Fix style, add docs

docs/en/interfaces/formats.md

@@ -1973,6 +1973,7 @@ To exchange data with Hadoop, you can use [HDFS table engine](/docs/en/engines/t
 - [input_format_parquet_skip_columns_with_unsupported_types_in_schema_inference](/docs/en/operations/settings/settings-formats.md/#input_format_parquet_skip_columns_with_unsupported_types_in_schema_inference) - allow skipping columns with unsupported types while schema inference for Parquet format. Default value - `false`.
 - [output_format_parquet_fixed_string_as_fixed_byte_array](/docs/en/operations/settings/settings-formats.md/#output_format_parquet_fixed_string_as_fixed_byte_array) - use Parquet FIXED_LENGTH_BYTE_ARRAY type instead of Binary/String for FixedString columns. Default value - `true`.
 - [output_format_parquet_version](/docs/en/operations/settings/settings-formats.md/#output_format_parquet_version) - The version of Parquet format used in output format. Default value - `2.latest`.
+- [output_format_parquet_compression_method](/docs/en/operations/settings/settings-formats.md/#output_format_parquet_compression_method) - compression method used in output Parquet format. Default value - `snappy`.
 
 ## Arrow {#data-format-arrow}
 
@@ -2041,6 +2042,7 @@ $ clickhouse-client --query="SELECT * FROM {some_table} FORMAT Arrow" > {filenam
 - [input_format_arrow_allow_missing_columns](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_allow_missing_columns) - allow missing columns while reading Arrow data. Default value - `false`.
 - [input_format_arrow_skip_columns_with_unsupported_types_in_schema_inference](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_skip_columns_with_unsupported_types_in_schema_inference) - allow skipping columns with unsupported types while schema inference for Arrow format. Default value - `false`.
 - [output_format_arrow_fixed_string_as_fixed_byte_array](/docs/en/operations/settings/settings-formats.md/#output_format_arrow_fixed_string_as_fixed_byte_array) - use Arrow FIXED_SIZE_BINARY type instead of Binary/String for FixedString columns. Default value - `true`.
+- [output_format_arrow_compression_method](/docs/en/operations/settings/settings-formats.md/#output_format_arrow_compression_method) - compression method used in output Arrow format. Default value - `none`.
 
 ## ArrowStream {#data-format-arrow-stream}
 
@@ -2096,6 +2098,7 @@ $ clickhouse-client --query="SELECT * FROM {some_table} FORMAT ORC" > {filename.
 ### Arrow format settings {#parquet-format-settings}
 
 - [output_format_arrow_string_as_string](/docs/en/operations/settings/settings-formats.md/#output_format_arrow_string_as_string) - use Arrow String type instead of Binary for String columns. Default value - `false`.
+- [output_format_orc_compression_method](/docs/en/operations/settings/settings-formats.md/#output_format_orc_compression_method) - compression method used in output ORC format. Default value - `none`.
 - [input_format_arrow_import_nested](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_import_nested) - allow inserting array of structs into Nested table in Arrow input format. Default value - `false`.
 - [input_format_arrow_case_insensitive_column_matching](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_case_insensitive_column_matching) - ignore case when matching Arrow columns with ClickHouse columns. Default value - `false`.
 - [input_format_arrow_allow_missing_columns](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_allow_missing_columns) - allow missing columns while reading Arrow data. Default value - `false`.

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

@@ -1014,6 +1014,12 @@ Use Arrow FIXED_SIZE_BINARY type instead of Binary/String for FixedString column
 
 Enabled by default.
 
+### output_format_arrow_compression_method {#output_format_arrow_compression_method}
+
+Compression method used in output Arrow format. Supported codecs: `lz4_frame`, `zstd`, `none` (uncompressed)
+
+Default value: `none`.
+
 ## ORC format settings {#orc-format-settings}
 
 ### input_format_orc_import_nested {#input_format_orc_import_nested}
@@ -1057,6 +1063,12 @@ Use ORC String type instead of Binary for String columns.
 
 Disabled by default.
 
+### output_format_orc_compression_method {#output_format_orc_compression_method}
+
+Compression method used in output ORC format. Supported codecs: `lz4`, `snappy`, `zlib`, `zstd`, `none` (uncompressed)
+
+Default value: `none`.
+
 ## Parquet format settings {#parquet-format-settings}
 
 ### input_format_parquet_import_nested {#input_format_parquet_import_nested}
@@ -1112,6 +1124,12 @@ The version of Parquet format used in output format. Supported versions: `1.0`,
 
 Default value: `2.latest`.
 
+### output_format_parquet_compression_method {#output_format_parquet_compression_method}
+
+Compression method used in output Parquet format. Supported codecs: `snappy`, `lz4`, `brotli`, `zstd`, `gzip`, `none` (uncompressed)
+
+Default value: `snappy`.
+
 ## Hive format settings {#hive-format-settings}
 
 ### input_format_hive_text_fields_delimiter {#input_format_hive_text_fields_delimiter}

src/Processors/Formats/Impl/ArrowBlockOutputFormat.cpp

@@ -8,7 +8,6 @@
 #include <arrow/result.h>
 #include "ArrowBufferedStreams.h"
 #include "CHColumnToArrowColumn.h"
-#include "config.h"
 
 
 namespace DB
@@ -36,7 +35,6 @@ arrow::Compression::type getArrowCompression(FormatSettings::ArrowCompression me
 
 }
 
-
 ArrowBlockOutputFormat::ArrowBlockOutputFormat(WriteBuffer & out_, const Block & header_, bool stream_, const FormatSettings & format_settings_)
     : IOutputFormat(header_, out_)
     , stream{stream_}

src/Processors/Formats/Impl/ORCBlockOutputFormat.cpp

@@ -22,8 +22,6 @@
 #include <DataTypes/DataTypeMap.h>
 #include <DataTypes/DataTypeLowCardinality.h>
 
-#include "config.h"
-
 namespace DB
 {
 

src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp

@@ -14,6 +14,7 @@ namespace DB
 namespace ErrorCodes
 {
     extern const int UNKNOWN_EXCEPTION;
+    extern const int NOT_IMPLEMENTED;
 }
 
 namespace
@@ -61,7 +62,6 @@ parquet::Compression::type getParquetCompression(FormatSettings::ParquetCompress
     throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Unsupported compression method");
 }
 
-
 }
 
 ParquetBlockOutputFormat::ParquetBlockOutputFormat(WriteBuffer & out_, const Block & header_, const FormatSettings & format_settings_)

tests/queries/0_stateless/02581_parquet_arrow_orc_compressions.sh

@@ -1,4 +1,5 @@
 #!/usr/bin/env bash
+# Tags: no-fasttest
 
 CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 # shellcheck source=../shell_config.sh