ParquetFile.h

Go to the documentation of this file.

// Copyright (c) 2002-present, OpenMS Inc. -- EKU Tuebingen, ETH Zurich, and FU Berlin
// SPDX-License-Identifier: BSD-3-Clause
//
// --------------------------------------------------------------------------
// $Maintainer: Justin Sing $
// $Authors: Justin Sing $
// --------------------------------------------------------------------------
 
#pragma once
 
#include <OpenMS/config.h>
#include <OpenMS/CONCEPT/Types.h>
#include <OpenMS/DATASTRUCTURES/String.h>
#include <OpenMS/SYSTEM/File.h>
 
#include <arrow/api.h>
#include <arrow/io/file.h>
#include <parquet/arrow/writer.h>
#include <parquet/arrow/reader.h>
 
#include <memory>
#include <string>
#include <vector>
 
namespace OpenMS
{
 
  class OPENMS_DLLAPI ParquetFile
  {
  public:
 
    static void appendOrThrow(const arrow::Status& status, const std::string& column);
 
    template <typename BuilderT>
    static std::shared_ptr<arrow::Array> finishArray(BuilderT& builder, const std::string& name)
    {
      std::shared_ptr<arrow::Array> array;
      auto status = builder.Finish(&array);
      if (!status.ok())
      {
        throw_finish_error_(name, status.ToString());
      }
      return array;
    }
 
 
 
    static void writeTable(const std::shared_ptr<arrow::Table>& table, const String& filename,
                int64_t row_group_size = 262144);
 
    static std::shared_ptr<arrow::Table> readTable(const String& filename);
    static std::shared_ptr<arrow::Table> readTable(const std::shared_ptr<arrow::io::RandomAccessFile>& infile);
 
 
 
    static std::shared_ptr<arrow::Array> getColumn(const std::shared_ptr<arrow::Table>& table,
                                                   const std::string& name);
 
    static std::shared_ptr<arrow::Array> getOptionalColumn(const std::shared_ptr<arrow::Table>& table,
                                                           const std::string& name);
 
 
 
    static int64_t getInt64(const std::shared_ptr<arrow::Array>& array, int64_t row,
                            int64_t default_value, bool allow_null);
 
    static double getDouble(const std::shared_ptr<arrow::Array>& array, int64_t row,
                            double default_value, bool allow_null);
 
    static bool getBool(const std::shared_ptr<arrow::Array>& array, int64_t row,
                        bool default_value, bool allow_null);
 
    static std::string getString(const std::shared_ptr<arrow::Array>& array, int64_t row);
 
    static std::vector<std::string> getStringList(const std::shared_ptr<arrow::Array>& array, int64_t row);
 
 
 
    static std::string jsonEscape(const String& input);
 
    static int64_t rowCount(const String& filename);
 
 
 
  private:
    static void throw_finish_error_(const std::string& name, const std::string& error);
  };
 
} // namespace OpenMS