reference/functions/hash-functions.md index b7adcfc1829..d7b90b09122 100644 --- a/docs/ru/sql-reference/functions/hash-functions.md +++ b/docs/ru/sql-reference/functions/hash-functions.md @@ -131,7 +131,7 @@ SELECT hex(sipHash128('foo', '\x01', 3)); **Синтаксис** ```sql -ripeMD160('input') +ripeMD160(input) ``` **Аргументы** @@ -143,6 +143,7 @@ ripeMD160('input') - [UInt256](../data-types/int-uint.md), где 160-битный хеш RIPEMD-160 хранится в первых 20 байтах. Оставшиеся 12 байт заполняются нулями. **Пример** + Используйте функцию [hex](../functions/encoding-functions.md#hex) для представления результата в виде строки с шестнадцатеричной кодировкой Запрос: From b0894bffe62722acee2fa5d832ceda9a75754bde Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Fri, 23 Aug 2024 12:01:17 +0200 Subject: [PATCH 240/260] change test file location --- .../sample.parquet | Bin 1 file changed, 0 insertions(+), 0 deletions(-) rename tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/{column0=Elizabeth => column0=Elizabeth1}/sample.parquet (100%) diff --git a/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth1/sample.parquet similarity index 100% rename from tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth/sample.parquet rename to tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth1/sample.parquet From 1165ae756d3a6ca1b9b7c7e9be77f1812390c527 Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 23 Aug 2024 12:16:16 +0000 Subject: [PATCH 241/260] Make dynamic structure selection more consistent --- src/Columns/ColumnDynamic.cpp | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/src/Columns/ColumnDynamic.cpp b/src/Columns/ColumnDynamic.cpp index 1f37add9d2d..efb835b2e17 100644 --- a/src/Columns/ColumnDynamic.cpp +++ b/src/Columns/ColumnDynamic.cpp @@ -1182,12 +1182,13 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source if (!canAddNewVariants(0, all_variants.size())) { /// Create list of variants with their sizes and sort it. - std::vector> variants_with_sizes; + std::vector> variants_with_sizes; variants_with_sizes.reserve(all_variants.size()); for (const auto & variant : all_variants) { - if (variant->getName() != getSharedVariantTypeName()) - variants_with_sizes.emplace_back(total_sizes[variant->getName()], variant); + auto variant_name = variant->getName(); + if (variant_name != getSharedVariantTypeName()) + variants_with_sizes.emplace_back(total_sizes[variant_name], variant_name, variant); } std::sort(variants_with_sizes.begin(), variants_with_sizes.end(), std::greater()); @@ -1196,14 +1197,14 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source result_variants.reserve(max_dynamic_types + 1); /// +1 for shared variant. /// Add shared variant. result_variants.push_back(getSharedVariantDataType()); - for (const auto & [size, variant] : variants_with_sizes) + for (const auto & [size, variant_name, variant_type] : variants_with_sizes) { /// Add variant to the resulting variants list until we reach max_dynamic_types. if (canAddNewVariant(result_variants.size())) - result_variants.push_back(variant); + result_variants.push_back(variant_type); /// Add all remaining variants into shared_variants_statistics until we reach its max size. else if (new_statistics.shared_variants_statistics.size() < Statistics::MAX_SHARED_VARIANT_STATISTICS_SIZE) - new_statistics.shared_variants_statistics[variant->getName()] = size; + new_statistics.shared_variants_statistics[variant_name] = size; else break; } From 6f5210644b95b41cc9d490d4e117c81bd61a1d06 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Fri, 23 Aug 2024 14:43:09 +0200 Subject: [PATCH 242/260] Update src/Columns/ColumnObject.cpp Co-authored-by: Alexander Gololobov --- src/Columns/ColumnObject.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Columns/ColumnObject.cpp b/src/Columns/ColumnObject.cpp index 999c0f6088e..e397b03b69e 100644 --- a/src/Columns/ColumnObject.cpp +++ b/src/Columns/ColumnObject.cpp @@ -1045,7 +1045,7 @@ void ColumnObject::forEachSubcolumnRecursively(DB::IColumn::RecursiveMutableColu bool ColumnObject::structureEquals(const IColumn & rhs) const { - /// 2 Object columns have equal structure if they have the same typed paths and max_dynamic_paths/max_dynamic_types. + /// 2 Object columns have equal structure if they have the same typed paths and global_max_dynamic_paths/max_dynamic_types. const auto * rhs_object = typeid_cast(&rhs); if (!rhs_object || typed_paths.size() != rhs_object->typed_paths.size() || global_max_dynamic_paths != rhs_object->global_max_dynamic_paths || max_dynamic_types != rhs_object->max_dynamic_types) return false; From 2b20b2d4de78acf4fbb08b3f106ebdf410e4587d Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Fri, 23 Aug 2024 15:02:43 +0200 Subject: [PATCH 243/260] Update src/Columns/ColumnDynamic.cpp Co-authored-by: Dmitry Novik --- src/Columns/ColumnDynamic.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Columns/ColumnDynamic.cpp b/src/Columns/ColumnDynamic.cpp index efb835b2e17..ef6cd7dcea2 100644 --- a/src/Columns/ColumnDynamic.cpp +++ b/src/Columns/ColumnDynamic.cpp @@ -1181,7 +1181,7 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source /// Check if the number of all dynamic types exceeds the limit. if (!canAddNewVariants(0, all_variants.size())) { - /// Create list of variants with their sizes and sort it. + /// Create a list of variants with their sizes and names and then sort it. std::vector> variants_with_sizes; variants_with_sizes.reserve(all_variants.size()); for (const auto & variant : all_variants) From 5d6b861ff055de0d04e0c574bf2ebb1e51215ace Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 23 Aug 2024 13:49:36 +0000 Subject: [PATCH 244/260] Fix index with limit=0 --- src/Columns/ColumnVariant.cpp | 2 +- .../03228_variant_permutation_issue.reference | 4 ++++ .../03228_variant_permutation_issue.sql | 15 +++++++++++++++ 3 files changed, 20 insertions(+), 1 deletion(-) diff --git a/src/Columns/ColumnVariant.cpp b/src/Columns/ColumnVariant.cpp index 2fea3eca123..c6511695f5c 100644 --- a/src/Columns/ColumnVariant.cpp +++ b/src/Columns/ColumnVariant.cpp @@ -953,7 +953,7 @@ ColumnPtr ColumnVariant::index(const IColumn & indexes, size_t limit) const { /// If we have only NULLs, index will take no effect, just return resized column. if (hasOnlyNulls()) - return cloneResized(limit); + return cloneResized(limit == 0 ? indexes.size(): limit); /// Optimization when we have only one non empty variant and no NULLs. /// In this case local_discriminators column is filled with identical values and offsets column diff --git a/tests/queries/0_stateless/03228_variant_permutation_issue.reference b/tests/queries/0_stateless/03228_variant_permutation_issue.reference index 10688253e15..be9cdedaf07 100644 --- a/tests/queries/0_stateless/03228_variant_permutation_issue.reference +++ b/tests/queries/0_stateless/03228_variant_permutation_issue.reference @@ -2,3 +2,7 @@ 3 {"foo2":"bar"} 1 2 {"foo2":"baz"} 2 3 {"foo2":"bar"} 1 +2 {"foo2":"bar"} 1 +3 {"foo2":"bar"} 1 +2 {"foo2":"baz"} 2 +3 {"foo2":"bar"} 1 diff --git a/tests/queries/0_stateless/03228_variant_permutation_issue.sql b/tests/queries/0_stateless/03228_variant_permutation_issue.sql index 088361d6430..81eb2ed69af 100644 --- a/tests/queries/0_stateless/03228_variant_permutation_issue.sql +++ b/tests/queries/0_stateless/03228_variant_permutation_issue.sql @@ -16,3 +16,18 @@ SELECT * FROM test_new_json_type FINAL WHERE data.foo2 is not null ORDER BY id; DROP TABLE test_new_json_type; +CREATE TABLE test_new_json_type(id Nullable(UInt32), data JSON, version UInt64) ENGINE=ReplacingMergeTree(version) ORDER BY id settings allow_nullable_key=1; +INSERT INTO test_new_json_type format JSONEachRow +{"id":1,"data":{"foo1":"bar"},"version":1} +{"id":2,"data":{"foo2":"bar"},"version":1} +{"id":3,"data":{"foo2":"bar"},"version":1} +; + +SELECT * FROM test_new_json_type FINAL WHERE data.foo2 is not null ORDER BY id; + +INSERT INTO test_new_json_type SELECT id, '{"foo2":"baz"}' AS _data, version+1 AS _version FROM test_new_json_type where id=2; + +SELECT * FROM test_new_json_type FINAL PREWHERE data.foo2 IS NOT NULL WHERE data.foo2 IS NOT NULL ORDER BY id ASC NULLS FIRST; + +DROP TABLE test_new_json_type; + From 61fa4e7a476b3db31c22030470341b131501f3b6 Mon Sep 17 00:00:00 2001 From: Han Fei Date: Fri, 23 Aug 2024 16:38:48 +0200 Subject: [PATCH 245/260] fix logical err of modify statistics --- src/Storages/AlterCommands.cpp | 2 +- src/Storages/StatisticsDescription.cpp | 6 ++++-- .../integration/test_manipulate_statistics/test.py | 14 ++++++++++++-- 3 files changed, 17 insertions(+), 5 deletions(-) diff --git a/src/Storages/AlterCommands.cpp b/src/Storages/AlterCommands.cpp index d5780e32db3..67b18217767 100644 --- a/src/Storages/AlterCommands.cpp +++ b/src/Storages/AlterCommands.cpp @@ -734,7 +734,7 @@ void AlterCommand::apply(StorageInMemoryMetadata & metadata, ContextPtr context) { if (!metadata.columns.has(statistics_column_name)) { - throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Cannot add statistics for column {}: this column is not found", statistics_column_name); + throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Cannot modify statistics for column {}: this column is not found", statistics_column_name); } } diff --git a/src/Storages/StatisticsDescription.cpp b/src/Storages/StatisticsDescription.cpp index 63c849e3806..acf600dd6f7 100644 --- a/src/Storages/StatisticsDescription.cpp +++ b/src/Storages/StatisticsDescription.cpp @@ -6,7 +6,6 @@ #include #include #include -#include #include @@ -115,8 +114,11 @@ void ColumnStatisticsDescription::merge(const ColumnStatisticsDescription & othe void ColumnStatisticsDescription::assign(const ColumnStatisticsDescription & other) { + /// If the statistics is empty, it's possible that we have not assign a column_name. + if (empty() && column_name == "") + column_name = other.column_name; if (other.column_name != column_name) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot assign statistics from column {} to {}", column_name, other.column_name); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot assign statistics from column {} to {}", other.column_name, column_name); types_to_desc = other.types_to_desc; data_type = other.data_type; diff --git a/tests/integration/test_manipulate_statistics/test.py b/tests/integration/test_manipulate_statistics/test.py index 2541c9b946f..ab5559e18fa 100644 --- a/tests/integration/test_manipulate_statistics/test.py +++ b/tests/integration/test_manipulate_statistics/test.py @@ -6,11 +6,13 @@ from helpers.cluster import ClickHouseCluster cluster = ClickHouseCluster(__file__) node1 = cluster.add_instance( - "node1", user_configs=["config/config.xml"], with_zookeeper=True + "node1", user_configs=["config/config.xml"], with_zookeeper=True, + macros={"replica": "a", "shard": "shard1"} ) node2 = cluster.add_instance( - "node2", user_configs=["config/config.xml"], with_zookeeper=True + "node2", user_configs=["config/config.xml"], with_zookeeper=True, + macros={"replica": "b", "shard": "shard1"} ) @@ -183,3 +185,11 @@ def test_replicated_table_ddl(started_cluster): ) check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_3", "a", True) check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_3", "b", True) + + +def test_replicated_db(started_cluster): + node1.query("CREATE DATABASE test ENGINE = Replicated('/test/shared_stats', '{shard}', '{replica}')") + node2.query("CREATE DATABASE test ENGINE = Replicated('/test/shared_stats', '{shard}', '{replica}')") + node1.query("CREATE TABLE test.test_stats (a Int64, b Int64) ENGINE = ReplicatedMergeTree() ORDER BY()") + node2.query("ALTER TABLE test.test_stats MODIFY COLUMN b Float64") + node2.query("ALTER TABLE test.test_stats MODIFY STATISTICS b TYPE tdigest") From 7aabd7d2fd4a03ddea5ef311cf89b2eb7520674c Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 23 Aug 2024 15:11:51 +0000 Subject: [PATCH 246/260] Fix resolving dynamic subcolumns from subqueries in analyzer --- src/Analyzer/Resolve/IdentifierResolver.cpp | 2 +- src/Analyzer/Resolve/QueryAnalyzer.cpp | 3 +++ src/Analyzer/Resolve/TableExpressionData.h | 1 + .../03228_dynamic_subcolumns_from_subquery.reference | 4 ++++ .../03228_dynamic_subcolumns_from_subquery.sql | 9 +++++++++ 5 files changed, 18 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.reference create mode 100644 tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.sql diff --git a/src/Analyzer/Resolve/IdentifierResolver.cpp b/src/Analyzer/Resolve/IdentifierResolver.cpp index 14d4acc7c9b..80e7d1e4445 100644 --- a/src/Analyzer/Resolve/IdentifierResolver.cpp +++ b/src/Analyzer/Resolve/IdentifierResolver.cpp @@ -692,7 +692,7 @@ QueryTreeNodePtr IdentifierResolver::tryResolveIdentifierFromStorage( result_column_node = it->second; } /// Check if it's a dynamic subcolumn - else + else if (table_expression_data.supports_subcolumns) { auto [column_name, dynamic_subcolumn_name] = Nested::splitName(identifier_full_name); auto jt = table_expression_data.column_name_to_column_node.find(column_name); diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 004da5ed341..a18c2901a58 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -4379,7 +4379,10 @@ void QueryAnalyzer::initializeTableExpressionData(const QueryTreeNodePtr & table auto get_column_options = GetColumnsOptions(GetColumnsOptions::All).withExtendedObjects().withVirtuals(); if (storage_snapshot->storage.supportsSubcolumns()) + { get_column_options.withSubcolumns(); + table_expression_data.supports_subcolumns = true; + } auto column_names_and_types = storage_snapshot->getColumns(get_column_options); table_expression_data.column_names_and_types = NamesAndTypes(column_names_and_types.begin(), column_names_and_types.end()); diff --git a/src/Analyzer/Resolve/TableExpressionData.h b/src/Analyzer/Resolve/TableExpressionData.h index 18cbfa32366..6770672d0c2 100644 --- a/src/Analyzer/Resolve/TableExpressionData.h +++ b/src/Analyzer/Resolve/TableExpressionData.h @@ -36,6 +36,7 @@ struct AnalysisTableExpressionData std::string database_name; std::string table_name; bool should_qualify_columns = true; + bool supports_subcolumns = false; NamesAndTypes column_names_and_types; ColumnNameToColumnNodeMap column_name_to_column_node; std::unordered_set subcolumn_names; /// Subset columns that are subcolumns of other columns diff --git a/tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.reference b/tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.reference new file mode 100644 index 00000000000..153ad78f694 --- /dev/null +++ b/tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.reference @@ -0,0 +1,4 @@ +str +42 +42 +42 diff --git a/tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.sql b/tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.sql new file mode 100644 index 00000000000..a10b0cb2809 --- /dev/null +++ b/tests/queries/0_stateless/03228_dynamic_subcolumns_from_subquery.sql @@ -0,0 +1,9 @@ +set allow_experimental_dynamic_type=1; +set allow_experimental_json_type=1; +set allow_experimental_analyzer=1; + +select d.String from (select 'str'::Dynamic as d); +select json.a from (select '{"a" : 42}'::JSON as json); +select json.a from (select '{"a" : 42}'::JSON(a UInt32) as json); +select json.a.:Int64 from (select materialize('{"a" : 42}')::JSON as json); + From 80504e7b9b52fec79a89e2fff5881ca397022107 Mon Sep 17 00:00:00 2001 From: Anton Popov Date: Fri, 23 Aug 2024 19:07:25 +0000 Subject: [PATCH 247/260] fix test 03228_virtual_column_merge_dist --- .../queries/0_stateless/03228_virtual_column_merge_dist.sql | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03228_virtual_column_merge_dist.sql b/tests/queries/0_stateless/03228_virtual_column_merge_dist.sql index caf00a2e407..e58c7f38d3b 100644 --- a/tests/queries/0_stateless/03228_virtual_column_merge_dist.sql +++ b/tests/queries/0_stateless/03228_virtual_column_merge_dist.sql @@ -1,3 +1,6 @@ +-- There is a bug in old analyzer with currentDatabase() and distributed query. +SET enable_analyzer = 1; + DROP TABLE IF EXISTS t_local_1; DROP TABLE IF EXISTS t_local_2; DROP TABLE IF EXISTS t_merge; @@ -10,7 +13,7 @@ INSERT INTO t_local_1 VALUES (1); INSERT INTO t_local_2 VALUES (2); CREATE TABLE t_merge AS t_local_1 ENGINE = Merge(currentDatabase(), '^(t_local_1|t_local_2)$'); -CREATE TABLE t_distr AS t_local_1 engine=Distributed('test_shard_localhost', currentDatabase(), t_merge, rand()); +CREATE TABLE t_distr AS t_local_1 ENGINE = Distributed('test_shard_localhost', currentDatabase(), t_merge, rand()); SELECT a, _table FROM t_merge ORDER BY a; SELECT a, _table FROM t_distr ORDER BY a; From a82421719383041a839289093d1882265a068cd1 Mon Sep 17 00:00:00 2001 From: Nikita Mikhaylov Date: Fri, 23 Aug 2024 20:29:04 +0000 Subject: [PATCH 248/260] Done --- ..._rewrite_sum_column_and_constant.reference | 26 +++++++++---------- ...alyzer_rewrite_sum_column_and_constant.sql | 11 ++++---- 2 files changed, 19 insertions(+), 18 deletions(-) diff --git a/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.reference b/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.reference index 802d920aaef..b41635f014e 100644 --- a/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.reference +++ b/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.reference @@ -1635,21 +1635,21 @@ QUERY id: 0 JOIN TREE TABLE id: 10, alias: __table1, table_name: default.test_table SELECT sum(float64 + 2) From test_table; -26.5 +26.875 SELECT sum(2 + float64) From test_table; -26.5 +26.875 SELECT sum(float64 - 2) From test_table; -6.5 +6.875 SELECT sum(2 - float64) From test_table; --6.5 +-6.875 SELECT sum(float64) + 2 * count(float64) From test_table; -26.5 +26.875 SELECT 2 * count(float64) + sum(float64) From test_table; -26.5 +26.875 SELECT sum(float64) - 2 * count(float64) From test_table; -6.5 +6.875 SELECT 2 * count(float64) - sum(float64) From test_table; --6.5 +-6.875 EXPLAIN QUERY TREE (SELECT sum(float64 + 2) From test_table); QUERY id: 0 PROJECTION COLUMNS @@ -2463,25 +2463,25 @@ QUERY id: 0 JOIN TREE TABLE id: 12, alias: __table1, table_name: default.test_table SELECT sum(float64 + 2) + sum(float64 + 3) From test_table; -58 +58.75 SELECT sum(float64 + 2) - sum(float64 + 3) From test_table; -5 SELECT sum(float64 - 2) + sum(float64 - 3) From test_table; -8 +8.75 SELECT sum(float64 - 2) - sum(float64 - 3) From test_table; 5 SELECT sum(2 - float64) - sum(3 - float64) From test_table; -5 SELECT (sum(float64) + 2 * count(float64)) + (sum(float64) + 3 * count(float64)) From test_table; -58 +58.75 SELECT (sum(float64) + 2 * count(float64)) - (sum(float64) + 3 * count(float64)) From test_table; -5 SELECT (sum(float64) - 2 * count(float64)) + (sum(float64) - 3 * count(float64)) From test_table; -8 +8.75 SELECT (sum(float64) - 2 * count(float64)) - (sum(float64) - 3 * count(float64)) From test_table; 5 SELECT (2 * count(float64) - sum(float64)) + (3 * count(float64) - sum(float64)) From test_table; --8 +-8.75 EXPLAIN QUERY TREE (SELECT sum(float64 + 2) + sum(float64 + 3) From test_table); QUERY id: 0 PROJECTION COLUMNS diff --git a/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql b/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql index 5492d061c12..b6fa097abe9 100644 --- a/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql +++ b/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql @@ -25,11 +25,12 @@ CREATE TABLE test_table decimal32 Decimal32(5), ) ENGINE=MergeTree ORDER BY uint64; -INSERT INTO test_table VALUES (1, 1.1, 1.11); -INSERT INTO test_table VALUES (2, 2.2, 2.22); -INSERT INTO test_table VALUES (3, 3.3, 3.33); -INSERT INTO test_table VALUES (4, 4.4, 4.44); -INSERT INTO test_table VALUES (5, 5.5, 5.55); +-- Use Float64 numbers divisible by 1/16 (or some other small power of two), so that their sum doesn't depend on summation order. +INSERT INTO test_table VALUES (1, 1.125, 1.11); +INSERT INTO test_table VALUES (2, 2.250, 2.22); +INSERT INTO test_table VALUES (3, 3.375, 3.33); +INSERT INTO test_table VALUES (4, 4.500, 4.44); +INSERT INTO test_table VALUES (5, 5.625, 5.55); -- { echoOn } SELECT sum(uint64 + 1 AS i) from test_table where i > 0; From 0f265ce33d857a9c7446698629b6517b71b4a71d Mon Sep 17 00:00:00 2001 From: Han Fei Date: Fri, 23 Aug 2024 23:13:53 +0200 Subject: [PATCH 249/260] address comments --- src/Interpreters/InterpreterCreateQuery.cpp | 1 - src/Storages/AlterCommands.cpp | 10 ++++------ src/Storages/ColumnsDescription.cpp | 4 ---- src/Storages/MergeTree/MergeTask.cpp | 2 +- src/Storages/MergeTree/MutateTask.cpp | 4 ++-- src/Storages/Statistics/Statistics.cpp | 16 ++++++++-------- src/Storages/Statistics/Statistics.h | 6 ++++-- src/Storages/StatisticsDescription.cpp | 21 +++++---------------- src/Storages/StatisticsDescription.h | 4 ++-- 9 files changed, 26 insertions(+), 42 deletions(-) diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp index 95143031707..467547e6c9e 100644 --- a/src/Interpreters/InterpreterCreateQuery.cpp +++ b/src/Interpreters/InterpreterCreateQuery.cpp @@ -700,7 +700,6 @@ ColumnsDescription InterpreterCreateQuery::getColumnsDescription( col_decl.codec, column.type, sanity_check_compression_codecs, allow_experimental_codecs, enable_deflate_qpl_codec, enable_zstd_qat_codec); } - column.statistics.column_name = column.name; /// We assign column name here for better exception error message. if (col_decl.statistics_desc) { if (!skip_checks && !context_->getSettingsRef().allow_experimental_statistics) diff --git a/src/Storages/AlterCommands.cpp b/src/Storages/AlterCommands.cpp index 67b18217767..07bc87b0162 100644 --- a/src/Storages/AlterCommands.cpp +++ b/src/Storages/AlterCommands.cpp @@ -705,9 +705,9 @@ void AlterCommand::apply(StorageInMemoryMetadata & metadata, ContextPtr context) } auto stats_vec = ColumnStatisticsDescription::fromAST(statistics_decl, metadata.columns); - for (const auto & stats : stats_vec) + for (const auto & [stats_column_name, stats] : stats_vec) { - metadata.columns.modify(stats.column_name, + metadata.columns.modify(stats_column_name, [&](ColumnDescription & column) { column.statistics.merge(stats, column.name, column.type, if_not_exists); }); } } @@ -739,9 +739,9 @@ void AlterCommand::apply(StorageInMemoryMetadata & metadata, ContextPtr context) } auto stats_vec = ColumnStatisticsDescription::fromAST(statistics_decl, metadata.columns); - for (const auto & stats : stats_vec) + for (const auto & [stats_column_name, stats] : stats_vec) { - metadata.columns.modify(stats.column_name, + metadata.columns.modify(stats_column_name, [&](ColumnDescription & column) { column.statistics.assign(stats); }); } } @@ -866,8 +866,6 @@ void AlterCommand::apply(StorageInMemoryMetadata & metadata, ContextPtr context) rename_visitor.visit(column_to_modify.default_desc.expression); if (column_to_modify.ttl) rename_visitor.visit(column_to_modify.ttl); - if (column_to_modify.name == column_name && !column_to_modify.statistics.empty()) - column_to_modify.statistics.column_name = rename_to; }); } if (metadata.table_ttl.definition_ast) diff --git a/src/Storages/ColumnsDescription.cpp b/src/Storages/ColumnsDescription.cpp index 0d724245b49..fdc3446aa46 100644 --- a/src/Storages/ColumnsDescription.cpp +++ b/src/Storages/ColumnsDescription.cpp @@ -209,11 +209,7 @@ void ColumnDescription::readText(ReadBuffer & buf) settings = col_ast->settings->as().changes; if (col_ast->statistics_desc) - { statistics = ColumnStatisticsDescription::fromColumnDeclaration(*col_ast, type); - /// every column has name `x` here, so we have to set the name manually. - statistics.column_name = name; - } } else throw Exception(ErrorCodes::CANNOT_PARSE_TEXT, "Cannot parse column description"); diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp index ce06adf110c..0d34eb7f630 100644 --- a/src/Storages/MergeTree/MergeTask.cpp +++ b/src/Storages/MergeTree/MergeTask.cpp @@ -62,7 +62,7 @@ static ColumnsStatistics getStatisticsForColumns( const auto * desc = all_columns.tryGet(column.name); if (desc && !desc->statistics.empty()) { - auto statistics = MergeTreeStatisticsFactory::instance().get(desc->statistics); + auto statistics = MergeTreeStatisticsFactory::instance().get(*desc); all_statistics.push_back(std::move(statistics)); } } diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp index b9b5333a61c..1119ca324d6 100644 --- a/src/Storages/MergeTree/MutateTask.cpp +++ b/src/Storages/MergeTree/MutateTask.cpp @@ -546,7 +546,7 @@ static std::set getStatisticsToRecalculate(const StorageMet { if (!col_desc.statistics.empty() && materialized_stats.contains(col_desc.name)) { - stats_to_recalc.insert(stats_factory.get(col_desc.statistics)); + stats_to_recalc.insert(stats_factory.get(col_desc)); } } return stats_to_recalc; @@ -1530,7 +1530,7 @@ private: if (ctx->materialized_statistics.contains(col.name)) { - stats_to_rewrite.push_back(MergeTreeStatisticsFactory::instance().get(col.statistics)); + stats_to_rewrite.push_back(MergeTreeStatisticsFactory::instance().get(col)); } else { diff --git a/src/Storages/Statistics/Statistics.cpp b/src/Storages/Statistics/Statistics.cpp index fd686c5f0aa..6372c804e0e 100644 --- a/src/Storages/Statistics/Statistics.cpp +++ b/src/Storages/Statistics/Statistics.cpp @@ -58,8 +58,8 @@ IStatistics::IStatistics(const SingleStatisticsDescription & stat_) { } -ColumnStatistics::ColumnStatistics(const ColumnStatisticsDescription & stats_desc_) - : stats_desc(stats_desc_) +ColumnStatistics::ColumnStatistics(const ColumnStatisticsDescription & stats_desc_, const String & column_name_) + : stats_desc(stats_desc_), column_name(column_name_) { } @@ -176,7 +176,7 @@ String ColumnStatistics::getFileName() const const String & ColumnStatistics::columnName() const { - return stats_desc.column_name; + return column_name; } UInt64 ColumnStatistics::rowCount() const @@ -227,15 +227,15 @@ void MergeTreeStatisticsFactory::validate(const ColumnStatisticsDescription & st } } -ColumnStatisticsPtr MergeTreeStatisticsFactory::get(const ColumnStatisticsDescription & stats) const +ColumnStatisticsPtr MergeTreeStatisticsFactory::get(const ColumnDescription & column_desc) const { - ColumnStatisticsPtr column_stat = std::make_shared(stats); - for (const auto & [type, desc] : stats.types_to_desc) + ColumnStatisticsPtr column_stat = std::make_shared(column_desc.statistics, column_desc.name); + for (const auto & [type, desc] : column_desc.statistics.types_to_desc) { auto it = creators.find(type); if (it == creators.end()) throw Exception(ErrorCodes::INCORRECT_QUERY, "Unknown statistic type '{}'. Available types: 'tdigest' 'uniq' and 'count_min'", type); - auto stat_ptr = (it->second)(desc, stats.data_type); + auto stat_ptr = (it->second)(desc, column_desc.type); column_stat->stats[type] = stat_ptr; } return column_stat; @@ -246,7 +246,7 @@ ColumnsStatistics MergeTreeStatisticsFactory::getMany(const ColumnsDescription & ColumnsStatistics result; for (const auto & col : columns) if (!col.statistics.empty()) - result.push_back(get(col.statistics)); + result.push_back(get(col)); return result; } diff --git a/src/Storages/Statistics/Statistics.h b/src/Storages/Statistics/Statistics.h index 2a30c0de315..98666ed73df 100644 --- a/src/Storages/Statistics/Statistics.h +++ b/src/Storages/Statistics/Statistics.h @@ -54,7 +54,7 @@ using StatisticsPtr = std::shared_ptr; class ColumnStatistics { public: - explicit ColumnStatistics(const ColumnStatisticsDescription & stats_desc_); + explicit ColumnStatistics(const ColumnStatisticsDescription & stats_desc_, const String & column_name_); void serialize(WriteBuffer & buf); void deserialize(ReadBuffer & buf); @@ -73,10 +73,12 @@ public: private: friend class MergeTreeStatisticsFactory; ColumnStatisticsDescription stats_desc; + String column_name; std::map stats; UInt64 rows = 0; /// the number of rows in the column }; +struct ColumnDescription; class ColumnsDescription; using ColumnStatisticsPtr = std::shared_ptr; using ColumnsStatistics = std::vector; @@ -91,7 +93,7 @@ public: using Validator = std::function; using Creator = std::function; - ColumnStatisticsPtr get(const ColumnStatisticsDescription & stats) const; + ColumnStatisticsPtr get(const ColumnDescription & column_desc) const; ColumnsStatistics getMany(const ColumnsDescription & columns) const; void registerValidator(StatisticsType type, Validator validator); diff --git a/src/Storages/StatisticsDescription.cpp b/src/Storages/StatisticsDescription.cpp index acf600dd6f7..64634124758 100644 --- a/src/Storages/StatisticsDescription.cpp +++ b/src/Storages/StatisticsDescription.cpp @@ -96,16 +96,13 @@ void ColumnStatisticsDescription::merge(const ColumnStatisticsDescription & othe { chassert(merging_column_type); - if (column_name.empty()) - column_name = merging_column_name; - data_type = merging_column_type; for (const auto & [stats_type, stats_desc]: other.types_to_desc) { if (!if_not_exists && types_to_desc.contains(stats_type)) { - throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Statistics type name {} has existed in column {}", stats_type, column_name); + throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Statistics type name {} has existed in column {}", stats_type, merging_column_name); } else if (!types_to_desc.contains(stats_type)) types_to_desc.emplace(stats_type, stats_desc); @@ -114,12 +111,6 @@ void ColumnStatisticsDescription::merge(const ColumnStatisticsDescription & othe void ColumnStatisticsDescription::assign(const ColumnStatisticsDescription & other) { - /// If the statistics is empty, it's possible that we have not assign a column_name. - if (empty() && column_name == "") - column_name = other.column_name; - if (other.column_name != column_name) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot assign statistics from column {} to {}", other.column_name, column_name); - types_to_desc = other.types_to_desc; data_type = other.data_type; } @@ -129,7 +120,7 @@ void ColumnStatisticsDescription::clear() types_to_desc.clear(); } -std::vector ColumnStatisticsDescription::fromAST(const ASTPtr & definition_ast, const ColumnsDescription & columns) +std::vector> ColumnStatisticsDescription::fromAST(const ASTPtr & definition_ast, const ColumnsDescription & columns) { const auto * stat_definition_ast = definition_ast->as(); if (!stat_definition_ast) @@ -147,7 +138,7 @@ std::vector ColumnStatisticsDescription::fromAST(co statistics_types.emplace(stat.type, stat); } - std::vector result; + std::vector> result; result.reserve(stat_definition_ast->columns->children.size()); for (const auto & column_ast : stat_definition_ast->columns->children) @@ -159,10 +150,9 @@ std::vector ColumnStatisticsDescription::fromAST(co throw Exception(ErrorCodes::INCORRECT_QUERY, "Incorrect column name {}", physical_column_name); const auto & column = columns.getPhysical(physical_column_name); - stats.column_name = column.name; stats.data_type = column.type; stats.types_to_desc = statistics_types; - result.push_back(stats); + result.emplace_back(physical_column_name, stats); } if (result.empty()) @@ -177,14 +167,13 @@ ColumnStatisticsDescription ColumnStatisticsDescription::fromColumnDeclaration(c if (stat_type_list_ast->children.empty()) throw Exception(ErrorCodes::INCORRECT_QUERY, "We expect at least one statistics type for column {}", queryToString(column)); ColumnStatisticsDescription stats; - stats.column_name = column.name; for (const auto & ast : stat_type_list_ast->children) { const auto & stat_type = ast->as().name; SingleStatisticsDescription stat(stringToStatisticsType(Poco::toLower(stat_type)), ast->clone()); if (stats.types_to_desc.contains(stat.type)) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Column {} already contains statistics type {}", stats.column_name, stat_type); + throw Exception(ErrorCodes::INCORRECT_QUERY, "Column {} already contains statistics type {}", column.name, stat_type); stats.types_to_desc.emplace(stat.type, std::move(stat)); } stats.data_type = data_type; diff --git a/src/Storages/StatisticsDescription.h b/src/Storages/StatisticsDescription.h index 03b8fb0d583..46927f1418c 100644 --- a/src/Storages/StatisticsDescription.h +++ b/src/Storages/StatisticsDescription.h @@ -55,12 +55,12 @@ struct ColumnStatisticsDescription ASTPtr getAST() const; - static std::vector fromAST(const ASTPtr & definition_ast, const ColumnsDescription & columns); + /// get a vector of pair + static std::vector> fromAST(const ASTPtr & definition_ast, const ColumnsDescription & columns); static ColumnStatisticsDescription fromColumnDeclaration(const ASTColumnDeclaration & column, DataTypePtr data_type); using StatisticsTypeDescMap = std::map; StatisticsTypeDescMap types_to_desc; - String column_name; DataTypePtr data_type; }; From 6fb8f2b4ee10a95104bf6f8880471d24d39095dc Mon Sep 17 00:00:00 2001 From: Han Fei Date: Fri, 23 Aug 2024 23:19:03 +0200 Subject: [PATCH 250/260] fix black --- .../test_manipulate_statistics/test.py | 24 +++++++++++++------ 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/tests/integration/test_manipulate_statistics/test.py b/tests/integration/test_manipulate_statistics/test.py index ab5559e18fa..aff943e4d20 100644 --- a/tests/integration/test_manipulate_statistics/test.py +++ b/tests/integration/test_manipulate_statistics/test.py @@ -6,13 +6,17 @@ from helpers.cluster import ClickHouseCluster cluster = ClickHouseCluster(__file__) node1 = cluster.add_instance( - "node1", user_configs=["config/config.xml"], with_zookeeper=True, - macros={"replica": "a", "shard": "shard1"} + "node1", + user_configs=["config/config.xml"], + with_zookeeper=True, + macros={"replica": "a", "shard": "shard1"}, ) node2 = cluster.add_instance( - "node2", user_configs=["config/config.xml"], with_zookeeper=True, - macros={"replica": "b", "shard": "shard1"} + "node2", + user_configs=["config/config.xml"], + with_zookeeper=True, + macros={"replica": "b", "shard": "shard1"}, ) @@ -188,8 +192,14 @@ def test_replicated_table_ddl(started_cluster): def test_replicated_db(started_cluster): - node1.query("CREATE DATABASE test ENGINE = Replicated('/test/shared_stats', '{shard}', '{replica}')") - node2.query("CREATE DATABASE test ENGINE = Replicated('/test/shared_stats', '{shard}', '{replica}')") - node1.query("CREATE TABLE test.test_stats (a Int64, b Int64) ENGINE = ReplicatedMergeTree() ORDER BY()") + node1.query( + "CREATE DATABASE test ENGINE = Replicated('/test/shared_stats', '{shard}', '{replica}')" + ) + node2.query( + "CREATE DATABASE test ENGINE = Replicated('/test/shared_stats', '{shard}', '{replica}')" + ) + node1.query( + "CREATE TABLE test.test_stats (a Int64, b Int64) ENGINE = ReplicatedMergeTree() ORDER BY()" + ) node2.query("ALTER TABLE test.test_stats MODIFY COLUMN b Float64") node2.query("ALTER TABLE test.test_stats MODIFY STATISTICS b TYPE tdigest") From 0a35b111ffb34f3d6a8a9e9bfa712b57b722c447 Mon Sep 17 00:00:00 2001 From: Anton Popov Date: Fri, 23 Aug 2024 20:03:38 +0000 Subject: [PATCH 251/260] fix test 03221_mutation_analyzer_skip_part --- .../03221_mutation_analyzer_skip_part.sh | 46 +++++++++++++++++++ .../03221_mutation_analyzer_skip_part.sql | 21 --------- 2 files changed, 46 insertions(+), 21 deletions(-) create mode 100755 tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sh delete mode 100644 tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sql diff --git a/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sh b/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sh new file mode 100755 index 00000000000..03fd15f54e2 --- /dev/null +++ b/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sh @@ -0,0 +1,46 @@ +#!/usr/bin/env bash +# Tags: no-random-settings, no-random-merge-tree-settings + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --query " + DROP TABLE IF EXISTS t_mutate_skip_part; + + CREATE TABLE t_mutate_skip_part (key UInt64, id UInt64, v1 UInt64, v2 UInt64) + ENGINE = MergeTree ORDER BY id PARTITION BY key + SETTINGS min_bytes_for_wide_part = 0; + + INSERT INTO t_mutate_skip_part SELECT 1, number, number, number FROM numbers(10000); + INSERT INTO t_mutate_skip_part SELECT 2, number, number, number FROM numbers(10000); + + SET mutations_sync = 2; + ALTER TABLE t_mutate_skip_part UPDATE v1 = 1000 WHERE key = 1; + ALTER TABLE t_mutate_skip_part DELETE WHERE key = 2 AND v2 % 10 = 0; +" + +# Mutation query may return before the entry is added to part log. +# So, we may have to retry the flush of logs until all entries are actually flushed. +for _ in {1..10}; do + ${CLICKHOUSE_CLIENT} --query "SYSTEM FLUSH LOGS" + res=$(${CLICKHOUSE_CLIENT} --query "SELECT count() FROM system.part_log WHERE database = currentDatabase() AND table = 't_mutate_skip_part' AND event_type = 'MutatePart'") + + if [[ $res -eq 4 ]]; then + break + fi + + sleep 2.0 +done + +${CLICKHOUSE_CLIENT} --query " + SYSTEM FLUSH LOGS; + + -- If part is skipped in mutation and hardlinked then read_rows must be 0. + SELECT part_name, read_rows + FROM system.part_log + WHERE database = currentDatabase() AND table = 't_mutate_skip_part' AND event_type = 'MutatePart' + ORDER BY part_name; + + DROP TABLE IF EXISTS t_mutate_skip_part; +" diff --git a/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sql b/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sql deleted file mode 100644 index bf9a10e2af4..00000000000 --- a/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sql +++ /dev/null @@ -1,21 +0,0 @@ -DROP TABLE IF EXISTS t_mutate_skip_part; - -CREATE TABLE t_mutate_skip_part (key UInt64, id UInt64, v1 UInt64, v2 UInt64) ENGINE = MergeTree ORDER BY id PARTITION BY key; - -INSERT INTO t_mutate_skip_part SELECT 1, number, number, number FROM numbers(10000); -INSERT INTO t_mutate_skip_part SELECT 2, number, number, number FROM numbers(10000); - -SET mutations_sync = 2; - -ALTER TABLE t_mutate_skip_part UPDATE v1 = 1000 WHERE key = 1; -ALTER TABLE t_mutate_skip_part DELETE WHERE key = 2 AND v2 % 10 = 0; - -SYSTEM FLUSH LOGS; - --- If part is skipped in mutation and hardlinked then read_rows must be 0. -SELECT part_name, read_rows -FROM system.part_log -WHERE database = currentDatabase() AND table = 't_mutate_skip_part' AND event_type = 'MutatePart' -ORDER BY part_name; - -DROP TABLE IF EXISTS t_mutate_skip_part; From 080b8f74be186738813ca9d9e12ed3e327129c33 Mon Sep 17 00:00:00 2001 From: Tanya Bragin Date: Fri, 23 Aug 2024 15:50:56 -0700 Subject: [PATCH 252/260] Update README.md Add Austin meetup --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 546f08afd3d..ba212852ea8 100644 --- a/README.md +++ b/README.md @@ -54,6 +54,7 @@ Other upcoming meetups * [Sydney Meetup](https://www.meetup.com/clickhouse-australia-user-group/events/302862966/) - September 5 * [Zurich Meetup](https://www.meetup.com/clickhouse-switzerland-meetup-group/events/302267429/) - September 5 * [Toronto Meetup (Shopify)](https://www.meetup.com/clickhouse-toronto-user-group/events/301490855/) - September 10 +* [Austin Meetup](https://www.meetup.com/clickhouse-austin-user-group/events/302558689/) - September 17 * [London Meetup](https://www.meetup.com/clickhouse-london-user-group/events/302977267) - September 17 ## Recent Recordings From 5fe151529ab58112f8fa8491d2bfff24562ff624 Mon Sep 17 00:00:00 2001 From: Han Fei Date: Sat, 24 Aug 2024 07:33:18 +0200 Subject: [PATCH 253/260] fix flacky although that is not actually flacky --- tests/integration/test_manipulate_statistics/test.py | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_manipulate_statistics/test.py b/tests/integration/test_manipulate_statistics/test.py index aff943e4d20..3a1c5ad5b96 100644 --- a/tests/integration/test_manipulate_statistics/test.py +++ b/tests/integration/test_manipulate_statistics/test.py @@ -135,8 +135,8 @@ def test_single_node_normal(started_cluster): def test_replicated_table_ddl(started_cluster): - node1.query("DROP TABLE IF EXISTS test_stat") - node2.query("DROP TABLE IF EXISTS test_stat") + node1.query("DROP TABLE IF EXISTS test_stat SYNC") + node2.query("DROP TABLE IF EXISTS test_stat SYNC") node1.query( """ @@ -192,6 +192,8 @@ def test_replicated_table_ddl(started_cluster): def test_replicated_db(started_cluster): + node1.query("DROP DATABASE IF EXISTS test SYNC") + node2.query("DROP DATABASE IF EXISTS test SYNC") node1.query( "CREATE DATABASE test ENGINE = Replicated('/test/shared_stats', '{shard}', '{replica}')" ) From e2aa953e700bfbabbfe69a5749f4d2806bd3610f Mon Sep 17 00:00:00 2001 From: Amos Bird Date: Sat, 24 Aug 2024 20:45:10 +0800 Subject: [PATCH 254/260] Fix empty tuple in array --- src/Functions/array/arrayElement.cpp | 3 +++ tests/queries/0_stateless/03229_empty_tuple_in_array.reference | 1 + tests/queries/0_stateless/03229_empty_tuple_in_array.sql | 1 + 3 files changed, 5 insertions(+) create mode 100644 tests/queries/0_stateless/03229_empty_tuple_in_array.reference create mode 100644 tests/queries/0_stateless/03229_empty_tuple_in_array.sql diff --git a/src/Functions/array/arrayElement.cpp b/src/Functions/array/arrayElement.cpp index 81f3f97979b..d0b2b49cc1c 100644 --- a/src/Functions/array/arrayElement.cpp +++ b/src/Functions/array/arrayElement.cpp @@ -1598,6 +1598,9 @@ ColumnPtr FunctionArrayElement::executeTuple(const ColumnsWithTypeAndName & argu const auto & tuple_columns = col_nested->getColumns(); size_t tuple_size = tuple_columns.size(); + if (tuple_size == 0) + return ColumnTuple::create(input_rows_count); + const DataTypes & tuple_types = typeid_cast( *typeid_cast(*arguments[0].type).getNestedType()).getElements(); diff --git a/tests/queries/0_stateless/03229_empty_tuple_in_array.reference b/tests/queries/0_stateless/03229_empty_tuple_in_array.reference new file mode 100644 index 00000000000..6a452c185a8 --- /dev/null +++ b/tests/queries/0_stateless/03229_empty_tuple_in_array.reference @@ -0,0 +1 @@ +() diff --git a/tests/queries/0_stateless/03229_empty_tuple_in_array.sql b/tests/queries/0_stateless/03229_empty_tuple_in_array.sql new file mode 100644 index 00000000000..09ba3595a5a --- /dev/null +++ b/tests/queries/0_stateless/03229_empty_tuple_in_array.sql @@ -0,0 +1 @@ +select [()][0]; From 78c175225b9b4c929ed918e718351c18a166458a Mon Sep 17 00:00:00 2001 From: Nikita Mikhaylov Date: Thu, 22 Aug 2024 14:50:10 +0000 Subject: [PATCH 255/260] Done --- contrib/replxx | 2 +- src/Client/ReplxxLineReader.cpp | 13 +++++++------ 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/contrib/replxx b/contrib/replxx index 5d04501f93a..5f696c6eb9a 160000 --- a/contrib/replxx +++ b/contrib/replxx @@ -1 +1 @@ -Subproject commit 5d04501f93a4fb7f0bb8b73b8f614bc986f9e25b +Subproject commit 5f696c6eb9a88eb9784e8ff1d68bd5f70285dcc5 diff --git a/src/Client/ReplxxLineReader.cpp b/src/Client/ReplxxLineReader.cpp index 78ae6c5eb15..37ceb471e5b 100644 --- a/src/Client/ReplxxLineReader.cpp +++ b/src/Client/ReplxxLineReader.cpp @@ -299,13 +299,14 @@ ReplxxLineReader::ReplxxLineReader( Patterns delimiters_, const char word_break_characters_[], replxx::Replxx::highlighter_callback_t highlighter_, - [[ maybe_unused ]] std::istream & input_stream_, - [[ maybe_unused ]] std::ostream & output_stream_, - [[ maybe_unused ]] int in_fd_, - [[ maybe_unused ]] int out_fd_, - [[ maybe_unused ]] int err_fd_ + std::istream & input_stream_, + std::ostream & output_stream_, + int in_fd_, + int out_fd_, + int err_fd_ ) : LineReader(history_file_path_, multiline_, std::move(extenders_), std::move(delimiters_), input_stream_, output_stream_, in_fd_) + , rx(input_stream_, output_stream_, in_fd_, out_fd_, err_fd_) , highlighter(std::move(highlighter_)) , word_break_characters(word_break_characters_) , editor(getEditor()) @@ -516,7 +517,7 @@ void ReplxxLineReader::addToHistory(const String & line) rx.history_add(line); // flush changes to the disk - if (!rx.history_save(history_file_path)) + if (history_file_fd >= 0 && !rx.history_save(history_file_path)) rx.print("Saving history failed: %s\n", errnoToString().c_str()); if (history_file_fd >= 0 && locked && 0 != flock(history_file_fd, LOCK_UN)) From 01523cce2a4ba21c9855ab4eb1398986cf66c64b Mon Sep 17 00:00:00 2001 From: Nikita Mikhaylov Date: Fri, 23 Aug 2024 12:14:40 +0000 Subject: [PATCH 256/260] Bump replxx --- contrib/replxx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/replxx b/contrib/replxx index 5f696c6eb9a..711c18e7f4d 160000 --- a/contrib/replxx +++ b/contrib/replxx @@ -1 +1 @@ -Subproject commit 5f696c6eb9a88eb9784e8ff1d68bd5f70285dcc5 +Subproject commit 711c18e7f4d951255aa8b0851e5a55d5a5fb0ddb From 385c8127cf4b7018a964705d0bdcaf17bdf494e4 Mon Sep 17 00:00:00 2001 From: Nikita Mikhaylov Date: Fri, 23 Aug 2024 17:25:34 +0200 Subject: [PATCH 257/260] Fix FreeBSD build --- cmake/freebsd/toolchain-x86_64.cmake | 3 +++ 1 file changed, 3 insertions(+) diff --git a/cmake/freebsd/toolchain-x86_64.cmake b/cmake/freebsd/toolchain-x86_64.cmake index 4635880b4a6..4d814693b39 100644 --- a/cmake/freebsd/toolchain-x86_64.cmake +++ b/cmake/freebsd/toolchain-x86_64.cmake @@ -8,4 +8,7 @@ set (CMAKE_CXX_COMPILER_TARGET "x86_64-pc-freebsd11") set (CMAKE_ASM_COMPILER_TARGET "x86_64-pc-freebsd11") set (CMAKE_SYSROOT "${CMAKE_CURRENT_LIST_DIR}/../../contrib/sysroot/freebsd-x86_64") +# dprintf is used in a patched version of replxx +add_compile_definitions(_WITH_DPRINTF) + set (CMAKE_TRY_COMPILE_TARGET_TYPE STATIC_LIBRARY) # disable linkage check - it doesn't work in CMake From d16388000497251856f62e8ac67ade58c29f8e85 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Sun, 25 Aug 2024 00:11:31 -0400 Subject: [PATCH 258/260] process possible SSL error on connection reset --- base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp b/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp index 4873d259ae5..14c877b30af 100644 --- a/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp +++ b/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp @@ -311,6 +311,14 @@ int SecureSocketImpl::sendBytes(const void* buffer, int length, int flags) while (mustRetry(rc, remaining_time)); if (rc <= 0) { + // At this stage we still can have last not yet recieved SSL message containing SSL error + // so make a read to force SSL to process possible SSL error + if (SSL_get_error(_pSSL, rc) == SSL_ERROR_SYSCALL && SocketImpl::lastError() == POCO_ECONNRESET) + { + char c = 0; + SSL_read(_pSSL, &c, 1); + } + rc = handleError(rc); if (rc == 0) throw SSLConnectionUnexpectedlyClosedException(); } From f7cc3e9c59947af5b753b154f5b1b59d26fe67d4 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Sun, 25 Aug 2024 00:13:12 -0400 Subject: [PATCH 259/260] postpone SSL handshake --- src/Client/Connection.cpp | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/Client/Connection.cpp b/src/Client/Connection.cpp index e89bd7a2bf5..da6e5baa3ad 100644 --- a/src/Client/Connection.cpp +++ b/src/Client/Connection.cpp @@ -145,6 +145,9 @@ void Connection::connect(const ConnectionTimeouts & timeouts) /// work we need to pass host name separately. It will be send into TLS Hello packet to let /// the server know which host we want to talk with (single IP can process requests for multiple hosts using SNI). static_cast(socket.get())->setPeerHostName(host); + /// we want to postpone SSL handshake until first read or write operation + /// so any errors during negotiation would be properly processed + static_cast(socket.get())->setLazyHandshake(true); #else throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "tcp_secure protocol is disabled because poco library was built without NetSSL support."); #endif From f38f95a144fa8840bc19647af3be9aa83a505196 Mon Sep 17 00:00:00 2001 From: Nikita Mikhaylov Date: Sun, 25 Aug 2024 14:26:21 +0200 Subject: [PATCH 260/260] Update base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp --- base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp b/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp index 14c877b30af..eaf267d8a8b 100644 --- a/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp +++ b/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp @@ -311,7 +311,7 @@ int SecureSocketImpl::sendBytes(const void* buffer, int length, int flags) while (mustRetry(rc, remaining_time)); if (rc <= 0) { - // At this stage we still can have last not yet recieved SSL message containing SSL error + // At this stage we still can have last not yet received SSL message containing SSL error // so make a read to force SSL to process possible SSL error if (SSL_get_error(_pSSL, rc) == SSL_ERROR_SYSCALL && SocketImpl::lastError() == POCO_ECONNRESET) {