[GitHub] [doris] zhangstar333 commented on a diff in pull request #12730: [Refactor](parquet) refactor parquet write to uniform and consistent logic

[GitBox]

zhangstar333 commented on code in PR #12730:
URL: https://github.com/apache/doris/pull/12730#discussion_r976116269


##########
be/src/exec/parquet_writer.cpp:
##########
@@ -40,6 +40,124 @@ ParquetOutputStream::~ParquetOutputStream() {
     }
 }
 
+void ParquetBuildHelper::build_schema_repetition_type(
+        parquet::Repetition::type& parquet_repetition_type,
+        const TParquetRepetitionType::type& column_repetition_type) {
+    switch (column_repetition_type) {
+    case TParquetRepetitionType::REQUIRED: {
+        parquet_repetition_type = parquet::Repetition::REQUIRED;
+        break;
+    }
+    case TParquetRepetitionType::REPEATED: {
+        parquet_repetition_type = parquet::Repetition::REPEATED;
+        break;
+    }
+    case TParquetRepetitionType::OPTIONAL: {
+        parquet_repetition_type = parquet::Repetition::OPTIONAL;
+        break;
+    }
+    default:
+        parquet_repetition_type = parquet::Repetition::UNDEFINED;
+    }
+}
+
+void ParquetBuildHelper::build_schema_data_type(parquet::Type::type& 
parquet_data_type,
+                                                const TParquetDataType::type& 
column_data_type) {
+    switch (column_data_type) {
+    case TParquetDataType::BOOLEAN: {
+        parquet_data_type = parquet::Type::BOOLEAN;
+        break;
+    }
+    case TParquetDataType::INT32: {
+        parquet_data_type = parquet::Type::INT32;
+        break;
+    }
+    case TParquetDataType::INT64: {
+        parquet_data_type = parquet::Type::INT64;
+        break;
+    }
+    case TParquetDataType::INT96: {
+        parquet_data_type = parquet::Type::INT96;
+        break;
+    }
+    case TParquetDataType::BYTE_ARRAY: {
+        parquet_data_type = parquet::Type::BYTE_ARRAY;
+        break;
+    }
+    case TParquetDataType::FLOAT: {
+        parquet_data_type = parquet::Type::FLOAT;
+        break;
+    }
+    case TParquetDataType::DOUBLE: {
+        parquet_data_type = parquet::Type::DOUBLE;
+        break;
+    }
+    case TParquetDataType::FIXED_LEN_BYTE_ARRAY: {
+        parquet_data_type = parquet::Type::FIXED_LEN_BYTE_ARRAY;
+        break;
+    }
+    default:
+        parquet_data_type = parquet::Type::UNDEFINED;
+    }
+}
+
+void ParquetBuildHelper::build_compression_type(
+        parquet::WriterProperties::Builder& builder,
+        const TParquetCompressionType::type& compression_type) {
+    switch (compression_type) {
+    case TParquetCompressionType::SNAPPY: {
+        builder.compression(parquet::Compression::SNAPPY);
+        break;
+    }
+    case TParquetCompressionType::GZIP: {
+        builder.compression(parquet::Compression::GZIP);
+        break;
+    }
+    case TParquetCompressionType::BROTLI: {
+        builder.compression(parquet::Compression::BROTLI);
+        break;
+    }
+    case TParquetCompressionType::ZSTD: {
+        builder.compression(parquet::Compression::ZSTD);
+        break;
+    }
+    case TParquetCompressionType::LZ4: {
+        builder.compression(parquet::Compression::LZ4);
+        break;
+    }
+    case TParquetCompressionType::LZO: {
+        builder.compression(parquet::Compression::LZO);
+        break;
+    }
+    case TParquetCompressionType::BZ2: {
+        builder.compression(parquet::Compression::BZ2);
+        break;
+    }
+    case TParquetCompressionType::UNCOMPRESSED: {
+        builder.compression(parquet::Compression::UNCOMPRESSED);
+        break;
+    }
+    default:
+        builder.compression(parquet::Compression::UNCOMPRESSED);
+    }
+}
+
+void ParquetBuildHelper::build_version(parquet::WriterProperties::Builder& 
builder,
+                                       const TParquetVersion::type& 
parquet_version) {
+    switch (parquet_version) {
+    case TParquetVersion::PARQUET_1_0: {
+        builder.version(parquet::ParquetVersion::PARQUET_1_0);
+        break;
+    }
+    case TParquetVersion::PARQUET_2_LATEST: {
+        builder.version(parquet::ParquetVersion::PARQUET_2_LATEST);
+        break;
+    }
+    default:
+        builder.version(parquet::ParquetVersion::PARQUET_1_0);

Review Comment:
   <img width="577" alt="image" 
src="https://user-images.githubusercontent.com/87313068/191435673-aa1baf79-15a6-4940-b5ab-1429e6ed354f.png";>
   
   before refactor, if haven't set version property, it will be PARQUET_1_0, 



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org
For additional commands, e-mail: commits-h...@doris.apache.org