/// <summary>
/// This creates a scan number-to-byte index for .mzML files that have no index.
/// This means that a dynamic connection can be created with unindexed .mzML files,
/// it just takes longer because we have to read the entire file one line at a time
/// to create the index via code.
///
/// This does NOT write an index to the .mzML file. This is intentional. This .mzML reader
/// does not modify the .mzML data file at all. It just creates an index in memory
/// via code if one is not provided in the .mzML file.
/// </summary>
private void CreateIndexFromUnindexedMzml()
{
reader.BaseStream.Position = 0;
reader.DiscardBufferedData();
int scanNumber = 0;
while (reader.Peek() > 0)
{
// this byte offset might be a little different than what it technically
// should be because of white space but it will work out ok
long byteOffset = TextFileReading.GetByteOffsetAtCurrentPosition(reader);
var line = reader.ReadLine();
line = line.Trim();
if (line.StartsWith("<spectrum ", StringComparison.InvariantCultureIgnoreCase))
{
Match result = nativeIdScanNumberParser.Match(line);
int ind = line.IndexOf("id=\"");
string nativeId;
if (ind >= 0)
{
StringBuilder nativeIdBuilder = new StringBuilder();
for (int r = ind + 4; r < line.Length; r++)
{
if (line[r] == '"')
{
break;
}
nativeIdBuilder.Append(line[r]);
}
nativeId = nativeIdBuilder.ToString();
}
else
{
throw new MzLibException("Could not get nativeID from line: " + line);
}
if (result.Groups[2].Success)
{
scanNumber = int.Parse(result.Groups[2].Value);
}
else
{
scanNumber++;
}
NativeIdToScanNumber.Add(nativeId, scanNumber);
ScanNumberToByteOffset.Add(scanNumber, byteOffset);
}
}
}
/// <summary>
/// Mgf files are not indexed, so the index will need to be built.
/// This method builds the index.
/// </summary>
private void BuildIndex()
{
scanToByteOffset = new Dictionary <int, long>();
int oneBasedScanNumber = 0;
long currentPositionByteOffset = 0;
long oneBasedScanByteOffset = 0;
bool scanHasAScanNumber = false;
while (reader.Peek() > 0)
{
currentPositionByteOffset = TextFileReading.GetByteOffsetAtCurrentPosition(reader);
string line = reader.ReadLine();
if (line.StartsWith("BEGIN IONS", StringComparison.InvariantCultureIgnoreCase))
{
oneBasedScanByteOffset = currentPositionByteOffset;
scanHasAScanNumber = false;
}
else if (line.StartsWith("SCANS=", StringComparison.InvariantCultureIgnoreCase))
{
scanHasAScanNumber = true;
Match result = scanNumberParser.Match(line);
var scanString = result.Groups[2].Value;
oneBasedScanNumber = int.Parse(scanString);
}
else if (line.StartsWith("END IONS", StringComparison.InvariantCultureIgnoreCase))
{
if (!scanHasAScanNumber)
{
oneBasedScanNumber++;
}
if (scanToByteOffset.ContainsKey(oneBasedScanNumber))
{
throw new MzLibException("Scan number " + oneBasedScanNumber.ToString() +
" appeared multiple times in " + FilePath + ", which is not allowed because we assume all scan numbers are unique.");
}
scanToByteOffset.Add(oneBasedScanNumber, oneBasedScanByteOffset);
}
}
}