Exemple #1
0
  ExitCodes main_(int, const char**)
  {
    //-------------------------------------------------------------
    // parameter handling
    //-------------------------------------------------------------

    //input file names
    String in = getStringOption_("in");

    //input file type
    FileHandler fh;
    FileTypes::Type in_type = FileTypes::nameToType(getStringOption_("in_type"));

    if (in_type == FileTypes::UNKNOWN)
    {
      in_type = fh.getType(in);
      writeDebug_(String("Input file type: ") + FileTypes::typeToName(in_type), 2);
    }

    if (in_type == FileTypes::UNKNOWN)
    {
      writeLog_("Error: Could not determine input file type!");
      return PARSE_ERROR;
    }


    //output file names and types
    String out = getStringOption_("out");
    FileTypes::Type out_type = FileTypes::nameToType(getStringOption_("out_type"));

    if (out_type == FileTypes::UNKNOWN)
    {
      out_type = fh.getTypeByFileName(out);
    }

    if (out_type == FileTypes::UNKNOWN)
    {
      writeLog_("Error: Could not determine output file type!");
      return PARSE_ERROR;
    }

    bool TIC_DTA2D = getFlag_("TIC_DTA2D");

    writeDebug_(String("Output file type: ") + FileTypes::typeToName(out_type), 1);

    //-------------------------------------------------------------
    // reading input
    //-------------------------------------------------------------
    typedef MSExperiment<Peak1D> MSExperimentType;
    MSExperimentType exp;

    typedef MSExperimentType::SpectrumType SpectrumType;

    typedef FeatureMap<> FeatureMapType;

    FeatureMapType fm;
    ConsensusMap cm;

    writeDebug_(String("Loading input file"), 1);

    if (in_type == FileTypes::CONSENSUSXML)
    {
      ConsensusXMLFile().load(in, cm);
      cm.sortByPosition();
      if ((out_type != FileTypes::FEATUREXML) &&
          (out_type != FileTypes::CONSENSUSXML))
      {
        // You you will lose information and waste memory. Enough reasons to issue a warning!
        writeLog_("Warning: Converting consensus features to peaks. You will lose information!");
        exp.set2DData(cm);
      }
    }
    else if (in_type == FileTypes::EDTA)
    {
      EDTAFile().load(in, cm);
      cm.sortByPosition();
      if ((out_type != FileTypes::FEATUREXML) &&
          (out_type != FileTypes::CONSENSUSXML))
      {
        // You you will lose information and waste memory. Enough reasons to issue a warning!
        writeLog_("Warning: Converting consensus features to peaks. You will lose information!");
        exp.set2DData(cm);
      }
    }
    else if (in_type == FileTypes::FEATUREXML ||
             in_type == FileTypes::TSV ||
             in_type == FileTypes::PEPLIST ||
             in_type == FileTypes::KROENIK)
    {
      fh.loadFeatures(in, fm, in_type);
      fm.sortByPosition();
      if ((out_type != FileTypes::FEATUREXML) &&
          (out_type != FileTypes::CONSENSUSXML))
      {
        // You will lose information and waste memory. Enough reasons to issue a warning!
        writeLog_("Warning: Converting features to peaks. You will lose information! Mass traces are added, if present as 'num_of_masstraces' and 'masstrace_intensity_<X>' (X>=0) meta values.");
        exp.set2DData<true>(fm);
      }
    }
    else
    {
      fh.loadExperiment(in, exp, in_type, log_type_);
    }

    //-------------------------------------------------------------
    // writing output
    //-------------------------------------------------------------

    writeDebug_(String("Writing output file"), 1);

    if (out_type == FileTypes::MZML)
    {
      //add data processing entry
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 CONVERSION_MZML));
      MzMLFile f;
      f.setLogType(log_type_);
      ChromatogramTools().convertSpectraToChromatograms(exp, true);
      f.store(out, exp);
    }
    else if (out_type == FileTypes::MZDATA)
    {
      //annotate output with data processing info
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 CONVERSION_MZDATA));
      MzDataFile f;
      f.setLogType(log_type_);
      ChromatogramTools().convertChromatogramsToSpectra<MSExperimentType>(exp);
      f.store(out, exp);
    }
    else if (out_type == FileTypes::MZXML)
    {
      //annotate output with data processing info
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 CONVERSION_MZXML));
      MzXMLFile f;
      f.setLogType(log_type_);
      ChromatogramTools().convertChromatogramsToSpectra<MSExperimentType>(exp);
      f.store(out, exp);
    }
    else if (out_type == FileTypes::DTA2D)
    {
      //add data processing entry
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 FORMAT_CONVERSION));
      DTA2DFile f;
      f.setLogType(log_type_);
      ChromatogramTools().convertChromatogramsToSpectra<MSExperimentType>(exp);
      if (TIC_DTA2D)
      {
        // store the total ion chromatogram (TIC)
        f.storeTIC(out, exp);
      }
      else
      {
        // store entire experiment
        f.store(out, exp);
      }


    }
    else if (out_type == FileTypes::MGF)
    {
      //add data processing entry
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 FORMAT_CONVERSION));
      MascotGenericFile f;
      f.setLogType(log_type_);
      f.store(out, exp);
    }
    else if (out_type == FileTypes::FEATUREXML)
    {
      if ((in_type == FileTypes::FEATUREXML) || (in_type == FileTypes::TSV) ||
          (in_type == FileTypes::PEPLIST) || (in_type == FileTypes::KROENIK))
      {
        fm.applyMemberFunction(&UniqueIdInterface::setUniqueId);
      }
      else if (in_type == FileTypes::CONSENSUSXML || in_type == FileTypes::EDTA)
      {
        ConsensusMap::convert(cm, true, fm);
      }
      else // not loaded as feature map or consensus map
      {
        // The feature specific information is only defaulted. Enough reasons to issue a warning!
        writeLog_("Warning: Converting peaks to features will lead to incomplete features!");
        fm.clear();
        fm.reserve(exp.getSize());
        typedef FeatureMapType::FeatureType FeatureType;
        FeatureType feature;
        feature.setQuality(0, 1); // override default
        feature.setQuality(1, 1); // override default
        feature.setOverallQuality(1); // override default
        for (MSExperimentType::ConstIterator spec_iter = exp.begin();
             spec_iter != exp.end();
             ++spec_iter
             )
        {
          feature.setRT(spec_iter->getRT());
          for (SpectrumType::ConstIterator peak1_iter = spec_iter->begin();
               peak1_iter != spec_iter->end();
               ++peak1_iter
               )
          {
            feature.setMZ(peak1_iter->getMZ());
            feature.setIntensity(peak1_iter->getIntensity());
            feature.setUniqueId();
            fm.push_back(feature);
          }
        }
        fm.updateRanges();
      }

      addDataProcessing_(fm, getProcessingInfo_(DataProcessing::
                                                FORMAT_CONVERSION));
      FeatureXMLFile().store(out, fm);
    }
    else if (out_type == FileTypes::CONSENSUSXML)
    {
      if ((in_type == FileTypes::FEATUREXML) || (in_type == FileTypes::TSV) ||
          (in_type == FileTypes::PEPLIST) || (in_type == FileTypes::KROENIK))
      {
        fm.applyMemberFunction(&UniqueIdInterface::setUniqueId);
        ConsensusMap::convert(0, fm, cm);
      }
      // nothing to do for consensus input
      else if (in_type == FileTypes::CONSENSUSXML || in_type == FileTypes::EDTA)
      {
      }
      else // experimental data
      {
        ConsensusMap::convert(0, exp, cm, exp.size());
      }

      addDataProcessing_(cm, getProcessingInfo_(DataProcessing::
                                                FORMAT_CONVERSION));
      ConsensusXMLFile().store(out, cm);
    }
    else if (out_type == FileTypes::EDTA)
    {
      if (fm.size() > 0 && cm.size() > 0)
      {
        LOG_ERROR << "Internal error: cannot decide on container (Consensus or Feature)! This is a bug. Please report it!";
        return INTERNAL_ERROR;
      }
      if (fm.size() > 0) EDTAFile().store(out, fm);
      else if (cm.size() > 0) EDTAFile().store(out, cm);
    }
    else
    {
      writeLog_("Unknown output file type given. Aborting!");
      printUsage_();
      return ILLEGAL_PARAMETERS;
    }

    return EXECUTION_OK;
  }
  ExitCodes main_(int, const char**)
  {
    //-------------------------------------------------------------
    // parameter handling
    //-------------------------------------------------------------

    //input file names
    String in = getStringOption_("in");
    bool write_mzML_index = getFlag_("write_mzML_index");

    //input file type
    FileHandler fh;
    FileTypes::Type in_type = FileTypes::nameToType(getStringOption_("in_type"));

    if (in_type == FileTypes::UNKNOWN)
    {
      in_type = fh.getType(in);
      writeDebug_(String("Input file type: ") + FileTypes::typeToName(in_type), 2);
    }

    if (in_type == FileTypes::UNKNOWN)
    {
      writeLog_("Error: Could not determine input file type!");
      return PARSE_ERROR;
    }


    //output file names and types
    String out = getStringOption_("out");
    FileTypes::Type out_type = FileTypes::nameToType(getStringOption_("out_type"));

    if (out_type == FileTypes::UNKNOWN)
    {
      out_type = fh.getTypeByFileName(out);
    }

    if (out_type == FileTypes::UNKNOWN)
    {
      writeLog_("Error: Could not determine output file type!");
      return PARSE_ERROR;
    }

    bool TIC_DTA2D = getFlag_("TIC_DTA2D");
    bool process_lowmemory = getFlag_("process_lowmemory");

    writeDebug_(String("Output file type: ") + FileTypes::typeToName(out_type), 1);

    String uid_postprocessing = getStringOption_("UID_postprocessing");
    //-------------------------------------------------------------
    // reading input
    //-------------------------------------------------------------
    typedef MSExperiment<Peak1D> MSExperimentType;
    MSExperimentType exp;

    typedef MSExperimentType::SpectrumType SpectrumType;

    typedef FeatureMap FeatureMapType;

    FeatureMapType fm;
    ConsensusMap cm;

    writeDebug_(String("Loading input file"), 1);

    if (in_type == FileTypes::CONSENSUSXML)
    {
      ConsensusXMLFile().load(in, cm);
      cm.sortByPosition();
      if ((out_type != FileTypes::FEATUREXML) &&
          (out_type != FileTypes::CONSENSUSXML))
      {
        // You you will lose information and waste memory. Enough reasons to issue a warning!
        writeLog_("Warning: Converting consensus features to peaks. You will lose information!");
        exp.set2DData(cm);
      }
    }
    else if (in_type == FileTypes::EDTA)
    {
      EDTAFile().load(in, cm);
      cm.sortByPosition();
      if ((out_type != FileTypes::FEATUREXML) &&
          (out_type != FileTypes::CONSENSUSXML))
      {
        // You you will lose information and waste memory. Enough reasons to issue a warning!
        writeLog_("Warning: Converting consensus features to peaks. You will lose information!");
        exp.set2DData(cm);
      }
    }
    else if (in_type == FileTypes::FEATUREXML ||
             in_type == FileTypes::TSV ||
             in_type == FileTypes::PEPLIST ||
             in_type == FileTypes::KROENIK)
    {
      fh.loadFeatures(in, fm, in_type);
      fm.sortByPosition();
      if ((out_type != FileTypes::FEATUREXML) &&
          (out_type != FileTypes::CONSENSUSXML))
      {
        // You will lose information and waste memory. Enough reasons to issue a warning!
        writeLog_("Warning: Converting features to peaks. You will lose information! Mass traces are added, if present as 'num_of_masstraces' and 'masstrace_intensity_<X>' (X>=0) meta values.");
        exp.set2DData<true>(fm);
      }
    }
    else if (process_lowmemory)
    {
      // Special switch for the low memory options:
      // We can transform the complete experiment directly without first
      // loading the complete data into memory. PlainMSDataWritingConsumer will
      // write out mzML to disk as they are read from the input.
      if (in_type == FileTypes::MZML && out_type == FileTypes::MZML)
      {
        PlainMSDataWritingConsumer consumer(out);
        consumer.getOptions().setWriteIndex(write_mzML_index);
        consumer.addDataProcessing(getProcessingInfo_(DataProcessing::CONVERSION_MZML));
        MzMLFile mzmlfile; 
        mzmlfile.setLogType(log_type_);
        mzmlfile.transform(in, &consumer);
        return EXECUTION_OK;
      }
      else if (in_type == FileTypes::MZXML && out_type == FileTypes::MZML)
      {
        PlainMSDataWritingConsumer consumer(out);
        consumer.getOptions().setWriteIndex(write_mzML_index);
        consumer.addDataProcessing(getProcessingInfo_(DataProcessing::CONVERSION_MZML));
        MzXMLFile mzxmlfile; 
        mzxmlfile.setLogType(log_type_);
        mzxmlfile.transform(in, &consumer);
        return EXECUTION_OK;
      }
      else
      {
        throw Exception::IllegalArgument(__FILE__, __LINE__, __PRETTY_FUNCTION__,
          "Process_lowmemory option can only be used with mzML / mzXML input and mzML output data types.");
      }
    }
    else
    {
      fh.loadExperiment(in, exp, in_type, log_type_);
    }

    //-------------------------------------------------------------
    // writing output
    //-------------------------------------------------------------

    writeDebug_(String("Writing output file"), 1);

    if (out_type == FileTypes::MZML)
    {
      //add data processing entry
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 CONVERSION_MZML));
      MzMLFile f;
      f.setLogType(log_type_);
      f.getOptions().setWriteIndex(write_mzML_index);
      ChromatogramTools().convertSpectraToChromatograms(exp, true);
      f.store(out, exp);
    }
    else if (out_type == FileTypes::MZDATA)
    {
      //annotate output with data processing info
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 CONVERSION_MZDATA));
      MzDataFile f;
      f.setLogType(log_type_);
      ChromatogramTools().convertChromatogramsToSpectra<MSExperimentType>(exp);
      f.store(out, exp);
    }
    else if (out_type == FileTypes::MZXML)
    {
      //annotate output with data processing info
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 CONVERSION_MZXML));
      MzXMLFile f;
      f.setLogType(log_type_);
      ChromatogramTools().convertChromatogramsToSpectra<MSExperimentType>(exp);
      f.store(out, exp);
    }
    else if (out_type == FileTypes::DTA2D)
    {
      //add data processing entry
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 FORMAT_CONVERSION));
      DTA2DFile f;
      f.setLogType(log_type_);
      ChromatogramTools().convertChromatogramsToSpectra<MSExperimentType>(exp);
      if (TIC_DTA2D)
      {
        // store the total ion chromatogram (TIC)
        f.storeTIC(out, exp);
      }
      else
      {
        // store entire experiment
        f.store(out, exp);
      }


    }
    else if (out_type == FileTypes::MGF)
    {
      //add data processing entry
      addDataProcessing_(exp, getProcessingInfo_(DataProcessing::
                                                 FORMAT_CONVERSION));
      MascotGenericFile f;
      f.setLogType(log_type_);
      f.store(out, exp, getFlag_("MGF_compact"));
    }
    else if (out_type == FileTypes::FEATUREXML)
    {
      if ((in_type == FileTypes::FEATUREXML) || (in_type == FileTypes::TSV) ||
          (in_type == FileTypes::PEPLIST) || (in_type == FileTypes::KROENIK))
      {
        if (uid_postprocessing == "ensure")
        {
          fm.applyMemberFunction(&UniqueIdInterface::ensureUniqueId);
        } else if (uid_postprocessing == "reassign")
        {
          fm.applyMemberFunction(&UniqueIdInterface::setUniqueId);
        }
      }
      else if (in_type == FileTypes::CONSENSUSXML || in_type == FileTypes::EDTA)
      {
        MapConversion::convert(cm, true, fm);
      }
      else // not loaded as feature map or consensus map
      {
        // The feature specific information is only defaulted. Enough reasons to issue a warning!
        writeLog_("Warning: Converting peaks to features will lead to incomplete features!");
        fm.clear();
        fm.reserve(exp.getSize());
        typedef FeatureMapType::FeatureType FeatureType;
        FeatureType feature;
        feature.setQuality(0, 1); // override default
        feature.setQuality(1, 1); // override default
        feature.setOverallQuality(1); // override default
        for (MSExperimentType::ConstIterator spec_iter = exp.begin();
             spec_iter != exp.end();
             ++spec_iter
             )
        {
          feature.setRT(spec_iter->getRT());
          for (SpectrumType::ConstIterator peak1_iter = spec_iter->begin();
               peak1_iter != spec_iter->end();
               ++peak1_iter
               )
          {
            feature.setMZ(peak1_iter->getMZ());
            feature.setIntensity(peak1_iter->getIntensity());
            feature.setUniqueId();
            fm.push_back(feature);
          }
        }
        fm.updateRanges();
      }

      addDataProcessing_(fm, getProcessingInfo_(DataProcessing::
                                                FORMAT_CONVERSION));
      FeatureXMLFile().store(out, fm);
    }
    else if (out_type == FileTypes::CONSENSUSXML)
    {
      if ((in_type == FileTypes::FEATUREXML) || (in_type == FileTypes::TSV) ||
          (in_type == FileTypes::PEPLIST) || (in_type == FileTypes::KROENIK))
      {
        if (uid_postprocessing == "ensure")
        {
          fm.applyMemberFunction(&UniqueIdInterface::ensureUniqueId);
        } else if (uid_postprocessing == "reassign")
        {
          fm.applyMemberFunction(&UniqueIdInterface::setUniqueId);
        }
        MapConversion::convert(0, fm, cm);
      }
      // nothing to do for consensus input
      else if (in_type == FileTypes::CONSENSUSXML || in_type == FileTypes::EDTA)
      {
      }
      else // experimental data
      {
        MapConversion::convert(0, exp, cm, exp.size());
      }

      addDataProcessing_(cm, getProcessingInfo_(DataProcessing::
                                                FORMAT_CONVERSION));
      ConsensusXMLFile().store(out, cm);
    }
    else if (out_type == FileTypes::EDTA)
    {
      if (fm.size() > 0 && cm.size() > 0)
      {
        LOG_ERROR << "Internal error: cannot decide on container (Consensus or Feature)! This is a bug. Please report it!";
        return INTERNAL_ERROR;
      }
      if (fm.size() > 0) EDTAFile().store(out, fm);
      else if (cm.size() > 0) EDTAFile().store(out, cm);
    }
    else if (out_type == FileTypes::CSV)
    {
      // as ibspectra is currently the only csv/text based format we assume
      // that out_type == FileTypes::CSV means ibspectra, if more formats
      // are added we need a more intelligent strategy to decide which
      // conversion is requested

      // IBSpectra selected as output type
      if (in_type != FileTypes::CONSENSUSXML)
      {
        LOG_ERROR << "Incompatible input data: FileConverter can only convert consensusXML files to ibspectra format.";
        return INCOMPATIBLE_INPUT_DATA;
      }

      IBSpectraFile ibfile;
      ibfile.store(out, cm);
    }
    else
    {
      writeLog_("Unknown output file type given. Aborting!");
      printUsage_();
      return ILLEGAL_PARAMETERS;
    }

    return EXECUTION_OK;
  }