public void Dispose()
{
_sequences.ForEach(seq => ((IDisposable)seq).Dispose());
if (MmFile != null)
{
MmFile.Dispose();
MmFile = null;
}
}
/// <summary>
/// This loads the sequences from the file
/// </summary>
/// <param name="bioValidator"></param>
/// <returns></returns>
private int LoadSequences(IBioValidator bioValidator)
{
var headerBuffer = new byte[512];
string currHeader = string.Empty;
long startPos = -1;
int totalSequenceCount = 0, startingNonGap = -1;
bool loadIntoMemory = _loadAllIntoMemory || (new FileInfo(Filename).Length < SmallFileSize);
MmFile = MemoryMappedFile.CreateFromFile(Filename, FileMode.Open);
using (var fs = MmFile.CreateViewStream(0, 0, MemoryMappedFileAccess.Read))
{
for (; fs != null;)
{
int db = fs.ReadByte();
if (db <= 0)
{
break;
}
if ((char)db == '>') // Header for species?
{
// Just finished a sequence?
if (currHeader != string.Empty)
{
if (MaxSequenceLength < totalSequenceCount)
{
MaxSequenceLength = totalSequenceCount;
}
if (startingNonGap == -1)
{
startingNonGap = totalSequenceCount;
}
_sequences.Add(
new FastASequence(this, loadIntoMemory, currHeader, startPos,
(int)(fs.Position - startPos), startingNonGap, totalSequenceCount, bioValidator));
}
// Start new sequence
currHeader = string.Empty;
totalSequenceCount = 0;
startingNonGap = -1;
for (int i = 0; currHeader == string.Empty;)
{
db = fs.ReadByte();
if (db == -1 || db == 0x0a || db == 0x0d)
{
currHeader = Encoding.ASCII.GetString(headerBuffer, 0, i);
}
else
{
headerBuffer[i++] = (byte)db;
}
}
// This is the start of the nucleotide chain
startPos = fs.Position;
}
else if (db != 0x0a && db != 0x0d)
{
if (startingNonGap == -1 && bioValidator.IsValid((char)db))
{
startingNonGap = totalSequenceCount;
}
totalSequenceCount++;
}
}
// Handle the final sequence in the file.
Debug.Assert(currHeader != string.Empty);
Debug.Assert(totalSequenceCount > 0);
Debug.Assert(startPos > 0);
if (MaxSequenceLength < totalSequenceCount)
{
MaxSequenceLength = totalSequenceCount;
}
if (startingNonGap == -1)
{
startingNonGap = totalSequenceCount;
}
_sequences.Add(
new FastASequence(this, loadIntoMemory, currHeader, startPos,
(int)(fs.Position - startPos) - 1, startingNonGap, totalSequenceCount, bioValidator));
}
// Force all sequences to have the same "virtual" length
_sequences.ForEach(ns => ((FastASequence)ns).ForceLength(MaxSequenceLength));
if (loadIntoMemory)
{
MmFile.Dispose();
MmFile = null;
}
return(_sequences.Count);
}
