public void ValidateFile(Stream stream, TextWriter outstream, Action<int> progress, Action<string> status)
{
try {
MZTabErrorList errorList = new MZTabErrorList(Level.Info);
try {
validate(stream, outstream, errorList, progress, status);
//refine();
} catch (MZTabException e) {
outstream.Write(MZTabProperties.MZTabExceptionMessage);
errorList.Add(e.Error);
} catch (MZTabErrorOverflowException) {
outstream.Write(MZTabProperties.MZTabErrorOverflowExceptionMessage);
}
errorList.print(outstream);
if (errorList.IsNullOrEmpty()) {
outstream.Write("No errors in this section!" + MZTabConstants.NEW_LINE);
}
outstream.Close();
//stream.Close();
} catch (Exception e) {
MessageBox.Show(e.Message, e.StackTrace);
}
}
/**
* For facing colunit definition line, for example:
* MTD colunit-protein retention_time=[UO, UO:000031, minute, ]
* after parse metadata and header lines, need calling
* {@link #refineColUnit(uk.ac.ebi.pride.jmztab.model.MZTabColumnFactory)} manually.
*/
public new void Parse(int lineNumber, string mtdLine, MZTabErrorList errorList)
{
base.Parse(lineNumber, mtdLine, errorList);
if (_items.Length != 3){
MZTabError error = new MZTabError(FormatErrorType.MTDLine, lineNumber, mtdLine);
throw new MZTabException(error);
}
string defineLabel = _items[1].Trim().ToLower();
string valueLabel = _items[2].Trim();
if (defineLabel.Contains("colunit")){
// ignore colunit parse. In the stage, just store them into colUnitMap<defineLabel, valueLabel>.
// after table section columns created, call checkColUnit manually.
_colUnitMap.Add(defineLabel, valueLabel);
if (! defineLabel.Equals("colunit-protein") &&
! defineLabel.Equals("colunit-peptide") &&
! defineLabel.Equals("colunit-psm") &&
! defineLabel.Equals("colunit-small_molecule")){
MZTabError error = new MZTabError(FormatErrorType.MTDDefineLabel, lineNumber, defineLabel);
throw new MZTabException(error);
}
}
else{
parseNormalMetadata(defineLabel, valueLabel);
}
}
/**
* Parse a header line into {@link MZTabColumnFactory} structure. There are two steps in this method:
* Step 1: {@link #parseStableOrderColumns()} focus on validate and parse for stable columns and optional
* columns which have stable order; and Step 2: {@link #parseOptionalColumns(int)} focus on {@link AbundanceColumn},
* {@link OptionColumn} and {@link CVParamOptionColumn} parse and validation.
*/
protected new void Parse(int lineNumber, string line, MZTabErrorList errorList)
{
base.Parse(lineNumber, line, errorList);
int offset = parseStableOrderColumns();
if (offset < _items.Length){
parseOptionalColumns(offset);
}
refine();
}
/**
* We assume that user before call this method, have parse the raw line
* is not empty line and start with section prefix.
*/
protected void Parse(int lineNumber, string line, MZTabErrorList errorList)
{
_lineNumber = lineNumber;
_line = line;
_errorList = errorList ?? new MZTabErrorList();
_items = line.Split(MZTabConstants.TAB);
_items[0] = _items[0].Trim();
_items[_items.Length - 1] = _items[_items.Length - 1].Trim();
_section = Section.findSection(_items[0]);
if (_section == null){
MZTabError error = new MZTabError(FormatErrorType.LinePrefix, lineNumber, _items[0]);
_errorList.Add(error);
}
}
/**
* Generate a mzTab data line parser. A couple of common method used to parse a data line into
* {@link MZTabRecord} structure.
*
* NOTICE: {@link MZTabColumnFactory} maintain a couple of {@link MZTabColumn} which have internal logical
* position and order. In physical mzTab file, we allow user not obey this logical position organized way,
* and provide their date with own order. In order to distinguish them, we use physical position (a positive
* integer) to record the column location in mzTab file. And use {@link PositionMapping} structure the maintain
* the mapping between them.
*
* @param factory SHOULD NOT set null
* @param positionMapping SHOULD NOT set null
* @param metadata SHOULD NOT set null
*/
protected MZTabDataLineParser(MZTabColumnFactory factory, PositionMapping positionMapping,
Metadata metadata, MZTabErrorList errorList)
{
if (factory == null){
throw new NullReferenceException("Column header factory should be create first.");
}
this.factory = factory;
this.positionMapping = positionMapping;
exchangeMapping = positionMapping.exchange();
mapping = factory.GetOffsetColumnsMap();
if (metadata == null){
throw new NullReferenceException("Metadata should be parser first.");
}
this.metadata = metadata;
_errorList = errorList ?? new MZTabErrorList();
}
public MZTabFileParser(string tabFile, TextWriter outstream, Level level)
{
init(tabFile);
try{
errorList = new MZTabErrorList(level);
check();
refine();
}
catch (MZTabException e){
outstream.Write(MZTabProperties.MZTabExceptionMessage);
errorList.Add(e.Error);
}
catch (MZTabErrorOverflowException){
outstream.Write(MZTabProperties.MZTabErrorOverflowExceptionMessage);
}
errorList.print(outstream);
if (errorList.IsNullOrEmpty()){
outstream.Write("not errors in " + tabFile + " file!" + MZTabConstants.NEW_LINE);
}
}
public PSMLineParser(MZTabColumnFactory factory, PositionMapping positionMapping, Metadata metadata,
MZTabErrorList errorList)
: base(factory, positionMapping, metadata, errorList)
{
}
/**
* Validate the data line, if there exist errors, add them into {@link MZTabErrorList}.
*
* NOTICE: this step just do validate, not do convert operation. Convert the data line into
* {@link MZTabRecord} implemented by {@link #getRecord(uk.ac.ebi.pride.jmztab.model.Section, string)}
* method.
*/
public new void Parse(int lineNumber, string line, MZTabErrorList errorList)
{
base.Parse(lineNumber, line, errorList);
checkCount();
checkStableData();
checkOptionalData();
}
private void validate(Stream stream, TextWriter outstream, MZTabErrorList errorList,
Action<int> progress, Action<string> status)
{
COMLineParser comParser = new COMLineParser();
MTDLineParser mtdParser = new MTDLineParser();
PRHLineParser prhParser = null;
PRTLineParser prtParser = null;
PEHLineParser pehParser = null;
PEPLineParser pepParser = null;
PSHLineParser pshParser = null;
PSMLineParser psmParser = null;
SMHLineParser smhParser = null;
SMLLineParser smlParser = null;
SortedDictionary<int, Comment> commentMap = new SortedDictionary<int, Comment>();
SortedDictionary<int, Protein> proteinMap = new SortedDictionary<int, Protein>();
SortedDictionary<int, Peptide> peptideMap = new SortedDictionary<int, Peptide>();
SortedDictionary<int, PSM> psmMap = new SortedDictionary<int, PSM>();
SortedDictionary<int, SmallMolecule> smallMoleculeMap = new SortedDictionary<int, SmallMolecule>();
PositionMapping prtPositionMapping = null;
PositionMapping pepPositionMapping = null;
PositionMapping psmPositionMapping = null;
PositionMapping smlPositionMapping = null;
StreamReader reader = new StreamReader(stream);
int highWaterMark = 1;
int lineNumber = 0;
try {
string line;
while ((line = reader.ReadLine()) != null) {
progress((int)(stream.Position * 100 / stream.Length));
status("Validate line " + lineNumber);
lineNumber++;
if (String.IsNullOrEmpty(line) || line.StartsWith("MTH") || line.StartsWith("#")) {
continue;
}
if (line.StartsWith(Section.Comment.Prefix)) {
comParser.Parse(lineNumber, line, errorList);
commentMap.Add(lineNumber, comParser.getComment());
continue;
}
Section section = MZTabFileParser.getSection(line);
MZTabError error;
if (section == null) {
error = new MZTabError(FormatErrorType.LinePrefix, lineNumber, MZTabFileParser.subString(line));
throw new MZTabException(error);
}
if (section.Level < highWaterMark) {
Section currentSection = Section.FindSection(highWaterMark);
error = new MZTabError(LogicalErrorType.LineOrder, lineNumber, currentSection.Name, section.Name);
throw new MZTabException(error);
}
highWaterMark = section.Level;
// There exists errors during checking metadata section.
if (highWaterMark == 1 && !errorList.IsNullOrEmpty()) {
break;
}
switch (highWaterMark) {
case 1:
// metadata section.
mtdParser.Parse(lineNumber, line, errorList);
break;
case 2:
if (prhParser != null) {
// header line only display once!
error = new MZTabError(LogicalErrorType.HeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
// protein header section
prhParser = new PRHLineParser(mtdParser.Metadata);
prhParser.Parse(lineNumber, line, errorList);
prtPositionMapping = new PositionMapping(prhParser.getFactory(), line);
// tell system to continue check protein data line.
highWaterMark = 3;
break;
case 3:
if (prhParser == null) {
// header line should be check first.
error = new MZTabError(LogicalErrorType.NoHeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
if (prtParser == null) {
prtParser = new PRTLineParser(prhParser.getFactory(), prtPositionMapping,
mtdParser.Metadata,
errorList);
}
prtParser.Parse(lineNumber, line, errorList);
proteinMap.Add(lineNumber, prtParser.getRecord(line));
break;
case 4:
if (pehParser != null) {
// header line only display once!
error = new MZTabError(LogicalErrorType.HeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
if (mtdParser.Metadata.MzTabType == MzTabType.Identification) {
errorList.Add(new MZTabError(LogicalErrorType.PeptideSection, lineNumber,
MZTabFileParser.subString(line)));
}
// peptide header section
pehParser = new PEHLineParser(mtdParser.Metadata);
pehParser.Parse(lineNumber, line, errorList);
pepPositionMapping = new PositionMapping(pehParser.getFactory(), line);
// tell system to continue check peptide data line.
highWaterMark = 5;
break;
case 5:
if (pehParser == null) {
// header line should be check first.
error = new MZTabError(LogicalErrorType.NoHeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
if (pepParser == null) {
pepParser = new PEPLineParser(pehParser.getFactory(), pepPositionMapping,
mtdParser.Metadata,
errorList);
}
pepParser.Parse(lineNumber, line, errorList);
peptideMap.Add(lineNumber, pepParser.getRecord(line));
break;
case 6:
if (pshParser != null) {
// header line only display once!
error = new MZTabError(LogicalErrorType.HeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
// psm header section
pshParser = new PSHLineParser(mtdParser.Metadata);
pshParser.Parse(lineNumber, line, errorList);
psmPositionMapping = new PositionMapping(pshParser.getFactory(), line);
// tell system to continue check peptide data line.
highWaterMark = 7;
break;
case 7:
if (pshParser == null) {
// header line should be check first.
error = new MZTabError(LogicalErrorType.NoHeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
if (psmParser == null) {
psmParser = new PSMLineParser(pshParser.getFactory(), psmPositionMapping,
mtdParser.Metadata,
errorList);
}
psmParser.Parse(lineNumber, line, errorList);
psmMap.Add(lineNumber, psmParser.getRecord(line));
break;
case 8:
if (smhParser != null) {
// header line only display once!
error = new MZTabError(LogicalErrorType.HeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
// small molecule header section
smhParser = new SMHLineParser(mtdParser.Metadata);
smhParser.Parse(lineNumber, line, errorList);
smlPositionMapping = new PositionMapping(smhParser.getFactory(), line);
// tell system to continue check small molecule data line.
highWaterMark = 9;
break;
case 9:
if (smhParser == null) {
// header line should be check first.
error = new MZTabError(LogicalErrorType.NoHeaderLine, lineNumber,
MZTabFileParser.subString(line));
throw new MZTabException(error);
}
if (smlParser == null) {
smlParser = new SMLLineParser(smhParser.getFactory(), smlPositionMapping,
mtdParser.Metadata,
errorList);
}
smlParser.Parse(lineNumber, line, errorList);
smallMoleculeMap.Add(lineNumber, smlParser.getRecord(line));
break;
}
}
} catch (Exception e) {
outstream.WriteLine("Line {0}: {1}", lineNumber, e.Message);
errorList.Add(new ParserError(lineNumber, e.Message));
}
if (reader != null) {
reader.Close();
}
if (errorList.IsNullOrEmpty()) {
MZTabFile mzTabFile = new MZTabFile(mtdParser.Metadata);
foreach (int id in commentMap.Keys) {
mzTabFile.addComment(id, commentMap[id]);
}
if (prhParser != null) {
MZTabColumnFactory proteinColumnFactory = prhParser.getFactory();
mzTabFile.setProteinColumnFactory(proteinColumnFactory);
foreach (int id in proteinMap.Keys) {
mzTabFile.addProtein(id, proteinMap[id]);
}
}
if (pehParser != null) {
MZTabColumnFactory peptideColumnFactory = pehParser.getFactory();
mzTabFile.setPeptideColumnFactory(peptideColumnFactory);
foreach (int id in peptideMap.Keys) {
mzTabFile.addPeptide(id, peptideMap[id]);
}
}
if (pshParser != null) {
MZTabColumnFactory psmColumnFactory = pshParser.getFactory();
mzTabFile.setPSMColumnFactory(psmColumnFactory);
foreach (int id in psmMap.Keys) {
mzTabFile.addPSM(id, psmMap[id]);
}
}
if (smhParser != null) {
MZTabColumnFactory smallMoleculeColumnFactory = smhParser.getFactory();
mzTabFile.setSmallMoleculeColumnFactory(smallMoleculeColumnFactory);
foreach (int id in smallMoleculeMap.Keys) {
mzTabFile.addSmallMolecule(id, smallMoleculeMap[id]);
}
}
}
}
public new void Parse(int lineNumber, String line, MZTabErrorList errorList)
{
base.Parse(lineNumber, line, errorList);
}