Merge pull request #68298 from Avogar/fix-nullable-schema-inference

Fix using schema_inference_make_columns_nullable=0

docs/en/interfaces/schema-inference.md

@@ -1389,7 +1389,7 @@ DESC format(JSONEachRow, '{"id" : 1, "age" : 25, "name" : "Josh", "status" : nul
 #### schema_inference_make_columns_nullable
 
 Controls making inferred types `Nullable` in schema inference for formats without information about nullability.
-If the setting is enabled, all inferred type will be `Nullable`, if disabled, the inferred type will be `Nullable` only if `input_format_null_as_default` is disabled and the column contains `NULL` in a sample that is parsed during schema inference.
+If the setting is enabled, all inferred type will be `Nullable`, if disabled, the inferred type will never be `Nullable`, if set to `auto`, the inferred type will be `Nullable` only if the column contains `NULL` in a sample that is parsed during schema inference or file metadata contains information about column nullability.
 
 Enabled by default.
 
@@ -1412,15 +1412,13 @@ DESC format(JSONEachRow, $$
 └─────────┴─────────────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
 ```
 ```sql
-SET schema_inference_make_columns_nullable = 0;
+SET schema_inference_make_columns_nullable = 'auto';
-SET input_format_null_as_default = 0;    
 DESC format(JSONEachRow, $$
                                 {"id" :  1, "age" :  25, "name" : "Josh", "status" : null, "hobbies" : ["football", "cooking"]}
                                 {"id" :  2, "age" :  19, "name" :  "Alan", "status" : "married", "hobbies" :  ["tennis", "art"]}
                          $$)
 ```
 ```response
-
 ┌─name────┬─type─────────────┬─default_type─┬─default_expression─┬─comment─┬─codec_expression─┬─ttl_expression─┐
 │ id      │ Int64            │              │                    │         │                  │                │
 │ age     │ Int64            │              │                    │         │                  │                │
@@ -1432,7 +1430,6 @@ DESC format(JSONEachRow, $$
 
 ```sql
 SET schema_inference_make_columns_nullable = 0;
-SET input_format_null_as_default = 1;    
 DESC format(JSONEachRow, $$
                                 {"id" :  1, "age" :  25, "name" : "Josh", "status" : null, "hobbies" : ["football", "cooking"]}
                                 {"id" :  2, "age" :  19, "name" :  "Alan", "status" : "married", "hobbies" :  ["tennis", "art"]}

docs/en/operations/settings/settings-formats.md

@@ -171,8 +171,8 @@ If the `schema_inference_hints` is not formated properly, or if there is a typo
 
 ## schema_inference_make_columns_nullable {#schema_inference_make_columns_nullable}
 
-Controls making inferred types `Nullable` in schema inference for formats without information about nullability.
+Controls making inferred types `Nullable` in schema inference.
-If the setting is enabled, the inferred type will be `Nullable` only if column contains `NULL` in a sample that is parsed during schema inference.
+If the setting is enabled, all inferred type will be `Nullable`, if disabled, the inferred type will never be `Nullable`, if set to `auto`, the inferred type will be `Nullable` only if the column contains `NULL` in a sample that is parsed during schema inference or file metadata contains information about column nullability.
 
 Default value: `true`.
 

src/Core/Settings.h

@@ -1120,7 +1120,7 @@ class IColumn;
     M(String, column_names_for_schema_inference, "", "The list of column names to use in schema inference for formats without column names. The format: 'column1,column2,column3,...'", 0) \
     M(String, schema_inference_hints, "", "The list of column names and types to use in schema inference for formats without column names. The format: 'column_name1 column_type1, column_name2 column_type2, ...'", 0) \
     M(SchemaInferenceMode, schema_inference_mode, "default", "Mode of schema inference. 'default' - assume that all files have the same schema and schema can be inferred from any file, 'union' - files can have different schemas and the resulting schema should be the a union of schemas of all files", 0) \
-    M(Bool, schema_inference_make_columns_nullable, true, "If set to true, all inferred types will be Nullable in schema inference for formats without information about nullability.", 0) \
+    M(UInt64Auto, schema_inference_make_columns_nullable, 1, "If set to true, all inferred types will be Nullable in schema inference. When set to false, no columns will be converted to Nullable. When set to 'auto', ClickHouse will use information about nullability from the data.", 0) \
     M(Bool, input_format_json_read_bools_as_numbers, true, "Allow to parse bools as numbers in JSON input formats", 0) \
     M(Bool, input_format_json_read_bools_as_strings, true, "Allow to parse bools as strings in JSON input formats", 0) \
     M(Bool, input_format_json_try_infer_numbers_from_strings, false, "Try to infer numbers from string fields while schema inference", 0) \

src/Formats/FormatFactory.cpp

@@ -257,7 +257,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se
     format_settings.max_bytes_to_read_for_schema_inference = settings.input_format_max_bytes_to_read_for_schema_inference;
     format_settings.column_names_for_schema_inference = settings.column_names_for_schema_inference;
     format_settings.schema_inference_hints = settings.schema_inference_hints;
-    format_settings.schema_inference_make_columns_nullable = settings.schema_inference_make_columns_nullable;
+    format_settings.schema_inference_make_columns_nullable = settings.schema_inference_make_columns_nullable.valueOr(2);
     format_settings.mysql_dump.table_name = settings.input_format_mysql_dump_table_name;
     format_settings.mysql_dump.map_column_names = settings.input_format_mysql_dump_map_column_names;
     format_settings.sql_insert.max_batch_size = settings.output_format_sql_insert_max_batch_size;

src/Formats/FormatSettings.h

@@ -77,7 +77,7 @@ struct FormatSettings
         Raw
     };
 
-    bool schema_inference_make_columns_nullable = true;
+    UInt64 schema_inference_make_columns_nullable = 1;
 
     DateTimeOutputFormat date_time_output_format = DateTimeOutputFormat::Simple;
 

src/Formats/SchemaInferenceUtils.cpp

@@ -1344,7 +1344,11 @@ namespace
         if (checkCharCaseInsensitive('n', buf))
         {
             if (checkStringCaseInsensitive("ull", buf))
-                return std::make_shared<DataTypeNullable>(std::make_shared<DataTypeNothing>());
+            {
+                if (settings.schema_inference_make_columns_nullable == 0)
+                    return std::make_shared<DataTypeNothing>();
+                return makeNullable(std::make_shared<DataTypeNothing>());
+            }
             else if (checkStringCaseInsensitive("an", buf))
                 return std::make_shared<DataTypeFloat64>();
         }

src/Processors/Formats/ISchemaReader.cpp

@@ -54,13 +54,8 @@ void checkFinalInferredType(
         type = default_type;
     }
 
-    if (settings.schema_inference_make_columns_nullable)
+    if (settings.schema_inference_make_columns_nullable == 1)
         type = makeNullableRecursively(type);
-    /// In case when data for some column could contain nulls and regular values,
-    /// resulting inferred type is Nullable.
-    /// If input_format_null_as_default is enabled, we should remove Nullable type.
-    else if (settings.null_as_default)
-        type = removeNullable(type);
 }
 
 void ISchemaReader::transformTypesIfNeeded(DB::DataTypePtr & type, DB::DataTypePtr & new_type)

src/Processors/Formats/Impl/ArrowBlockInputFormat.cpp

@@ -204,8 +204,11 @@ NamesAndTypesList ArrowSchemaReader::readSchema()
         schema = file_reader->schema();
 
     auto header = ArrowColumnToCHColumn::arrowSchemaToCHHeader(
-        *schema, stream ? "ArrowStream" : "Arrow", format_settings.arrow.skip_columns_with_unsupported_types_in_schema_inference);
+        *schema,
-    if (format_settings.schema_inference_make_columns_nullable)
+        stream ? "ArrowStream" : "Arrow",
+        format_settings.arrow.skip_columns_with_unsupported_types_in_schema_inference,
+        format_settings.schema_inference_make_columns_nullable != 0);
+    if (format_settings.schema_inference_make_columns_nullable == 1)
         return getNamesAndRecursivelyNullableTypes(header);
     return header.getNamesAndTypesList();
 }

src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp

@@ -727,6 +727,7 @@ struct ReadColumnFromArrowColumnSettings
     FormatSettings::DateTimeOverflowBehavior date_time_overflow_behavior;
     bool allow_arrow_null_type;
     bool skip_columns_with_unsupported_types;
+    bool allow_inferring_nullable_columns;
 };
 
 static ColumnWithTypeAndName readColumnFromArrowColumn(
@@ -1109,7 +1110,7 @@ static ColumnWithTypeAndName readColumnFromArrowColumn(
     bool is_map_nested_column,
     const ReadColumnFromArrowColumnSettings & settings)
 {
-    bool read_as_nullable_column = arrow_column->null_count() || is_nullable_column || (type_hint && type_hint->isNullable());
+    bool read_as_nullable_column = (arrow_column->null_count() || is_nullable_column || (type_hint && type_hint->isNullable())) && settings.allow_inferring_nullable_columns;
     if (read_as_nullable_column &&
         arrow_column->type()->id() != arrow::Type::LIST &&
         arrow_column->type()->id() != arrow::Type::LARGE_LIST &&
@@ -1173,14 +1174,16 @@ static std::shared_ptr<arrow::ChunkedArray> createArrowColumn(const std::shared_
 Block ArrowColumnToCHColumn::arrowSchemaToCHHeader(
     const arrow::Schema & schema,
     const std::string & format_name,
-    bool skip_columns_with_unsupported_types)
+    bool skip_columns_with_unsupported_types,
+    bool allow_inferring_nullable_columns)
 {
     ReadColumnFromArrowColumnSettings settings
     {
         .format_name = format_name,
         .date_time_overflow_behavior = FormatSettings::DateTimeOverflowBehavior::Ignore,
         .allow_arrow_null_type = false,
-        .skip_columns_with_unsupported_types = skip_columns_with_unsupported_types
+        .skip_columns_with_unsupported_types = skip_columns_with_unsupported_types,
+        .allow_inferring_nullable_columns = allow_inferring_nullable_columns,
     };
 
     ColumnsWithTypeAndName sample_columns;
@@ -1254,7 +1257,8 @@ Chunk ArrowColumnToCHColumn::arrowColumnsToCHChunk(const NameToArrowColumn & nam
         .format_name = format_name,
         .date_time_overflow_behavior = date_time_overflow_behavior,
         .allow_arrow_null_type = true,
-        .skip_columns_with_unsupported_types = false
+        .skip_columns_with_unsupported_types = false,
+        .allow_inferring_nullable_columns = true
     };
 
     Columns columns;

src/Processors/Formats/Impl/ArrowColumnToCHColumn.h

@@ -34,7 +34,8 @@ public:
     static Block arrowSchemaToCHHeader(
         const arrow::Schema & schema,
         const std::string & format_name,
-        bool skip_columns_with_unsupported_types = false);
+        bool skip_columns_with_unsupported_types = false,
+        bool allow_inferring_nullable_columns = true);
 
     struct DictionaryInfo
     {

src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp

@@ -1002,7 +1002,7 @@ NamesAndTypesList NativeORCSchemaReader::readSchema()
             header.insert(ColumnWithTypeAndName{type, name});
     }
 
-    if (format_settings.schema_inference_make_columns_nullable)
+    if (format_settings.schema_inference_make_columns_nullable == 1)
         return getNamesAndRecursivelyNullableTypes(header);
     return header.getNamesAndTypesList();
 }

src/Processors/Formats/Impl/ORCBlockInputFormat.cpp

@@ -160,8 +160,11 @@ NamesAndTypesList ORCSchemaReader::readSchema()
 {
     initializeIfNeeded();
     auto header = ArrowColumnToCHColumn::arrowSchemaToCHHeader(
-        *schema, "ORC", format_settings.orc.skip_columns_with_unsupported_types_in_schema_inference);
+        *schema,
-    if (format_settings.schema_inference_make_columns_nullable)
+        "ORC",
+        format_settings.orc.skip_columns_with_unsupported_types_in_schema_inference,
+        format_settings.schema_inference_make_columns_nullable != 0);
+    if (format_settings.schema_inference_make_columns_nullable == 1)
         return getNamesAndRecursivelyNullableTypes(header);
     return header.getNamesAndTypesList();
 }

src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp

@@ -869,8 +869,11 @@ NamesAndTypesList ParquetSchemaReader::readSchema()
     THROW_ARROW_NOT_OK(parquet::arrow::FromParquetSchema(metadata->schema(), &schema));
 
     auto header = ArrowColumnToCHColumn::arrowSchemaToCHHeader(
-        *schema, "Parquet", format_settings.parquet.skip_columns_with_unsupported_types_in_schema_inference);
+        *schema,
-    if (format_settings.schema_inference_make_columns_nullable)
+        "Parquet",
+        format_settings.parquet.skip_columns_with_unsupported_types_in_schema_inference,
+        format_settings.schema_inference_make_columns_nullable != 0);
+    if (format_settings.schema_inference_make_columns_nullable == 1)
         return getNamesAndRecursivelyNullableTypes(header);
     return header.getNamesAndTypesList();
 }

tests/queries/0_stateless/02497_schema_inference_nulls.sql

@@ -18,7 +18,7 @@ desc format(JSONEachRow, '{"x" : [[], [null], [1, 2, 3]]}');
 desc format(JSONEachRow, '{"x" : [{"a" : null}, {"b" : 1}]}');
 desc format(JSONEachRow, '{"x" : [["2020-01-01", null, "1234"], ["abcd"]]}');
 
-set schema_inference_make_columns_nullable=0;
+set schema_inference_make_columns_nullable='auto';
 desc format(JSONEachRow, '{"x" : [1, 2]}');
 desc format(JSONEachRow, '{"x" : [null, 1]}');
 desc format(JSONEachRow, '{"x" : [1, 2]}, {"x" : [3]}');
@@ -40,7 +40,7 @@ desc format(JSONCompactEachRow, '[[[], [null], [1, 2, 3]]]');
 desc format(JSONCompactEachRow, '[[{"a" : null}, {"b" : 1}]]');
 desc format(JSONCompactEachRow, '[[["2020-01-01", null, "1234"], ["abcd"]]]');
 
-set schema_inference_make_columns_nullable=0;
+set schema_inference_make_columns_nullable='auto';
 desc format(JSONCompactEachRow, '[[1, 2]]');
 desc format(JSONCompactEachRow, '[[null, 1]]');
 desc format(JSONCompactEachRow, '[[1, 2]], [[3]]');
@@ -59,7 +59,7 @@ desc format(CSV, '"[[], [null], [1, 2, 3]]"');
 desc format(CSV, '"[{\'a\' : null}, {\'b\' : 1}]"');
 desc format(CSV, '"[[\'2020-01-01\', null, \'1234\'], [\'abcd\']]"');
 
-set schema_inference_make_columns_nullable=0;
+set schema_inference_make_columns_nullable='auto';
 desc format(CSV, '"[1,2]"');
 desc format(CSV, '"[NULL, 1]"');
 desc format(CSV, '"[1, 2]"\n"[3]"');

tests/queries/0_stateless/02784_schema_inference_null_as_default.sql

@@ -1,7 +1,7 @@
 desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=1;
 select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=1;
-desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=0;
+desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable='auto', input_format_null_as_default=0;
-select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=0;
+select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable='auto', input_format_null_as_default=0;
 desc format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=1;
 select * from format(JSONEachRow, '{"x" : null}, {"x" : 42}') settings schema_inference_make_columns_nullable=0, input_format_null_as_default=1;
 

tests/queries/0_stateless/03036_parquet_arrow_nullable.reference

@@ -1,40 +1,66 @@
 Parquet
 a	UInt64					
 a_nullable	Nullable(UInt64)					
+a	UInt64					
+a_nullable	UInt64					
 Arrow
 a	UInt64					
 a_nullable	Nullable(UInt64)					
+a	UInt64					
+a_nullable	UInt64					
 Parquet
 b	Array(UInt64)					
 b_nullable	Array(Nullable(UInt64))					
+b	Array(UInt64)					
+b_nullable	Array(UInt64)					
 Arrow
 b	Array(Nullable(UInt64))					
 b_nullable	Array(Nullable(UInt64))					
+b	Array(UInt64)					
+b_nullable	Array(UInt64)					
 Parquet
 c	Tuple(\n    a UInt64,\n    b String)					
 c_nullable	Tuple(\n    a Nullable(UInt64),\n    b Nullable(String))					
+c	Tuple(\n    a UInt64,\n    b String)					
+c_nullable	Tuple(\n    a UInt64,\n    b String)					
 Arrow
 c	Tuple(\n    a UInt64,\n    b String)					
 c_nullable	Tuple(\n    a Nullable(UInt64),\n    b Nullable(String))					
+c	Tuple(\n    a UInt64,\n    b String)					
+c_nullable	Tuple(\n    a UInt64,\n    b String)					
 Parquet
 d	Tuple(\n    a UInt64,\n    b Tuple(\n        a UInt64,\n        b String),\n    d_nullable Tuple(\n        a UInt64,\n        b Tuple(\n            a Nullable(UInt64),\n            b Nullable(String))))					
+d	Tuple(\n    a UInt64,\n    b Tuple(\n        a UInt64,\n        b String),\n    d_nullable Tuple(\n        a UInt64,\n        b Tuple(\n            a UInt64,\n            b String)))					
 Arrow
 d	Tuple(\n    a UInt64,\n    b Tuple(\n        a UInt64,\n        b String),\n    d_nullable Tuple(\n        a UInt64,\n        b Tuple(\n            a Nullable(UInt64),\n            b Nullable(String))))					
+d	Tuple(\n    a UInt64,\n    b Tuple(\n        a UInt64,\n        b String),\n    d_nullable Tuple(\n        a UInt64,\n        b Tuple(\n            a UInt64,\n            b String)))					
 Parquet
 e	Map(UInt64, String)					
 e_nullable	Map(UInt64, Nullable(String))					
+e	Map(UInt64, String)					
+e_nullable	Map(UInt64, String)					
 Arrow
 e	Map(UInt64, Nullable(String))					
 e_nullable	Map(UInt64, Nullable(String))					
+e	Map(UInt64, String)					
+e_nullable	Map(UInt64, String)					
 Parquet
 f	Map(UInt64, Map(UInt64, String))					
 f_nullables	Map(UInt64, Map(UInt64, Nullable(String)))					
+f	Map(UInt64, Map(UInt64, String))					
+f_nullables	Map(UInt64, Map(UInt64, String))					
 Arrow
 f	Map(UInt64, Map(UInt64, Nullable(String)))					
 f_nullables	Map(UInt64, Map(UInt64, Nullable(String)))					
+f	Map(UInt64, Map(UInt64, String))					
+f_nullables	Map(UInt64, Map(UInt64, String))					
 Parquet
 g	String					
 g_nullable	Nullable(String)					
+g	String					
+g_nullable	String					
 Arrow
 g	LowCardinality(String)					
 g_nullable	LowCardinality(String)					
+g	LowCardinality(String)					
+g_nullable	LowCardinality(String)					

tests/queries/0_stateless/03036_parquet_arrow_nullable.sh

@@ -14,6 +14,7 @@ for format in $formats
 do
     echo $format
     $CLICKHOUSE_LOCAL -q "select * from generateRandom('a UInt64, a_nullable Nullable(UInt64)', 42) limit 10 format $format" > $DATA_FILE
+    $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'"
     $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0"
 done
 
@@ -21,6 +22,7 @@ for format in $formats
 do
     echo $format
     $CLICKHOUSE_LOCAL -q "select * from generateRandom('b Array(UInt64), b_nullable Array(Nullable(UInt64))', 42) limit 10 format $format" > $DATA_FILE
+    $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'"
     $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0"
 done
 
@@ -28,6 +30,7 @@ for format in $formats
 do
     echo $format
     $CLICKHOUSE_LOCAL -q "select * from generateRandom('c Tuple(a UInt64, b String), c_nullable Tuple(a Nullable(UInt64), b Nullable(String))', 42) limit 10 format $format" > $DATA_FILE
+    $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'"
     $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0"
 done
 
@@ -35,6 +38,7 @@ for format in $formats
 do
     echo $format
     $CLICKHOUSE_LOCAL -q "select * from generateRandom('d Tuple(a UInt64, b Tuple(a UInt64, b String), d_nullable Tuple(a UInt64, b Tuple(a Nullable(UInt64), b Nullable(String))))', 42) limit 10 format $format" > $DATA_FILE
+    $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'"
     $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0"
 done
 
@@ -42,6 +46,7 @@ for format in $formats
 do
     echo $format
     $CLICKHOUSE_LOCAL -q "select * from generateRandom('e Map(UInt64, String), e_nullable Map(UInt64, Nullable(String))', 42) limit 10 format $format" > $DATA_FILE
+    $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'"
     $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0"
 done
 
@@ -49,6 +54,7 @@ for format in $formats
 do
     echo $format
     $CLICKHOUSE_LOCAL -q "select * from generateRandom('f Map(UInt64, Map(UInt64, String)), f_nullables Map(UInt64, Map(UInt64, Nullable(String)))', 42) limit 10 format $format" > $DATA_FILE
+    $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'"
     $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0"
 done
 
@@ -56,6 +62,7 @@ for format in $formats
 do
     echo $format
     $CLICKHOUSE_LOCAL -q "select * from generateRandom('g LowCardinality(String), g_nullable LowCardinality(Nullable(String))', 42) limit 10 settings output_format_arrow_low_cardinality_as_dictionary=1, allow_suspicious_low_cardinality_types=1 format $format" > $DATA_FILE
+    $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 'auto'"
     $CLICKHOUSE_LOCAL -q "desc file('$DATA_FILE') SETTINGS schema_inference_make_columns_nullable = 0"
 done
 

tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference

@@ -1,2 +1,2 @@
 x	Nullable(Int64)					
-schema_inference_hints=, max_rows_to_read_for_schema_inference=25000, max_bytes_to_read_for_schema_inference=1000, schema_inference_make_columns_nullable=true, try_infer_integers=true, try_infer_dates=true, try_infer_datetimes=true, try_infer_datetimes_only_datetime64=false, try_infer_numbers_from_strings=false, read_bools_as_numbers=true, read_bools_as_strings=true, read_objects_as_strings=true, read_numbers_as_strings=true, read_arrays_as_strings=true, try_infer_objects_as_tuples=true, infer_incomplete_types_as_strings=true, try_infer_objects=false, use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects=false
+schema_inference_hints=, max_rows_to_read_for_schema_inference=25000, max_bytes_to_read_for_schema_inference=1000, schema_inference_make_columns_nullable=1, try_infer_integers=true, try_infer_dates=true, try_infer_datetimes=true, try_infer_datetimes_only_datetime64=false, try_infer_numbers_from_strings=false, read_bools_as_numbers=true, read_bools_as_strings=true, read_objects_as_strings=true, read_numbers_as_strings=true, read_arrays_as_strings=true, try_infer_objects_as_tuples=true, infer_incomplete_types_as_strings=true, try_infer_objects=false, use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects=false