public IList<PeakData> ImportFile(TextReader peakViewReader)
{
var peakDatas = new List<PeakData>();
var msFileNames = new List<string>();
var fileReader = new DsvFileReader(peakViewReader, TextUtil.SEPARATOR_TSV);
Assert.AreEqual(fileReader.GetFieldIndex(PROTEIN), 0);
Assert.AreEqual(fileReader.GetFieldIndex(PEPTIDE), 1);
Assert.AreEqual(fileReader.GetFieldIndex(LABEL), 2);
Assert.AreEqual(fileReader.GetFieldIndex(PRECURSOR), 3);
Assert.AreEqual(fileReader.GetFieldIndex(CHARGE), 4);
Assert.AreEqual(fileReader.GetFieldIndex(RT), 5);
Assert.AreEqual(fileReader.GetFieldIndex(DECOY), 6);
for (int i = 7; i < fileReader.NumberOfFields; ++i)
{
msFileNames.Add(fileReader.FieldNames[i]);
}
while (fileReader.ReadLine() != null)
{
string modifiedSequence = fileReader.GetFieldByName(PEPTIDE);
string decoyString = fileReader.GetFieldByName(DECOY);
bool decoy;
Assert.IsTrue(bool.TryParse(decoyString, out decoy));
foreach (var msFileName in msFileNames)
{
string dataFieldString = fileReader.GetFieldByName(msFileName);
double dataField;
Assert.IsTrue(double.TryParse(dataFieldString, out dataField));
var peakData = new PeakData(dataField, modifiedSequence, msFileName, decoy);
peakDatas.Add(peakData);
}
}
return peakDatas;
}
public static IList<Dictionary<string, string>> ReadCsvFile(DsvFileReader fileReader)
{
var result = new List<Dictionary<string, string>>();
while (null != fileReader.ReadLine())
{
Dictionary<string, string> row = new Dictionary<string, string>();
for (int i = 0; i < fileReader.NumberOfFields; i++)
{
var value = fileReader.GetFieldByIndex(i);
if (null != value)
{
row.Add(fileReader.FieldNames[i], value);
}
}
result.Add(row);
}
return result;
}
public static IDictionary<string, LinearFitResult> ReadExpectedResults(Type type, string resourceName)
{
var result = new Dictionary<string, LinearFitResult>();
using (var reader = GetTextReaderForManifestResource(type, resourceName))
{
var csvReader = new DsvFileReader(reader, ',');
while (null != csvReader.ReadLine())
{
string protein = csvReader.GetFieldByName("Protein");
var linearFitResult = new LinearFitResult(Convert.ToDouble(csvReader.GetFieldByName("log2FC"), CultureInfo.InvariantCulture))
.SetStandardError(Convert.ToDouble(csvReader.GetFieldByName("SE"), CultureInfo.InvariantCulture))
.SetTValue(Convert.ToDouble(csvReader.GetFieldByName("Tvalue"), CultureInfo.InvariantCulture))
.SetDegreesOfFreedom(Convert.ToInt32(csvReader.GetFieldByName("DF"), CultureInfo.InvariantCulture))
.SetPValue(Convert.ToDouble(csvReader.GetFieldByName("pvalue"), CultureInfo.InvariantCulture));
result.Add(protein, linearFitResult);
}
}
return result;
}
public void DsvHeadersTest()
{
// test reading DSV files with and without headers
string[] lines = {"dog,1.1,cat,2.2", "pony,3.3,fish,4.4"}; // Not L10N
const char csvSeperator = ',';
var withheaders = new DsvFileReader(new StringReader("0,1,2,3\n" + String.Join("\n", lines)), csvSeperator); // Not L10N
var withoutheaders = new DsvFileReader(new StringReader(String.Join("\n", lines)), csvSeperator, hasHeaders: false); // Not L10N
Assert.AreEqual(withheaders.NumberOfFields, withoutheaders.NumberOfFields);
for (int h = 0; h < withoutheaders.NumberOfFields; h++)
{
var f = String.Format("{0}", h); // verify that a headerless CSV file will pretend to have a header of form "0,1,2,..." // Not L10N
Assert.AreEqual(withheaders.GetFieldIndex(f), withoutheaders.GetFieldIndex(f));
}
int linenumber = 0;
while (true)
{
var fields = withoutheaders.ReadLine();
var fieldsTest = withheaders.ReadLine();
Assert.IsTrue((fields == null) == (fieldsTest == null));
if (fields == null)
break;
for (int f = 0; f < withoutheaders.NumberOfFields; f++)
{
Assert.AreEqual(fields[f], fieldsTest[f]); // quick string compare
Assert.AreEqual(lines[linenumber].Split(csvSeperator)[f],fields[f]);
}
linenumber++;
}
Assert.AreEqual(2,linenumber);
// let's also check the precision handling and field exception in AssertEx.FieldsEqual
var A = String.Join("\n", lines).Replace("cat","hamster"); // change col 2, then ignore it // Not L10N
var B = String.Join("\n", lines).Replace("1.1","1.09"); // add error at limits of precision, then allow for it // Not L10N
AssertEx.FieldsEqual(A, B, 4, 2, true);
}
public List<string> TranscribeAndModifyFile(StreamWriter writer, TextReader reader, List<string> fields, bool first, int currentFileCount)
{
var fileReader = new DsvFileReader(reader, SEPARATOR);
if (first)
{
fields = fileReader.FieldNames;
for (int i = 0; i < fields.Count; ++i)
{
if (i > 0)
writer.Write(SEPARATOR);
writer.WriteDsvField(fileReader.FieldNames[i], SEPARATOR);
}
writer.WriteLine();
}
Assert.AreEqual(fileReader.NumberOfFields, fields.Count);
for (int i = 0; i < fields.Count; ++i)
{
Assert.AreEqual(fileReader.FieldNames[i], fields[i]);
}
while (fileReader.ReadLine() != null)
{
for (int i = 0; i < fields.Count; ++i)
{
string modifiedField = fileReader.GetFieldByIndex(i);
switch (fileReader.FieldNames[i])
{
case FILE_NAME:
modifiedField = Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(modifiedField)) + MS_FILE_TYPE;
break;
case TRANSITION_GROUP:
modifiedField = modifiedField + currentFileCount;
break;
case RUN_ID:
modifiedField = currentFileCount.ToString(CultureInfo.CurrentCulture);
break;
}
if (i > 0)
writer.Write(SEPARATOR);
writer.WriteDsvField(modifiedField, SEPARATOR);
}
writer.WriteLine();
}
return fields;
}