zclllyybb commented on code in PR #47307: URL: https://github.com/apache/doris/pull/47307#discussion_r1938448935
########## be/src/vec/functions/function_compress.cpp: ########## @@ -0,0 +1,247 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. +#include <glog/logging.h> + +#include <cctype> +#include <cstddef> +#include <cstring> +#include <functional> +#include <memory> +#include <string> +#include <utility> + +#include "common/status.h" +#include "util/block_compression.h" +#include "util/faststring.h" +#include "vec/aggregate_functions/aggregate_function.h" +#include "vec/columns/column.h" +#include "vec/columns/column_nullable.h" +#include "vec/columns/column_string.h" +#include "vec/columns/column_vector.h" +#include "vec/columns/columns_number.h" +#include "vec/common/assert_cast.h" +#include "vec/core/block.h" +#include "vec/core/column_numbers.h" +#include "vec/core/column_with_type_and_name.h" +#include "vec/core/types.h" +#include "vec/data_types/data_type.h" +#include "vec/data_types/data_type_nullable.h" +#include "vec/data_types/data_type_number.h" +#include "vec/data_types/data_type_string.h" +#include "vec/functions/function.h" +#include "vec/functions/simple_function_factory.h" + +namespace doris { +class FunctionContext; +} // namespace doris + +namespace doris::vectorized { + +class FunctionCompress : public IFunction { + string hex_itoc = "0123456789ABCDEF"; + +public: + static constexpr auto name = "compress"; + static FunctionPtr create() { return std::make_shared<FunctionCompress>(); } + + String get_name() const override { return name; } + + size_t get_number_of_arguments() const override { return 1; } + + DataTypePtr get_return_type_impl(const DataTypes& arguments) const override { + return std::make_shared<DataTypeString>(); + } + + Status execute_impl(FunctionContext* context, Block& block, const ColumnNumbers& arguments, + uint32_t result, size_t input_rows_count) const override { + // Get the compression algorithm object + BlockCompressionCodec* compression_codec; + RETURN_IF_ERROR(get_block_compression_codec(segment_v2::CompressionTypePB::ZLIB, + &compression_codec)); + + const auto& arg_column = + assert_cast<const ColumnString&>(*block.get_by_position(arguments[0]).column); + auto result_column = ColumnString::create(); + + auto& arg_data = arg_column.get_chars(); + auto& arg_offset = arg_column.get_offsets(); + const char* arg_begin = reinterpret_cast<const char*>(arg_data.data()); + + auto& col_data = result_column->get_chars(); + auto& col_offset = result_column->get_offsets(); + col_offset.resize(input_rows_count); + + faststring compressed_str; + Slice data; + + // When the original string is large, the result is roughly this value + size_t total = arg_offset[input_rows_count - 1]; + col_data.reserve(total / 1000); + + for (size_t row = 0; row < input_rows_count; row++) { + size_t length = arg_offset[row] - arg_offset[row - 1]; + data = Slice(arg_begin + arg_offset[row - 1], length); + + // Z_MEM_ERROR and Z_BUF_ERROR are already handled in compress, making sure st is always Z_OK + auto st = compression_codec->compress(data, &compressed_str); + + size_t idx = col_data.size(); + if (!length) { // data is '' + col_data.resize(col_data.size() + 2); + col_data[idx] = '0', col_data[idx + 1] = 'x'; + col_offset[row] = col_offset[row - 1] + 2; + continue; + } + + // first ten digits represent the length of the uncompressed string + col_data.resize(col_data.size() + 10 + 2 * compressed_str.size()); Review Comment: It may be not rery reasonable. There's reason for Mysql to behave like this. 1. after compressing, the bytes in corresponding memory just a stream of bytes. so any case is possible. just interpret it as chars doesn’t keep consistency. consider a memory region of “a\b”. after printing it’s “” because ‘\b’ deletes ‘a’. 2. for the compression ratio, it’s guaranteed by compression algorithm. it has a very large ratio. so even we print it as chars which would double it length, it doesn’t matter. -- 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
