arrow/csv/column_decoder.h

0001 // Licensed to the Apache Software Foundation (ASF) under one
0002 // or more contributor license agreements.  See the NOTICE file
0003 // distributed with this work for additional information
0004 // regarding copyright ownership.  The ASF licenses this file
0005 // to you under the Apache License, Version 2.0 (the
0006 // "License"); you may not use this file except in compliance
0007 // with the License.  You may obtain a copy of the License at
0008 //
0009 //   http://www.apache.org/licenses/LICENSE-2.0
0010 //
0011 // Unless required by applicable law or agreed to in writing,
0012 // software distributed under the License is distributed on an
0013 // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
0014 // KIND, either express or implied.  See the License for the
0015 // specific language governing permissions and limitations
0016 // under the License.
0017
0018 #pragma once
0019
0020 #include <cstdint>
0021 #include <memory>
0022 #include <utility>
0023
0024 #include "arrow/result.h"
0025 #include "arrow/type_fwd.h"
0026 #include "arrow/util/type_fwd.h"
0027 #include "arrow/util/visibility.h"
0028
0029 namespace arrow {
0030 namespace csv {
0031
0032 class BlockParser;
0033 struct ConvertOptions;
0034
0035 class ARROW_EXPORT ColumnDecoder {
0036  public:
0037   virtual ~ColumnDecoder() = default;
0038
0039   /// Spawn a task that will try to convert and insert the given CSV block
0040   virtual Future<std::shared_ptr<Array>> Decode(
0041       const std::shared_ptr<BlockParser>& parser) = 0;
0042
0043   /// Construct a strictly-typed ColumnDecoder.
0044   static Result<std::shared_ptr<ColumnDecoder>> Make(MemoryPool* pool,
0045                                                      std::shared_ptr<DataType> type,
0046                                                      int32_t col_index,
0047                                                      const ConvertOptions& options);
0048
0049   /// Construct a type-inferring ColumnDecoder.
0050   /// Inference will run only on the first block, the type will be frozen afterwards.
0051   static Result<std::shared_ptr<ColumnDecoder>> Make(MemoryPool* pool, int32_t col_index,
0052                                                      const ConvertOptions& options);
0053
0054   /// Construct a ColumnDecoder for a column of nulls
0055   /// (i.e. not present in the CSV file).
0056   static Result<std::shared_ptr<ColumnDecoder>> MakeNull(MemoryPool* pool,
0057                                                          std::shared_ptr<DataType> type);
0058
0059  protected:
0060   ColumnDecoder() = default;
0061 };
0062
0063 }  // namespace csv
0064 }  // namespace arrow