From 4978869d2f709a9ad93cefc04cda43bcc739fb22 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Tue, 23 Jul 2024 18:04:38 +0800 Subject: [PATCH 01/56] 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 02/56] 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 03/56] 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 04/56] 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 05/56] 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 06/56] 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 07/56] 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 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 08/56] 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 09/56] 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 10/56] 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 11/56] 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 12/56] 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 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 13/56] 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 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 14/56] 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 15/56] 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 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 16/56] 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 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 17/56] 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 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 18/56] 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 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 19/56] 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 29a0161cdfeeed739578af279092e7694dd7fbda Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Mon, 19 Aug 2024 18:25:01 +0800 Subject: [PATCH 20/56] change as request --- .../functions/string-replace-functions.md | 5 ++-- src/Functions/overlay.cpp | 24 ++++++++++++------- 2 files changed, 18 insertions(+), 11 deletions(-) diff --git a/docs/en/sql-reference/functions/string-replace-functions.md b/docs/en/sql-reference/functions/string-replace-functions.md index 408a896e607..55e97688b18 100644 --- a/docs/en/sql-reference/functions/string-replace-functions.md +++ b/docs/en/sql-reference/functions/string-replace-functions.md @@ -251,7 +251,8 @@ select printf('%%%s %s %d', 'Hello', 'World', 2024); ## overlay -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`. +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** @@ -286,7 +287,7 @@ Result: ## overlayUTF8 -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`. +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. diff --git a/src/Functions/overlay.cpp b/src/Functions/overlay.cpp index 73ca0acbb8e..20988c775a5 100644 --- a/src/Functions/overlay.cpp +++ b/src/Functions/overlay.cpp @@ -11,12 +11,6 @@ namespace DB { -namespace ErrorCodes -{ -extern const int ILLEGAL_TYPE_OF_ARGUMENT; -extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; -} - namespace { @@ -708,8 +702,20 @@ private: REGISTER_FUNCTION(Overlay) { - factory.registerFunction>({}, FunctionFactory::Case::Insensitive); - factory.registerFunction>({}, FunctionFactory::Case::Sensitive); -} + 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. +)", + .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. + +Assumes that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. +)", + .categories{"String"}}, + FunctionFactory::Case::Sensitive); +} } From 8653c547c9e5543b46cead484eb446725ec737d6 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Mon, 19 Aug 2024 18:50:04 +0800 Subject: [PATCH 21/56] change as request --- .../0_stateless/03205_overlay.reference | 88 ++++++++++--------- tests/queries/0_stateless/03205_overlay.sql | 72 +++++---------- 2 files changed, 70 insertions(+), 90 deletions(-) diff --git a/tests/queries/0_stateless/03205_overlay.reference b/tests/queries/0_stateless/03205_overlay.reference index 383a26986d6..67a699944e0 100644 --- a/tests/queries/0_stateless/03205_overlay.reference +++ b/tests/queries/0_stateless/03205_overlay.reference @@ -1,22 +1,5 @@ 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 +Test with 3 arguments and 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 @@ -25,16 +8,7 @@ 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 +Test with 4 arguments and 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 @@ -47,16 +21,48 @@ 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 +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 +-1 Spark ANSI Spark ANSI H +0 Spark ANSI SQL Spark ANSI SQL和CH +1 Spark ANSI QL Spark ANSI QL和CH +2 Spark ANSI L Spark ANSI L和CH +3 Spark ANSI Spark ANSI 和CH +4 Spark ANSI Spark ANSI CH +5 Spark ANSI Spark ANSI H +6 Spark ANSI Spark ANSI +Test with different input and replace values +_ _ +Spark SQL Spark SQL和CH +ANSI ANSI +Spark SQL Spark SQL和CH diff --git a/tests/queries/0_stateless/03205_overlay.sql b/tests/queries/0_stateless/03205_overlay.sql index 4fd0791521d..4d0b5ecbe03 100644 --- a/tests/queries/0_stateless/03205_overlay.sql +++ b/tests/queries/0_stateless/03205_overlay.sql @@ -7,7 +7,17 @@ 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 'Positive test 1 with various combinations of const/non-const columns'; +SELECT 'Test with 3 arguments and 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 'Test with 4 arguments and 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); @@ -24,51 +34,15 @@ SELECT overlay(materialize('Spark SQL'), 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)); + +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 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 different 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); +SELECT overlay('Spark SQL', '', 7, 0), overlayUTF8('Spark SQL和CH', '', 7, 0); From c37fc005d39f880b6ab12b01e80954f162817790 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Mon, 19 Aug 2024 13:28:23 +0200 Subject: [PATCH 22/56] init --- src/Functions/FunctionSQLJSON.h | 48 ++++++++++++++++++++------------- 1 file changed, 30 insertions(+), 18 deletions(-) diff --git a/src/Functions/FunctionSQLJSON.h b/src/Functions/FunctionSQLJSON.h index 83ed874c47b..66e0b4c2efd 100644 --- a/src/Functions/FunctionSQLJSON.h +++ b/src/Functions/FunctionSQLJSON.h @@ -123,7 +123,7 @@ public: class Executor { public: - static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, uint32_t parse_backtracks, const ContextPtr & context) + static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, uint32_t parse_backtracks, const bool & function_json_value_return_type_allow_complex) { MutableColumnPtr to{result_type->createColumn()}; to->reserve(input_rows_count); @@ -191,7 +191,7 @@ public: { /// Instead of creating a new generator for each row, we can reuse the same one. generator_json_path.reinitialize(); - added_to_column = impl.insertResultToColumn(*to, document, generator_json_path, context); + added_to_column = impl.insertResultToColumn(*to, document, generator_json_path, function_json_value_return_type_allow_complex); } if (!added_to_column) { @@ -204,13 +204,25 @@ public: }; template typename Impl> -class FunctionSQLJSON : public IFunction, WithConstContext +class FunctionSQLJSON : public IFunction { public: static FunctionPtr create(ContextPtr context_) { return std::make_shared(context_); } - explicit FunctionSQLJSON(ContextPtr context_) : WithConstContext(context_) { } + explicit FunctionSQLJSON(ContextPtr context_) + { + max_parser_depth = context_->getSettingsRef().max_parser_depth; + max_parser_backtracks = context_->getSettingsRef().max_parser_backtracks; + allow_simdjson = context_->getSettingsRef().allow_simdjson; + function_json_value_return_type_allow_complex = context_->getSettingsRef().function_json_value_return_type_allow_complex; + function_json_value_return_type_allow_nullable = context_->getSettingsRef().function_json_value_return_type_allow_nullable; + } static constexpr auto name = Name::name; + bool max_parser_depth; + bool max_parser_backtracks; + bool allow_simdjson; + bool function_json_value_return_type_allow_complex; + bool function_json_value_return_type_allow_nullable; String getName() const override { return Name::name; } bool isVariadic() const override { return true; } size_t getNumberOfArguments() const override { return 0; } @@ -221,7 +233,7 @@ public: DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { return Impl>::getReturnType( - Name::name, arguments, getContext()); + Name::name, arguments, function_json_value_return_type_allow_nullable); } ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override @@ -231,18 +243,18 @@ public: /// 2. Create ASTPtr /// 3. Parser(Tokens, ASTPtr) -> complete AST /// 4. Execute functions: call getNextItem on generator and handle each item - unsigned parse_depth = static_cast(getContext()->getSettingsRef().max_parser_depth); - unsigned parse_backtracks = static_cast(getContext()->getSettingsRef().max_parser_backtracks); + unsigned parse_depth = static_cast(max_parser_depth); + unsigned parse_backtracks = static_cast(max_parser_backtracks); #if USE_SIMDJSON - if (getContext()->getSettingsRef().allow_simdjson) + if (allow_simdjson) return FunctionSQLJSONHelpers::Executor< Name, Impl>, - SimdJSONParser>::run(arguments, result_type, input_rows_count, parse_depth, parse_backtracks, getContext()); + SimdJSONParser>::run(arguments, result_type, input_rows_count, parse_depth, parse_backtracks, function_json_value_return_type_allow_complex); #endif return FunctionSQLJSONHelpers:: Executor>, DummyJSONParser>::run( - arguments, result_type, input_rows_count, parse_depth, parse_backtracks, getContext()); + arguments, result_type, input_rows_count, parse_depth, parse_backtracks, function_json_value_return_type_allow_complex); } }; @@ -267,11 +279,11 @@ class JSONExistsImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const ContextPtr &) { return std::make_shared(); } + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool &) { return std::make_shared(); } static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const ContextPtr &) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool &) { Element current_element = root; VisitorStatus status; @@ -305,9 +317,9 @@ class JSONValueImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const ContextPtr & context) + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool & function_json_value_return_type_allow_nullable) { - if (context->getSettingsRef().function_json_value_return_type_allow_nullable) + if (function_json_value_return_type_allow_nullable) { DataTypePtr string_type = std::make_shared(); return std::make_shared(string_type); @@ -320,7 +332,7 @@ public: static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const ContextPtr & context) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool & function_json_value_return_type_allow_complex) { Element current_element = root; VisitorStatus status; @@ -329,7 +341,7 @@ public: { if (status == VisitorStatus::Ok) { - if (context->getSettingsRef().function_json_value_return_type_allow_complex) + if (function_json_value_return_type_allow_complex) { break; } @@ -383,11 +395,11 @@ class JSONQueryImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const ContextPtr &) { return std::make_shared(); } + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool &) { return std::make_shared(); } static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const ContextPtr &) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool &) { ColumnString & col_str = assert_cast(dest); From 0d444e6708f04cf4a581f34246e57745552ab001 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Mon, 19 Aug 2024 15:39:24 +0200 Subject: [PATCH 23/56] review fixes --- src/Functions/FunctionSQLJSON.h | 35 +++++++++++++++++---------------- 1 file changed, 18 insertions(+), 17 deletions(-) diff --git a/src/Functions/FunctionSQLJSON.h b/src/Functions/FunctionSQLJSON.h index 66e0b4c2efd..1c2a8d527f8 100644 --- a/src/Functions/FunctionSQLJSON.h +++ b/src/Functions/FunctionSQLJSON.h @@ -123,7 +123,7 @@ public: class Executor { public: - static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, uint32_t parse_backtracks, const bool & function_json_value_return_type_allow_complex) + static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, uint32_t parse_backtracks, const bool function_json_value_return_type_allow_complex) { MutableColumnPtr to{result_type->createColumn()}; to->reserve(input_rows_count); @@ -209,20 +209,15 @@ class FunctionSQLJSON : public IFunction public: static FunctionPtr create(ContextPtr context_) { return std::make_shared(context_); } explicit FunctionSQLJSON(ContextPtr context_) + : max_parser_depth(context_->getSettingsRef().max_parser_depth), + max_parser_backtracks(context_->getSettingsRef().max_parser_backtracks), + allow_simdjson(context_->getSettingsRef().allow_simdjson), + function_json_value_return_type_allow_complex(context_->getSettingsRef().function_json_value_return_type_allow_complex), + function_json_value_return_type_allow_nullable(context_->getSettingsRef().function_json_value_return_type_allow_nullable) { - max_parser_depth = context_->getSettingsRef().max_parser_depth; - max_parser_backtracks = context_->getSettingsRef().max_parser_backtracks; - allow_simdjson = context_->getSettingsRef().allow_simdjson; - function_json_value_return_type_allow_complex = context_->getSettingsRef().function_json_value_return_type_allow_complex; - function_json_value_return_type_allow_nullable = context_->getSettingsRef().function_json_value_return_type_allow_nullable; } static constexpr auto name = Name::name; - bool max_parser_depth; - bool max_parser_backtracks; - bool allow_simdjson; - bool function_json_value_return_type_allow_complex; - bool function_json_value_return_type_allow_nullable; String getName() const override { return Name::name; } bool isVariadic() const override { return true; } size_t getNumberOfArguments() const override { return 0; } @@ -256,6 +251,12 @@ public: Executor>, DummyJSONParser>::run( arguments, result_type, input_rows_count, parse_depth, parse_backtracks, function_json_value_return_type_allow_complex); } +private: + const size_t max_parser_depth; + const size_t max_parser_backtracks; + const bool allow_simdjson; + const bool function_json_value_return_type_allow_complex; + const bool function_json_value_return_type_allow_nullable; }; struct NameJSONExists @@ -279,11 +280,11 @@ class JSONExistsImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool &) { return std::make_shared(); } + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool) { return std::make_shared(); } static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool &) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool) { Element current_element = root; VisitorStatus status; @@ -317,7 +318,7 @@ class JSONValueImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool & function_json_value_return_type_allow_nullable) + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool function_json_value_return_type_allow_nullable) { if (function_json_value_return_type_allow_nullable) { @@ -332,7 +333,7 @@ public: static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool & function_json_value_return_type_allow_complex) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool function_json_value_return_type_allow_complex) { Element current_element = root; VisitorStatus status; @@ -395,11 +396,11 @@ class JSONQueryImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool &) { return std::make_shared(); } + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool) { return std::make_shared(); } static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool &) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool) { ColumnString & col_str = assert_cast(dest); From 7e2804fad5e66b28074eb28fa1f6332c741cdad2 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Mon, 19 Aug 2024 15:48:27 +0200 Subject: [PATCH 24/56] remove const (review) --- src/Functions/FunctionSQLJSON.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/src/Functions/FunctionSQLJSON.h b/src/Functions/FunctionSQLJSON.h index 1c2a8d527f8..4721f858f5c 100644 --- a/src/Functions/FunctionSQLJSON.h +++ b/src/Functions/FunctionSQLJSON.h @@ -123,7 +123,7 @@ public: class Executor { public: - static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, uint32_t parse_backtracks, const bool function_json_value_return_type_allow_complex) + static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, uint32_t parse_depth, uint32_t parse_backtracks, bool function_json_value_return_type_allow_complex) { MutableColumnPtr to{result_type->createColumn()}; to->reserve(input_rows_count); @@ -280,11 +280,11 @@ class JSONExistsImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool) { return std::make_shared(); } + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, bool) { return std::make_shared(); } static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, bool) { Element current_element = root; VisitorStatus status; @@ -318,7 +318,7 @@ class JSONValueImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool function_json_value_return_type_allow_nullable) + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, bool function_json_value_return_type_allow_nullable) { if (function_json_value_return_type_allow_nullable) { @@ -333,7 +333,7 @@ public: static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool function_json_value_return_type_allow_complex) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, bool function_json_value_return_type_allow_complex) { Element current_element = root; VisitorStatus status; @@ -396,11 +396,11 @@ class JSONQueryImpl public: using Element = typename JSONParser::Element; - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, const bool) { return std::make_shared(); } + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &, bool) { return std::make_shared(); } static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, const bool) + static bool insertResultToColumn(IColumn & dest, const Element & root, GeneratorJSONPath & generator_json_path, bool) { ColumnString & col_str = assert_cast(dest); From bc9cac605a58b4dce38bf06da889639f5bd62438 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 19 Aug 2024 22:00:35 +0000 Subject: [PATCH 25/56] Fix enumerating dynamic subcolumns --- src/DataTypes/IDataType.cpp | 1 + src/DataTypes/Serializations/ISerialization.h | 1 + .../Serializations/SerializationDynamic.cpp | 2 +- .../Serializations/SerializationObject.cpp | 2 +- ...dynamic_subcolumns_enumerate_streams.reference | 15 +++++++++++++++ ...03227_dynamic_subcolumns_enumerate_streams.sql | 9 +++++++++ 6 files changed, 28 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.reference create mode 100644 tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.sql diff --git a/src/DataTypes/IDataType.cpp b/src/DataTypes/IDataType.cpp index 7fd8a85aeca..d14461a4ff1 100644 --- a/src/DataTypes/IDataType.cpp +++ b/src/DataTypes/IDataType.cpp @@ -150,6 +150,7 @@ std::unique_ptr IDataType::getSubcolumnData( ISerialization::EnumerateStreamsSettings settings; settings.position_independent_encoding = false; + settings.enumerate_dynamic_streams = false; data.serialization->enumerateStreams(settings, callback_with_data, data); if (!res && data.type->hasDynamicSubcolumnsData()) diff --git a/src/DataTypes/Serializations/ISerialization.h b/src/DataTypes/Serializations/ISerialization.h index 480d5a4f7c4..d19cb0dd365 100644 --- a/src/DataTypes/Serializations/ISerialization.h +++ b/src/DataTypes/Serializations/ISerialization.h @@ -241,6 +241,7 @@ public: { SubstreamPath path; bool position_independent_encoding = true; + bool enumerate_dynamic_streams = true; }; virtual void enumerateStreams( diff --git a/src/DataTypes/Serializations/SerializationDynamic.cpp b/src/DataTypes/Serializations/SerializationDynamic.cpp index 6bba87c40fa..10a80aa8a0d 100644 --- a/src/DataTypes/Serializations/SerializationDynamic.cpp +++ b/src/DataTypes/Serializations/SerializationDynamic.cpp @@ -64,7 +64,7 @@ void SerializationDynamic::enumerateStreams( const auto * deserialize_state = data.deserialize_state ? checkAndGetState(data.deserialize_state) : nullptr; /// If column is nullptr and we don't have deserialize state yet, nothing to enumerate as we don't have any variants. - if (!column_dynamic && !deserialize_state) + if (!settings.enumerate_dynamic_streams || (!column_dynamic && !deserialize_state)) return; const auto & variant_type = column_dynamic ? column_dynamic->getVariantInfo().variant_type : checkAndGetState(deserialize_state->structure_state)->variant_type; diff --git a/src/DataTypes/Serializations/SerializationObject.cpp b/src/DataTypes/Serializations/SerializationObject.cpp index 2dd25e540cc..45121b942e6 100644 --- a/src/DataTypes/Serializations/SerializationObject.cpp +++ b/src/DataTypes/Serializations/SerializationObject.cpp @@ -130,7 +130,7 @@ void SerializationObject::enumerateStreams(EnumerateStreamsSettings & settings, } /// If column or deserialization state was provided, iterate over dynamic paths, - if (column_object || structure_state) + if (settings.enumerate_dynamic_streams && (column_object || structure_state)) { /// Enumerate dynamic paths in sorted order for consistency. const auto * dynamic_paths = column_object ? &column_object->getDynamicPaths() : nullptr; diff --git a/tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.reference b/tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.reference new file mode 100644 index 00000000000..b9d711bf9f0 --- /dev/null +++ b/tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.reference @@ -0,0 +1,15 @@ +\N +\N +\N +\N +\N +str_0 +str_1 +str_2 +str_3 +str_4 +\N +\N +\N +\N +\N diff --git a/tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.sql b/tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.sql new file mode 100644 index 00000000000..e451521faca --- /dev/null +++ b/tests/queries/0_stateless/03227_dynamic_subcolumns_enumerate_streams.sql @@ -0,0 +1,9 @@ +set allow_experimental_json_type=1; +drop table if exists test; +create table test (json JSON) engine=Memory; +insert into test select toJSONString(map('a', 'str_' || number)) from numbers(5); +select json.a.String from test; +select json.a.:String from test; +select json.a.UInt64 from test; +drop table test; + From 4f799467ecaf1bf10e39f7da59aa459ab1903810 Mon Sep 17 00:00:00 2001 From: Michal Tabaszewski Date: Fri, 16 Aug 2024 22:54:09 +0200 Subject: [PATCH 26/56] Added removing query cache by tag. --- src/Common/CacheBase.h | 6 ++++++ src/Common/ICachePolicy.h | 1 + src/Common/LRUCachePolicy.h | 17 +++++++++++++++++ src/Common/SLRUCachePolicy.h | 21 +++++++++++++++++++++ src/Common/TTLCachePolicy.h | 17 +++++++++++++++++ src/Interpreters/Cache/QueryCache.cpp | 9 +++++++++ src/Interpreters/Cache/QueryCache.h | 1 + src/Interpreters/Context.cpp | 8 ++++++++ src/Interpreters/Context.h | 1 + src/Interpreters/InterpreterSystemQuery.cpp | 7 +++++-- src/Parsers/ASTSystemQuery.h | 1 + src/Parsers/CommonParsers.h | 1 + src/Parsers/ParserSystemQuery.cpp | 10 ++++++++++ 13 files changed, 98 insertions(+), 2 deletions(-) diff --git a/src/Common/CacheBase.h b/src/Common/CacheBase.h index a809136f451..e0f7ffbfd5f 100644 --- a/src/Common/CacheBase.h +++ b/src/Common/CacheBase.h @@ -197,6 +197,12 @@ public: cache_policy->remove(key); } + void removeWithPredicate(std::function predicate) + { + std::lock_guard lock(mutex); + cache_policy->removeWithPredicate(predicate); + } + size_t sizeInBytes() const { std::lock_guard lock(mutex); diff --git a/src/Common/ICachePolicy.h b/src/Common/ICachePolicy.h index 301a5c6cbbd..71543e0aed0 100644 --- a/src/Common/ICachePolicy.h +++ b/src/Common/ICachePolicy.h @@ -55,6 +55,7 @@ public: virtual void set(const Key & key, const MappedPtr & mapped) = 0; virtual void remove(const Key & key) = 0; + virtual void removeWithPredicate(std::function predicate) = 0; virtual void clear() = 0; virtual std::vector dump() const = 0; diff --git a/src/Common/LRUCachePolicy.h b/src/Common/LRUCachePolicy.h index f833e46a821..39bf5067aea 100644 --- a/src/Common/LRUCachePolicy.h +++ b/src/Common/LRUCachePolicy.h @@ -68,6 +68,23 @@ public: current_size_in_bytes = 0; } + void removeWithPredicate(std::function predicate) override + { + for(auto it = cells.begin(); it != cells.end();) + { + if(predicate(it->first, it->second.value)) + { + auto & cell = it->second; + current_size_in_bytes -= cell.size; + queue.erase(cell.queue_iterator); + cells.erase(it); + it = cells.erase(it); + } + else + ++it; + } + } + void remove(const Key & key) override { auto it = cells.find(key); diff --git a/src/Common/SLRUCachePolicy.h b/src/Common/SLRUCachePolicy.h index 354ec1d36d6..62c6d72a3e8 100644 --- a/src/Common/SLRUCachePolicy.h +++ b/src/Common/SLRUCachePolicy.h @@ -95,6 +95,27 @@ public: cells.erase(it); } + void removeWithPredicate(std::function predicate) override + { + for(auto it = cells.begin(); it != cells.end();) + { + if(predicate(it->first, it->second.value)) + { + auto & cell = it->second; + + current_size_in_bytes -= cell.size; + if (cell.is_protected) + current_protected_size -= cell.size; + + auto & queue = cell.is_protected ? protected_queue : probationary_queue; + queue.erase(cell.queue_iterator); + it = cells.erase(it); + } + else + ++it; + } + } + MappedPtr get(const Key & key) override { auto it = cells.find(key); diff --git a/src/Common/TTLCachePolicy.h b/src/Common/TTLCachePolicy.h index 6401835b0d7..6caab1d6c78 100644 --- a/src/Common/TTLCachePolicy.h +++ b/src/Common/TTLCachePolicy.h @@ -133,6 +133,23 @@ public: Base::user_quotas->clear(); } + void removeWithPredicate(std::function predicate) override + { + for(auto it = cache.begin(); it != cache.end();) + { + if(predicate(it->first, it->second)) + { + size_t sz = weight_function(*it->second); + if (it->first.user_id.has_value()) + Base::user_quotas->decreaseActual(*it->first.user_id, sz); + it = cache.erase(it); + size_in_bytes -= sz; + } + else + ++it; + } + } + void remove(const Key & key) override { auto it = cache.find(key); diff --git a/src/Interpreters/Cache/QueryCache.cpp b/src/Interpreters/Cache/QueryCache.cpp index ab926037c67..9c4d48fbd44 100644 --- a/src/Interpreters/Cache/QueryCache.cpp +++ b/src/Interpreters/Cache/QueryCache.cpp @@ -619,6 +619,15 @@ QueryCache::Writer QueryCache::createWriter(const Key & key, std::chrono::millis return Writer(cache, key, max_entry_size_in_bytes, max_entry_size_in_rows, min_query_runtime, squash_partial_results, max_block_size); } +void QueryCache::clearWithTag(const String & tag) +{ + auto removeWithTag = [tag](const Key & k, const Cache::MappedPtr & _){ + return k.tag == tag; + }; + cache.removeWithPredicate(removeWithTag); + std::lock_guard lock(mutex); +} + void QueryCache::clear() { cache.clear(); diff --git a/src/Interpreters/Cache/QueryCache.h b/src/Interpreters/Cache/QueryCache.h index c7ebaf4d26a..dec43747e0a 100644 --- a/src/Interpreters/Cache/QueryCache.h +++ b/src/Interpreters/Cache/QueryCache.h @@ -212,6 +212,7 @@ public: Writer createWriter(const Key & key, std::chrono::milliseconds min_query_runtime, bool squash_partial_results, size_t max_block_size, size_t max_query_cache_size_in_bytes_quota, size_t max_query_cache_entries_quota); void clear(); + void clearWithTag(const String & tag); size_t sizeInBytes() const; size_t count() const; diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index 3cc09370e86..ef5539f0b13 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -3236,6 +3236,14 @@ void Context::clearQueryCache() const shared->query_cache->clear(); } +void Context::clearQueryCacheWithTag(const String & tag) const +{ + std::lock_guard lock(shared->mutex); + + if (shared->query_cache) + shared->query_cache->clearWithTag(tag); +} + void Context::clearCaches() const { std::lock_guard lock(shared->mutex); diff --git a/src/Interpreters/Context.h b/src/Interpreters/Context.h index 3da4f124553..076bff462f9 100644 --- a/src/Interpreters/Context.h +++ b/src/Interpreters/Context.h @@ -1069,6 +1069,7 @@ public: void updateQueryCacheConfiguration(const Poco::Util::AbstractConfiguration & config); std::shared_ptr getQueryCache() const; void clearQueryCache() const; + void clearQueryCacheWithTag(const String & tag) const; /** Clear the caches of the uncompressed blocks and marks. * This is usually done when renaming tables, changing the type of columns, deleting a table. diff --git a/src/Interpreters/InterpreterSystemQuery.cpp b/src/Interpreters/InterpreterSystemQuery.cpp index 21c8b44b374..c71982d0934 100644 --- a/src/Interpreters/InterpreterSystemQuery.cpp +++ b/src/Interpreters/InterpreterSystemQuery.cpp @@ -369,9 +369,12 @@ BlockIO InterpreterSystemQuery::execute() system_context->clearMMappedFileCache(); break; case Type::DROP_QUERY_CACHE: - getContext()->checkAccess(AccessType::SYSTEM_DROP_QUERY_CACHE); - getContext()->clearQueryCache(); + { + getContext()->checkAccess(AccessType::SYSTEM_DROP_QUERY_CACHE); + !query.query_cache_tag.empty() ? getContext()->clearQueryCacheWithTag(query.query_cache_tag): getContext()->clearQueryCache(); break; + } + case Type::DROP_COMPILED_EXPRESSION_CACHE: #if USE_EMBEDDED_COMPILER getContext()->checkAccess(AccessType::SYSTEM_DROP_COMPILED_EXPRESSION_CACHE); diff --git a/src/Parsers/ASTSystemQuery.h b/src/Parsers/ASTSystemQuery.h index 59de90b1d8e..739d71678df 100644 --- a/src/Parsers/ASTSystemQuery.h +++ b/src/Parsers/ASTSystemQuery.h @@ -129,6 +129,7 @@ public: String storage_policy; String volume; String disk; + String query_cache_tag; UInt64 seconds{}; String filesystem_cache_name; diff --git a/src/Parsers/CommonParsers.h b/src/Parsers/CommonParsers.h index cc2ee79cd36..a079fba5d53 100644 --- a/src/Parsers/CommonParsers.h +++ b/src/Parsers/CommonParsers.h @@ -471,6 +471,7 @@ namespace DB MR_MACROS(TABLE, "TABLE") \ MR_MACROS(TABLES, "TABLES") \ MR_MACROS(TAGS, "TAGS") \ + MR_MACROS(TAG, "TAG") \ MR_MACROS(TAGS_INNER_UUID, "TAGS INNER UUID") \ MR_MACROS(TEMPORARY_TABLE, "TEMPORARY TABLE") \ MR_MACROS(TEMPORARY, "TEMPORARY") \ diff --git a/src/Parsers/ParserSystemQuery.cpp b/src/Parsers/ParserSystemQuery.cpp index efabbbfa479..95d89e6fde1 100644 --- a/src/Parsers/ParserSystemQuery.cpp +++ b/src/Parsers/ParserSystemQuery.cpp @@ -489,6 +489,16 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & return false; break; } + case Type::DROP_QUERY_CACHE: + { + ParserLiteral tag_parser; + ASTPtr ast; + if (ParserKeyword{Keyword::TAG}.ignore(pos, expected) && tag_parser.parse(pos, ast, expected)) + res->query_cache_tag = ast->as()->value.safeGet(); + if (!parseQueryWithOnCluster(res, pos, expected)) + return false; + break; + } case Type::SYNC_FILESYSTEM_CACHE: { ParserLiteral path_parser; From cd51535a67319d06fae7ca212123c9cb9114c91c Mon Sep 17 00:00:00 2001 From: Michal Tabaszewski Date: Sat, 17 Aug 2024 01:54:46 +0200 Subject: [PATCH 27/56] Fixed style. --- src/Common/LRUCachePolicy.h | 4 ++-- src/Common/SLRUCachePolicy.h | 4 ++-- src/Common/TTLCachePolicy.h | 4 ++-- src/Interpreters/Cache/QueryCache.cpp | 5 +++-- src/Interpreters/InterpreterSystemQuery.cpp | 2 +- 5 files changed, 10 insertions(+), 9 deletions(-) diff --git a/src/Common/LRUCachePolicy.h b/src/Common/LRUCachePolicy.h index 39bf5067aea..9f2ea548c32 100644 --- a/src/Common/LRUCachePolicy.h +++ b/src/Common/LRUCachePolicy.h @@ -70,9 +70,9 @@ public: void removeWithPredicate(std::function predicate) override { - for(auto it = cells.begin(); it != cells.end();) + for (auto it = cells.begin(); it != cells.end();) { - if(predicate(it->first, it->second.value)) + if (predicate(it->first, it->second.value)) { auto & cell = it->second; current_size_in_bytes -= cell.size; diff --git a/src/Common/SLRUCachePolicy.h b/src/Common/SLRUCachePolicy.h index 62c6d72a3e8..e6931bbd59c 100644 --- a/src/Common/SLRUCachePolicy.h +++ b/src/Common/SLRUCachePolicy.h @@ -97,9 +97,9 @@ public: void removeWithPredicate(std::function predicate) override { - for(auto it = cells.begin(); it != cells.end();) + for (auto it = cells.begin(); it != cells.end();) { - if(predicate(it->first, it->second.value)) + if (predicate(it->first, it->second.value)) { auto & cell = it->second; diff --git a/src/Common/TTLCachePolicy.h b/src/Common/TTLCachePolicy.h index 6caab1d6c78..100d753e53b 100644 --- a/src/Common/TTLCachePolicy.h +++ b/src/Common/TTLCachePolicy.h @@ -135,9 +135,9 @@ public: void removeWithPredicate(std::function predicate) override { - for(auto it = cache.begin(); it != cache.end();) + for (auto it = cache.begin(); it != cache.end();) { - if(predicate(it->first, it->second)) + if (predicate(it->first, it->second)) { size_t sz = weight_function(*it->second); if (it->first.user_id.has_value()) diff --git a/src/Interpreters/Cache/QueryCache.cpp b/src/Interpreters/Cache/QueryCache.cpp index 9c4d48fbd44..84aa9cd0342 100644 --- a/src/Interpreters/Cache/QueryCache.cpp +++ b/src/Interpreters/Cache/QueryCache.cpp @@ -621,10 +621,11 @@ QueryCache::Writer QueryCache::createWriter(const Key & key, std::chrono::millis void QueryCache::clearWithTag(const String & tag) { - auto removeWithTag = [tag](const Key & k, const Cache::MappedPtr & _){ + auto remove_with_tag = [tag](const Key & k, const Cache::MappedPtr & _) + { return k.tag == tag; }; - cache.removeWithPredicate(removeWithTag); + cache.removeWithPredicate(remove_with_tag); std::lock_guard lock(mutex); } diff --git a/src/Interpreters/InterpreterSystemQuery.cpp b/src/Interpreters/InterpreterSystemQuery.cpp index c71982d0934..c7540084451 100644 --- a/src/Interpreters/InterpreterSystemQuery.cpp +++ b/src/Interpreters/InterpreterSystemQuery.cpp @@ -370,7 +370,7 @@ BlockIO InterpreterSystemQuery::execute() break; case Type::DROP_QUERY_CACHE: { - getContext()->checkAccess(AccessType::SYSTEM_DROP_QUERY_CACHE); + getContext()->checkAccess(AccessType::SYSTEM_DROP_QUERY_CACHE); !query.query_cache_tag.empty() ? getContext()->clearQueryCacheWithTag(query.query_cache_tag): getContext()->clearQueryCache(); break; } From ad20d24c6e0aa50ea408627389acb4c83080d3e0 Mon Sep 17 00:00:00 2001 From: Michal Tabaszewski Date: Mon, 19 Aug 2024 20:52:51 +0200 Subject: [PATCH 28/56] Fixed binary-tidy complaint --- src/Interpreters/Cache/QueryCache.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/Cache/QueryCache.cpp b/src/Interpreters/Cache/QueryCache.cpp index 84aa9cd0342..6c1ffe9cecb 100644 --- a/src/Interpreters/Cache/QueryCache.cpp +++ b/src/Interpreters/Cache/QueryCache.cpp @@ -621,7 +621,7 @@ QueryCache::Writer QueryCache::createWriter(const Key & key, std::chrono::millis void QueryCache::clearWithTag(const String & tag) { - auto remove_with_tag = [tag](const Key & k, const Cache::MappedPtr & _) + auto remove_with_tag = [tag](const Key & k, const Cache::MappedPtr &) { return k.tag == tag; }; From eee618b3196cecb44f6314a4199d5aa95ff2ca66 Mon Sep 17 00:00:00 2001 From: Michal Tabaszewski Date: Mon, 19 Aug 2024 22:37:28 +0200 Subject: [PATCH 29/56] - From 129d1e387e1b436ac572ca5db1a50dd9ed79ddae Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Tue, 20 Aug 2024 13:21:42 +0000 Subject: [PATCH 30/56] Some fixups --- docs/en/operations/query-cache.md | 2 ++ docs/en/sql-reference/statements/system.md | 6 ++++ src/Common/CacheBase.h | 4 +-- src/Common/ICachePolicy.h | 2 +- src/Common/LRUCachePolicy.h | 33 +++++++++---------- src/Common/SLRUCachePolicy.h | 2 +- src/Common/TTLCachePolicy.h | 26 +++++++-------- src/Interpreters/Cache/QueryCache.cpp | 19 +++++------ src/Interpreters/Cache/QueryCache.h | 3 +- src/Interpreters/Context.cpp | 12 ++----- src/Interpreters/Context.h | 3 +- src/Interpreters/InterpreterSystemQuery.cpp | 2 +- src/Parsers/ASTSystemQuery.h | 3 +- src/Parsers/CommonParsers.h | 2 +- src/Parsers/ParserSystemQuery.cpp | 20 +++++------ .../02494_query_cache_drop_cache.reference | 14 ++++++++ .../02494_query_cache_drop_cache.sql | 25 ++++++++++++-- 17 files changed, 105 insertions(+), 73 deletions(-) diff --git a/docs/en/operations/query-cache.md b/docs/en/operations/query-cache.md index 384938e28f6..955cec0234e 100644 --- a/docs/en/operations/query-cache.md +++ b/docs/en/operations/query-cache.md @@ -155,6 +155,8 @@ SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'tag 1'; SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'tag 2'; ``` +To remove only entries with tag `tag` from the query cache, you can use statement `SYSTEM DROP QUERY CACHE TAG 'tag'`. + ClickHouse reads table data in blocks of [max_block_size](settings/settings.md#setting-max_block_size) rows. Due to filtering, aggregation, etc., result blocks are typically much smaller than 'max_block_size' but there are also cases where they are much bigger. Setting [query_cache_squash_partial_results](settings/settings.md#query-cache-squash-partial-results) (enabled by default) controls if result blocks diff --git a/docs/en/sql-reference/statements/system.md b/docs/en/sql-reference/statements/system.md index 3ebcf617491..77d023b67ce 100644 --- a/docs/en/sql-reference/statements/system.md +++ b/docs/en/sql-reference/statements/system.md @@ -136,7 +136,13 @@ The compiled expression cache is enabled/disabled with the query/user/profile-le ## DROP QUERY CACHE +```sql +SYSTEM DROP QUERY CACHE; +SYSTEM DROP QUERY CACHE TAG '' +```` + Clears the [query cache](../../operations/query-cache.md). +If a tag is specified, only query cache entries with the specified tag are deleted. ## DROP FORMAT SCHEMA CACHE {#system-drop-schema-format} diff --git a/src/Common/CacheBase.h b/src/Common/CacheBase.h index e0f7ffbfd5f..23e6a6fc91c 100644 --- a/src/Common/CacheBase.h +++ b/src/Common/CacheBase.h @@ -197,10 +197,10 @@ public: cache_policy->remove(key); } - void removeWithPredicate(std::function predicate) + void remove(std::function predicate) { std::lock_guard lock(mutex); - cache_policy->removeWithPredicate(predicate); + cache_policy->remove(predicate); } size_t sizeInBytes() const diff --git a/src/Common/ICachePolicy.h b/src/Common/ICachePolicy.h index 71543e0aed0..567fa35d977 100644 --- a/src/Common/ICachePolicy.h +++ b/src/Common/ICachePolicy.h @@ -55,7 +55,7 @@ public: virtual void set(const Key & key, const MappedPtr & mapped) = 0; virtual void remove(const Key & key) = 0; - virtual void removeWithPredicate(std::function predicate) = 0; + virtual void remove(std::function predicate) = 0; virtual void clear() = 0; virtual std::vector dump() const = 0; diff --git a/src/Common/LRUCachePolicy.h b/src/Common/LRUCachePolicy.h index 9f2ea548c32..cb8fdbd2b9c 100644 --- a/src/Common/LRUCachePolicy.h +++ b/src/Common/LRUCachePolicy.h @@ -68,23 +68,6 @@ public: current_size_in_bytes = 0; } - void removeWithPredicate(std::function predicate) override - { - for (auto it = cells.begin(); it != cells.end();) - { - if (predicate(it->first, it->second.value)) - { - auto & cell = it->second; - current_size_in_bytes -= cell.size; - queue.erase(cell.queue_iterator); - cells.erase(it); - it = cells.erase(it); - } - else - ++it; - } - } - void remove(const Key & key) override { auto it = cells.find(key); @@ -96,6 +79,22 @@ public: cells.erase(it); } + void remove(std::function predicate) override + { + for (auto it = cells.begin(); it != cells.end();) + { + if (predicate(it->first, it->second.value)) + { + Cell & cell = it->second; + current_size_in_bytes -= cell.size; + queue.erase(cell.queue_iterator); + it = cells.erase(it); + } + else + ++it; + } + } + MappedPtr get(const Key & key) override { auto it = cells.find(key); diff --git a/src/Common/SLRUCachePolicy.h b/src/Common/SLRUCachePolicy.h index e6931bbd59c..5321110f3e5 100644 --- a/src/Common/SLRUCachePolicy.h +++ b/src/Common/SLRUCachePolicy.h @@ -95,7 +95,7 @@ public: cells.erase(it); } - void removeWithPredicate(std::function predicate) override + void remove(std::function predicate) override { for (auto it = cells.begin(); it != cells.end();) { diff --git a/src/Common/TTLCachePolicy.h b/src/Common/TTLCachePolicy.h index 100d753e53b..6c548e5042b 100644 --- a/src/Common/TTLCachePolicy.h +++ b/src/Common/TTLCachePolicy.h @@ -133,7 +133,19 @@ public: Base::user_quotas->clear(); } - void removeWithPredicate(std::function predicate) override + void remove(const Key & key) override + { + auto it = cache.find(key); + if (it == cache.end()) + return; + size_t sz = weight_function(*it->second); + if (it->first.user_id.has_value()) + Base::user_quotas->decreaseActual(*it->first.user_id, sz); + cache.erase(it); + size_in_bytes -= sz; + } + + void remove(std::function predicate) override { for (auto it = cache.begin(); it != cache.end();) { @@ -150,18 +162,6 @@ public: } } - void remove(const Key & key) override - { - auto it = cache.find(key); - if (it == cache.end()) - return; - size_t sz = weight_function(*it->second); - if (it->first.user_id.has_value()) - Base::user_quotas->decreaseActual(*it->first.user_id, sz); - cache.erase(it); - size_in_bytes -= sz; - } - MappedPtr get(const Key & key) override { auto it = cache.find(key); diff --git a/src/Interpreters/Cache/QueryCache.cpp b/src/Interpreters/Cache/QueryCache.cpp index 6c1ffe9cecb..4312b35e18c 100644 --- a/src/Interpreters/Cache/QueryCache.cpp +++ b/src/Interpreters/Cache/QueryCache.cpp @@ -619,19 +619,18 @@ QueryCache::Writer QueryCache::createWriter(const Key & key, std::chrono::millis return Writer(cache, key, max_entry_size_in_bytes, max_entry_size_in_rows, min_query_runtime, squash_partial_results, max_block_size); } -void QueryCache::clearWithTag(const String & tag) +void QueryCache::clear(const std::optional & tag) { - auto remove_with_tag = [tag](const Key & k, const Cache::MappedPtr &) + if (tag) { - return k.tag == tag; - }; - cache.removeWithPredicate(remove_with_tag); - std::lock_guard lock(mutex); -} + auto predicate = [tag](const Key & key, const Cache::MappedPtr &) { return key.tag == tag.value(); }; + cache.remove(predicate); + } + else + { + cache.clear(); + } -void QueryCache::clear() -{ - cache.clear(); std::lock_guard lock(mutex); times_executed.clear(); } diff --git a/src/Interpreters/Cache/QueryCache.h b/src/Interpreters/Cache/QueryCache.h index dec43747e0a..64407633a8d 100644 --- a/src/Interpreters/Cache/QueryCache.h +++ b/src/Interpreters/Cache/QueryCache.h @@ -211,8 +211,7 @@ public: Reader createReader(const Key & key); Writer createWriter(const Key & key, std::chrono::milliseconds min_query_runtime, bool squash_partial_results, size_t max_block_size, size_t max_query_cache_size_in_bytes_quota, size_t max_query_cache_entries_quota); - void clear(); - void clearWithTag(const String & tag); + void clear(const std::optional & tag); size_t sizeInBytes() const; size_t count() const; diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index ef5539f0b13..cfcaf437510 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -3228,20 +3228,12 @@ QueryCachePtr Context::getQueryCache() const return shared->query_cache; } -void Context::clearQueryCache() const +void Context::clearQueryCache(const std::optional & tag) const { std::lock_guard lock(shared->mutex); if (shared->query_cache) - shared->query_cache->clear(); -} - -void Context::clearQueryCacheWithTag(const String & tag) const -{ - std::lock_guard lock(shared->mutex); - - if (shared->query_cache) - shared->query_cache->clearWithTag(tag); + shared->query_cache->clear(tag); } void Context::clearCaches() const diff --git a/src/Interpreters/Context.h b/src/Interpreters/Context.h index 076bff462f9..e0c69471e60 100644 --- a/src/Interpreters/Context.h +++ b/src/Interpreters/Context.h @@ -1068,8 +1068,7 @@ public: void setQueryCache(size_t max_size_in_bytes, size_t max_entries, size_t max_entry_size_in_bytes, size_t max_entry_size_in_rows); void updateQueryCacheConfiguration(const Poco::Util::AbstractConfiguration & config); std::shared_ptr getQueryCache() const; - void clearQueryCache() const; - void clearQueryCacheWithTag(const String & tag) const; + void clearQueryCache(const std::optional & tag) const; /** Clear the caches of the uncompressed blocks and marks. * This is usually done when renaming tables, changing the type of columns, deleting a table. diff --git a/src/Interpreters/InterpreterSystemQuery.cpp b/src/Interpreters/InterpreterSystemQuery.cpp index c7540084451..d4e2f22036c 100644 --- a/src/Interpreters/InterpreterSystemQuery.cpp +++ b/src/Interpreters/InterpreterSystemQuery.cpp @@ -371,7 +371,7 @@ BlockIO InterpreterSystemQuery::execute() case Type::DROP_QUERY_CACHE: { getContext()->checkAccess(AccessType::SYSTEM_DROP_QUERY_CACHE); - !query.query_cache_tag.empty() ? getContext()->clearQueryCacheWithTag(query.query_cache_tag): getContext()->clearQueryCache(); + getContext()->clearQueryCache(query.query_cache_tag); break; } diff --git a/src/Parsers/ASTSystemQuery.h b/src/Parsers/ASTSystemQuery.h index 739d71678df..d9f5b425182 100644 --- a/src/Parsers/ASTSystemQuery.h +++ b/src/Parsers/ASTSystemQuery.h @@ -129,9 +129,10 @@ public: String storage_policy; String volume; String disk; - String query_cache_tag; UInt64 seconds{}; + std::optional query_cache_tag; + String filesystem_cache_name; std::string key_to_drop; std::optional offset_to_drop; diff --git a/src/Parsers/CommonParsers.h b/src/Parsers/CommonParsers.h index a079fba5d53..ab0e70eb0e5 100644 --- a/src/Parsers/CommonParsers.h +++ b/src/Parsers/CommonParsers.h @@ -470,8 +470,8 @@ namespace DB MR_MACROS(TABLE_OVERRIDE, "TABLE OVERRIDE") \ MR_MACROS(TABLE, "TABLE") \ MR_MACROS(TABLES, "TABLES") \ - MR_MACROS(TAGS, "TAGS") \ MR_MACROS(TAG, "TAG") \ + MR_MACROS(TAGS, "TAGS") \ MR_MACROS(TAGS_INNER_UUID, "TAGS INNER UUID") \ MR_MACROS(TEMPORARY_TABLE, "TEMPORARY TABLE") \ MR_MACROS(TEMPORARY, "TEMPORARY") \ diff --git a/src/Parsers/ParserSystemQuery.cpp b/src/Parsers/ParserSystemQuery.cpp index 95d89e6fde1..af84dd10bfa 100644 --- a/src/Parsers/ParserSystemQuery.cpp +++ b/src/Parsers/ParserSystemQuery.cpp @@ -471,6 +471,16 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & res->seconds = seconds->as()->value.safeGet(); break; } + case Type::DROP_QUERY_CACHE: + { + ParserLiteral tag_parser; + ASTPtr ast; + if (ParserKeyword{Keyword::TAG}.ignore(pos, expected) && tag_parser.parse(pos, ast, expected)) + res->query_cache_tag = std::make_optional(ast->as()->value.safeGet()); + if (!parseQueryWithOnCluster(res, pos, expected)) + return false; + break; + } case Type::DROP_FILESYSTEM_CACHE: { ParserLiteral path_parser; @@ -489,16 +499,6 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & return false; break; } - case Type::DROP_QUERY_CACHE: - { - ParserLiteral tag_parser; - ASTPtr ast; - if (ParserKeyword{Keyword::TAG}.ignore(pos, expected) && tag_parser.parse(pos, ast, expected)) - res->query_cache_tag = ast->as()->value.safeGet(); - if (!parseQueryWithOnCluster(res, pos, expected)) - return false; - break; - } case Type::SYNC_FILESYSTEM_CACHE: { ParserLiteral path_parser; diff --git a/tests/queries/0_stateless/02494_query_cache_drop_cache.reference b/tests/queries/0_stateless/02494_query_cache_drop_cache.reference index 2f1465d1598..6481b5e0770 100644 --- a/tests/queries/0_stateless/02494_query_cache_drop_cache.reference +++ b/tests/queries/0_stateless/02494_query_cache_drop_cache.reference @@ -1,3 +1,17 @@ +Cache query result in query cache 1 1 +DROP entries with a certain tag, no entry will match +1 +After a full DROP, the cache is empty now +0 +Cache query result with different or no tag in query cache +1 +1 +1 +2 +4 +DROP entries with certain tags +2 +1 0 diff --git a/tests/queries/0_stateless/02494_query_cache_drop_cache.sql b/tests/queries/0_stateless/02494_query_cache_drop_cache.sql index bc2e7f442fc..3d064169a4e 100644 --- a/tests/queries/0_stateless/02494_query_cache_drop_cache.sql +++ b/tests/queries/0_stateless/02494_query_cache_drop_cache.sql @@ -4,10 +4,31 @@ -- (it's silly to use what will be tested below but we have to assume other tests cluttered the query cache) SYSTEM DROP QUERY CACHE; --- Cache query result in query cache +SELECT 'Cache query result in query cache'; SELECT 1 SETTINGS use_query_cache = true; SELECT count(*) FROM system.query_cache; --- No query results are cached after DROP +SELECT 'DROP entries with a certain tag, no entry will match'; +SYSTEM DROP QUERY CACHE TAG 'tag'; +SELECT count(*) FROM system.query_cache; + +SELECT 'After a full DROP, the cache is empty now'; SYSTEM DROP QUERY CACHE; SELECT count(*) FROM system.query_cache; + +-- More tests for DROP with tags: + +SELECT 'Cache query result with different or no tag in query cache'; +SELECT 1 SETTINGS use_query_cache = true; +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'abc'; +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'def'; +SELECT 2 SETTINGS use_query_cache = true; +SELECT count(*) FROM system.query_cache; + +SELECT 'DROP entries with certain tags'; +SYSTEM DROP QUERY CACHE TAG ''; +SELECT count(*) FROM system.query_cache; +SYSTEM DROP QUERY CACHE TAG 'def'; +SELECT count(*) FROM system.query_cache; +SYSTEM DROP QUERY CACHE TAG 'abc'; +SELECT count(*) FROM system.query_cache; From d10e65291ccea45e46aab5edcc21ee047de6d5fb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?= Date: Tue, 2 Jul 2024 10:32:51 +0000 Subject: [PATCH 31/56] Fix docs --- docs/en/operations/settings/settings.md | 12 +++++++++++- src/Storages/MergeTree/MergeTreeSettings.h | 2 +- 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index 19db4be17db..3617e6a3167 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -2855,7 +2855,7 @@ The minimum chunk size in bytes, which each thread will parse in parallel. ## merge_selecting_sleep_ms {#merge_selecting_sleep_ms} -Sleep time for merge selecting when no part is selected. A lower setting triggers selecting tasks in `background_schedule_pool` frequently, which results in a large number of requests to ClickHouse Keeper in large-scale clusters. +Minimum sleep time for merge selecting when no part is selected. A lower setting triggers selecting tasks in `background_schedule_pool` frequently, which results in a large number of requests to ClickHouse Keeper in large-scale clusters. Possible values: @@ -2863,6 +2863,16 @@ Possible values: Default value: `5000`. +## max_merge_selecting_sleep_ms + +Maximum sleep time for merge selecting when no part is selected. A lower setting triggers selecting tasks in `background_schedule_pool` frequently, which results in a large number of requests to ClickHouse Keeper in large-scale clusters. + +Possible values: + +- Any positive integer. + +Default value: `60000`. + ## parallel_distributed_insert_select {#parallel_distributed_insert_select} Enables parallel distributed `INSERT ... SELECT` query. diff --git a/src/Storages/MergeTree/MergeTreeSettings.h b/src/Storages/MergeTree/MergeTreeSettings.h index de1f0f60cfc..676f776df54 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.h +++ b/src/Storages/MergeTree/MergeTreeSettings.h @@ -67,7 +67,7 @@ struct Settings; M(Bool, fsync_part_directory, false, "Do fsync for part directory after all part operations (writes, renames, etc.).", 0) \ M(UInt64, non_replicated_deduplication_window, 0, "How many last blocks of hashes should be kept on disk (0 - disabled).", 0) \ M(UInt64, max_parts_to_merge_at_once, 100, "Max amount of parts which can be merged at once (0 - disabled). Doesn't affect OPTIMIZE FINAL query.", 0) \ - M(UInt64, merge_selecting_sleep_ms, 5000, "Maximum sleep time for merge selecting, a lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ + M(UInt64, merge_selecting_sleep_ms, 5000, "Minimum sleep time for merge selecting, a lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ M(UInt64, max_merge_selecting_sleep_ms, 60000, "Maximum sleep time for merge selecting, a lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ M(Float, merge_selecting_sleep_slowdown_factor, 1.2f, "The sleep time for merge selecting task is multiplied by this factor when there's nothing to merge and divided when a merge was assigned", 0) \ M(UInt64, merge_tree_clear_old_temporary_directories_interval_seconds, 60, "The period of executing the clear old temporary directories operation in background.", 0) \ From e547875a9cdffc53bb601c1f9d32c2a6fb6af560 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A1nos=20Benjamin=20Antal?= Date: Fri, 5 Jul 2024 13:27:16 +0000 Subject: [PATCH 32/56] Improve wording --- docs/en/operations/settings/settings.md | 4 ++-- src/Storages/MergeTree/MergeTreeSettings.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index 3617e6a3167..7bd36ccd00f 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -2855,7 +2855,7 @@ The minimum chunk size in bytes, which each thread will parse in parallel. ## merge_selecting_sleep_ms {#merge_selecting_sleep_ms} -Minimum sleep time for merge selecting when no part is selected. A lower setting triggers selecting tasks in `background_schedule_pool` frequently, which results in a large number of requests to ClickHouse Keeper in large-scale clusters. +Minimum time to wait before trying to select parts to merge again after no parts were selected. A lower setting triggers selecting tasks in `background_schedule_pool` frequently, which results in a large number of requests to ClickHouse Keeper in large-scale clusters. Possible values: @@ -2865,7 +2865,7 @@ Default value: `5000`. ## max_merge_selecting_sleep_ms -Maximum sleep time for merge selecting when no part is selected. A lower setting triggers selecting tasks in `background_schedule_pool` frequently, which results in a large number of requests to ClickHouse Keeper in large-scale clusters. +Maximum time to wait before trying to select parts to merge again after no parts were selected. A lower setting triggers selecting tasks in `background_schedule_pool` frequently, which results in a large number of requests to ClickHouse Keeper in large-scale clusters. Possible values: diff --git a/src/Storages/MergeTree/MergeTreeSettings.h b/src/Storages/MergeTree/MergeTreeSettings.h index 676f776df54..0769b60dc6b 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.h +++ b/src/Storages/MergeTree/MergeTreeSettings.h @@ -67,8 +67,8 @@ struct Settings; M(Bool, fsync_part_directory, false, "Do fsync for part directory after all part operations (writes, renames, etc.).", 0) \ M(UInt64, non_replicated_deduplication_window, 0, "How many last blocks of hashes should be kept on disk (0 - disabled).", 0) \ M(UInt64, max_parts_to_merge_at_once, 100, "Max amount of parts which can be merged at once (0 - disabled). Doesn't affect OPTIMIZE FINAL query.", 0) \ - M(UInt64, merge_selecting_sleep_ms, 5000, "Minimum sleep time for merge selecting, a lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ - M(UInt64, max_merge_selecting_sleep_ms, 60000, "Maximum sleep time for merge selecting, a lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ + M(UInt64, merge_selecting_sleep_ms, 5000, "Minimum time to wait before trying to select parts to merge again after no parts were selected. A lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ + M(UInt64, max_merge_selecting_sleep_ms, 60000, "Maximum time to wait before trying to select parts to merge again after no parts were selected. A lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ M(Float, merge_selecting_sleep_slowdown_factor, 1.2f, "The sleep time for merge selecting task is multiplied by this factor when there's nothing to merge and divided when a merge was assigned", 0) \ M(UInt64, merge_tree_clear_old_temporary_directories_interval_seconds, 60, "The period of executing the clear old temporary directories operation in background.", 0) \ M(UInt64, merge_tree_clear_old_parts_interval_seconds, 1, "The period of executing the clear old parts operation in background.", 0) \ From 9fd9f649db6bd3149cda217a200cfee3d7bc8238 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Tue, 20 Aug 2024 21:53:42 +0200 Subject: [PATCH 33/56] Add comment in IDataType.cpp --- src/DataTypes/IDataType.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/DataTypes/IDataType.cpp b/src/DataTypes/IDataType.cpp index d14461a4ff1..1a274c7f993 100644 --- a/src/DataTypes/IDataType.cpp +++ b/src/DataTypes/IDataType.cpp @@ -150,6 +150,7 @@ std::unique_ptr IDataType::getSubcolumnData( ISerialization::EnumerateStreamsSettings settings; settings.position_independent_encoding = false; + /// Don't enumerate dynamic subcolumns, they are handled separately. settings.enumerate_dynamic_streams = false; data.serialization->enumerateStreams(settings, callback_with_data, data); From 5d280053c2512332bc0c171e57dbc36cb3e0b675 Mon Sep 17 00:00:00 2001 From: Kruglov Pavel <48961922+Avogar@users.noreply.github.com> Date: Tue, 20 Aug 2024 21:55:10 +0200 Subject: [PATCH 34/56] Add comment in ISerialization.h --- src/DataTypes/Serializations/ISerialization.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/DataTypes/Serializations/ISerialization.h b/src/DataTypes/Serializations/ISerialization.h index d19cb0dd365..33575a07177 100644 --- a/src/DataTypes/Serializations/ISerialization.h +++ b/src/DataTypes/Serializations/ISerialization.h @@ -241,6 +241,9 @@ public: { SubstreamPath path; bool position_independent_encoding = true; + /// If set to false, don't enumerate dynamic subcolumns + /// (such as dynamic types in Dynamic column or dynamic paths in JSON column). + /// It may be needed when dynamic subcolumns are processed separately. bool enumerate_dynamic_streams = true; }; From e01a448bcc62a7e292766cddc0c817b9e44558d4 Mon Sep 17 00:00:00 2001 From: Zhigao Hong Date: Wed, 21 Aug 2024 15:35:33 +0800 Subject: [PATCH 35/56] Fix invalid characters in replica_name --- src/Storages/MergeTree/registerStorageMergeTree.cpp | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/Storages/MergeTree/registerStorageMergeTree.cpp b/src/Storages/MergeTree/registerStorageMergeTree.cpp index 44548e33d46..9a65d590453 100644 --- a/src/Storages/MergeTree/registerStorageMergeTree.cpp +++ b/src/Storages/MergeTree/registerStorageMergeTree.cpp @@ -538,6 +538,9 @@ static StoragePtr create(const StorageFactory::Arguments & args) if (replica_name.empty()) throw Exception(ErrorCodes::NO_REPLICA_NAME_GIVEN, "No replica name in config{}", verbose_help_message); + // '\t' and '\n' will interrupt parsing 'source replica' in ReplicatedMergeTreeLogEntryData::readText + if (replica_name.find('\t') != String::npos || replica_name.find('\n') != String::npos) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Replica name must not contain '\\t' or '\\n'"); arg_cnt = engine_args.size(); /// Update `arg_cnt` here because extractZooKeeperPathAndReplicaNameFromEngineArgs() could add arguments. arg_num = 2; /// zookeeper_path and replica_name together are always two arguments. From e87de3cfcd22870bf7aea3dfaf1607b180b2b1d8 Mon Sep 17 00:00:00 2001 From: Anton Popov Date: Wed, 21 Aug 2024 15:19:29 +0000 Subject: [PATCH 36/56] return back virtual columns to distributed tables --- src/Storages/StorageDistributed.cpp | 4 ++++ .../03228_virtual_column_merge_dist.reference | 8 +++++++ .../03228_virtual_column_merge_dist.sql | 24 +++++++++++++++++++ 3 files changed, 36 insertions(+) create mode 100644 tests/queries/0_stateless/03228_virtual_column_merge_dist.reference create mode 100644 tests/queries/0_stateless/03228_virtual_column_merge_dist.sql diff --git a/src/Storages/StorageDistributed.cpp b/src/Storages/StorageDistributed.cpp index e146e95f89f..c4668159759 100644 --- a/src/Storages/StorageDistributed.cpp +++ b/src/Storages/StorageDistributed.cpp @@ -290,6 +290,10 @@ VirtualColumnsDescription StorageDistributed::createVirtuals() desc.addEphemeral("_shard_num", std::make_shared(), "Deprecated. Use function shardNum instead"); + /// Add virtual columns from table of storage Merges. + 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"); + return desc; } diff --git a/tests/queries/0_stateless/03228_virtual_column_merge_dist.reference b/tests/queries/0_stateless/03228_virtual_column_merge_dist.reference new file mode 100644 index 00000000000..28f00bafdfe --- /dev/null +++ b/tests/queries/0_stateless/03228_virtual_column_merge_dist.reference @@ -0,0 +1,8 @@ +1 t_local_1 +2 t_local_2 +1 t_local_1 +2 t_local_2 +1 1 +2 1 +1 1 +2 1 diff --git a/tests/queries/0_stateless/03228_virtual_column_merge_dist.sql b/tests/queries/0_stateless/03228_virtual_column_merge_dist.sql new file mode 100644 index 00000000000..caf00a2e407 --- /dev/null +++ b/tests/queries/0_stateless/03228_virtual_column_merge_dist.sql @@ -0,0 +1,24 @@ +DROP TABLE IF EXISTS t_local_1; +DROP TABLE IF EXISTS t_local_2; +DROP TABLE IF EXISTS t_merge; +DROP TABLE IF EXISTS t_distr; + +CREATE TABLE t_local_1 (a UInt32) ENGINE = MergeTree ORDER BY a; +CREATE TABLE t_local_2 (a UInt32) ENGINE = MergeTree ORDER BY a; + +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()); + +SELECT a, _table FROM t_merge ORDER BY a; +SELECT a, _table FROM t_distr ORDER BY a; + +SELECT a, _database = currentDatabase() FROM t_merge ORDER BY a; +SELECT a, _database = currentDatabase() FROM t_distr ORDER BY a; + +DROP TABLE IF EXISTS t_local_1; +DROP TABLE IF EXISTS t_local_2; +DROP TABLE IF EXISTS t_merge; +DROP TABLE IF EXISTS t_distr; From d350f7bc1ab9e9454f7d84a1cc27176f3790be9b Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 21 Aug 2024 20:20:32 +0000 Subject: [PATCH 37/56] Improve accuracy of ICU-correcting code by 12.6% --- src/Functions/LowerUpperUTF8Impl.h | 11 ++++++----- .../0_stateless/00170_lower_upper_utf8.reference | 1 + tests/queries/0_stateless/00170_lower_upper_utf8.sql | 3 +++ 3 files changed, 10 insertions(+), 5 deletions(-) diff --git a/src/Functions/LowerUpperUTF8Impl.h b/src/Functions/LowerUpperUTF8Impl.h index 5da085f48e5..8469bedde0c 100644 --- a/src/Functions/LowerUpperUTF8Impl.h +++ b/src/Functions/LowerUpperUTF8Impl.h @@ -6,7 +6,6 @@ #include #include -#include #include #include @@ -43,7 +42,7 @@ struct LowerUpperUTF8Impl String output; size_t curr_offset = 0; - for (size_t i = 0; i < offsets.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const auto * data_start = reinterpret_cast(&data[offsets[i - 1]]); size_t size = offsets[i] - offsets[i - 1]; @@ -57,13 +56,15 @@ struct LowerUpperUTF8Impl output.clear(); input.toUTF8String(output); - /// For valid UTF-8 input strings, ICU sometimes produces output with extra '\0's at the end. Only the data before the first + /// 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 /// case, the behavior is also reasonable. - const char * res_end = find_last_not_symbols_or_null<'\0'>(output.data(), output.data() + output.size()); - size_t valid_size = res_end ? res_end - output.data() + 1 : 0; + size_t valid_size = output.size(); + if (!output.empty() && output.back() == '\0') + --valid_size; res_data.resize(curr_offset + valid_size + 1); + memcpy(&res_data[curr_offset], output.data(), valid_size); res_data[curr_offset + valid_size] = 0; diff --git a/tests/queries/0_stateless/00170_lower_upper_utf8.reference b/tests/queries/0_stateless/00170_lower_upper_utf8.reference index 3c644f22b9b..b1cb9ad5b57 100644 --- a/tests/queries/0_stateless/00170_lower_upper_utf8.reference +++ b/tests/queries/0_stateless/00170_lower_upper_utf8.reference @@ -26,3 +26,4 @@ 1 1 1 +2 diff --git a/tests/queries/0_stateless/00170_lower_upper_utf8.sql b/tests/queries/0_stateless/00170_lower_upper_utf8.sql index 85b6c5c6095..7c7bbac0df3 100644 --- a/tests/queries/0_stateless/00170_lower_upper_utf8.sql +++ b/tests/queries/0_stateless/00170_lower_upper_utf8.sql @@ -38,3 +38,6 @@ select lowerUTF8('ır') = 'ır'; -- German language select upper('öäüß') = 'öäüß'; select lower('ÖÄÜẞ') = 'ÖÄÜẞ'; + +-- Bug 68680 +SELECT lengthUTF8(lowerUTF8('Ä\0')); From dbee1542b6541e6aa1634706a5ce1a681fc717c2 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 21 Aug 2024 23:54:19 +0200 Subject: [PATCH 38/56] It --> 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 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 39/56] 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 40/56] 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 41/56] 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 add4718634317304f652579a9f201c3b81c96a7d Mon Sep 17 00:00:00 2001 From: Tanya Bragin Date: Thu, 22 Aug 2024 06:37:27 -0700 Subject: [PATCH 42/56] 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 43/56] 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 44/56] 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 5340ac5fbc7fba75d6a743d345c0f79dc466df0b Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 14:39:19 +0000 Subject: [PATCH 45/56] 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 0dc18247df3a290b4fb312325ff3b2a44a3f8357 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 15:10:24 +0000 Subject: [PATCH 46/56] 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 51fbc629c6dff4653e687228b0507947516072bb Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 15:42:17 +0000 Subject: [PATCH 47/56] 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 48/56] 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 49/56] 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 50/56] 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 0bd8ebf62616ce882b0ebc46945c837a5a91ba44 Mon Sep 17 00:00:00 2001 From: Tyler Hannan Date: Thu, 22 Aug 2024 17:58:56 +0200 Subject: [PATCH 51/56] 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 1692360233593e635c5a7797847bdfd8a0ffa33e Mon Sep 17 00:00:00 2001 From: Tyler Hannan Date: Thu, 22 Aug 2024 18:12:38 +0200 Subject: [PATCH 52/56] 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 53/56] 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 f89193fa416cc333f549d72bb8ba453907edc951 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 22 Aug 2024 19:12:19 +0000 Subject: [PATCH 54/56] 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 4c790999eb6ad74e3a8f99c072dcc12c956a63d8 Mon Sep 17 00:00:00 2001 From: Max Kainov Date: Fri, 23 Aug 2024 02:18:26 +0200 Subject: [PATCH 55/56] 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 e5380806653f8d391c6e88664b0096c3c51240f5 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 23 Aug 2024 07:09:03 +0000 Subject: [PATCH 56/56] 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