/// <summary>
/// Parses all the sequences in a SAM file.
/// This method is used only in data virtualization scenarios.
/// </summary>
/// <param name="mbfReader">A reader for the sequence alignment text.</param>
private void ParseSequences(MBFStreamReader mbfReader)
{
// if DV enabled
if (IsDataVirtualizationEnabled && _sidecarFileProvider.SidecarFileExists)
{
try
{
while (mbfReader.HasLines && !mbfReader.Line.StartsWith(@"@", StringComparison.OrdinalIgnoreCase))
{
SequencePointer sequencePointer = new SequencePointer { AlphabetName = Alphabets.DNA.Name };
// sequence starting index
sequencePointer.IndexOffsets[0] = mbfReader.CurrentLineStartingIndex;
// sequence ending index
sequencePointer.IndexOffsets[1] = mbfReader.CurrentLineStartingIndex + mbfReader.Line.Length;
// Write each sequence pointer to the sidecar file immediately
_sidecarFileProvider.WritePointer(sequencePointer);
mbfReader.GoToNextLine();
_lineCount++;
}
_sidecarFileProvider.Close();
}
catch (Exception)
{
_sidecarFileProvider.Cleanup();
}
}
}
/// <summary>
/// Parses file with DV and returns Virtual Qualitative Sequences list.
/// </summary>
/// <param name="isReadOnly">Flag to indicate whether the sequences returned should be set to readonly or not.</param>
private VirtualQualitativeSequenceList ParseWithDV(bool isReadOnly)
{
SidecarFileProvider sidecarFileProvider = null;
sidecarFileProvider = new SidecarFileProvider(_fileName);
sidecarFileProvider.Close();
// if valid sidecar file exists
if (sidecarFileProvider.IsSidecarValid)
{
// Create virtual list and return
return(new VirtualQualitativeSequenceList(sidecarFileProvider, this, sidecarFileProvider.Count)
{
CreateSequenceAsReadOnly = isReadOnly
});
}
// else create new sidecar
using (sidecarFileProvider = new SidecarFileProvider(_fileName, true))
{
using (_mbfStreamReader = new MBFStreamReader(_fileName))
{
if (sidecarFileProvider.SidecarFileExists)
{
try
{
while (_mbfStreamReader.HasLines)
{
ParseOne(_mbfStreamReader, isReadOnly);
}
// Create sidecar
sidecarFileProvider.CreateSidecarFile(_mbfStreamReader.FileName, _sequencePointers);
VirtualQualitativeSequenceList virtualSequences =
new VirtualQualitativeSequenceList(sidecarFileProvider, this, _sequencePointers.Count)
{
CreateSequenceAsReadOnly = isReadOnly
};
_sequencePointers.Clear();
return(virtualSequences);
}
catch (Exception)
{
sidecarFileProvider.Cleanup();
}
}
}
}
return(null);
}
/// <summary>
/// Parses a list of biological sequence texts from a file.
/// </summary>
/// <param name="filename">The name of a biological sequence file.</param>
/// <param name="isReadOnly">
/// Flag to indicate whether the resulting sequences should be in readonly mode or not.
/// If this flag is set to true then the resulting sequences's isReadOnly property
/// will be set to true, otherwise it will be set to false.
/// </param>
/// <returns>The list of parsed ISequence objects.</returns>
public IList <ISequence> Parse(string filename, bool isReadOnly)
{
// default to full load
_blockSize = FileLoadHelper.DefaultFullLoadBlockSize;
_maxNumberOfBlocks = 0;
// check if DV is required
if (filename != null)
{
_fileName = filename;
FileInfo fileInfo = new FileInfo(_fileName);
_enforceDataVirtualizationByFileSize = EnforceDataVirtualizationByFileSize * FileLoadHelper.KBytes;
if ((_enforceDataVirtualizationByFileSize != 0 && fileInfo.Length >= _enforceDataVirtualizationByFileSize) ||
_isDataVirtualizationEnforced)
{
_blockSize = FileLoadHelper.DefaultBlockSize;
_maxNumberOfBlocks = FileLoadHelper.DefaultMaxNumberOfBlocks;
}
}
// Check for sidecar
if (IsDataVirtualizationEnabled)
{
_sidecarFileProvider = new SidecarFileProvider(_fileName);
_sidecarFileProvider.Close();
// if valid sidecar file exists
if (_sidecarFileProvider.IsSidecarValid)
{
// Create virtual list and return
return(new VirtualSequenceList(_sidecarFileProvider, this, _sidecarFileProvider.Count)
{
CreateSequenceAsReadOnly = isReadOnly
});
}
// else create new sidecar
_sidecarFileProvider = new SidecarFileProvider(_fileName, true);
if (_sidecarFileProvider.SidecarFileExists)
{
using (_mbfStreamReader = new MBFStreamReader(_fileName))
{
try
{
while (_mbfStreamReader.HasLines)
{
// Parse and forget as the list is now maintained by DV using sequence pointers
ParseOne(_mbfStreamReader, isReadOnly);
}
_sidecarFileProvider.Close();
VirtualSequenceList virtualSequences =
new VirtualSequenceList(_sidecarFileProvider, this, _sidecarFileProvider.Count)
{
CreateSequenceAsReadOnly = isReadOnly
};
return(virtualSequences);
}
catch (Exception)
{
_sidecarFileProvider.Cleanup();
}
}
}
}
// non-DV parsing
using (MBFTextReader mbfReader = new MBFTextReader(filename))
{
return(Parse(mbfReader, isReadOnly));
}
}