/**
* Uncompress a file to a temporary file.
*
* @param context Step context
* @param file file to uncompress
* @return the path to the uncompressed file
* @throws IOException if an error occurs while creating the uncompressed file
*/
private File uncompressFile(final TaskContext context, final DataFile file) throws IOException {
checkNotNull(file, "file argument cannot be null");
final DataFile realFile;
final DataProtocol protocol = file.getProtocol();
// Get the underlying file if the file protocol is a storage protocol
if (protocol instanceof StorageDataProtocol) {
realFile = ((StorageDataProtocol) protocol).getUnderLyingData(file);
} else {
realFile = file;
}
final File outputFile =
Files.createTempFile(
context.getLocalTempDirectory().toPath(),
MODULE_NAME + "-",
realFile.getExtension())
.toFile();
context.getLogger().fine("Uncompress/copy " + realFile + " to " + outputFile);
DataFiles.copy(realFile, new DataFile(outputFile));
return outputFile;
}
/**
* Uncompress a file if compressed.
*
* @param context the step context
* @param temporaryFiles the list of temporary files
* @param file the file to process
* @return the absolute path of the file (once uncompressed or not)
* @throws IOException if an error occurs while uncompressing the file
*/
private File uncompressFileIfNecessary(
final TaskContext context, List<File> temporaryFiles, final DataFile file)
throws IOException {
checkNotNull(file, "file argument cannot be null");
final File result;
if (file.getCompressionType() != CompressionType.NONE || !file.isLocalFile()) {
// Uncompress file
final File uncompressedFile = uncompressFile(context, file);
// Add the temporary file to the file of the file to remove
temporaryFiles.add(uncompressedFile);
result = uncompressedFile;
} else {
result = file.toFile();
}
return result;
}
/**
* Public constructor.
*
* @param file the workflow file
* @throws IOException if an error occurs while opening the file
*/
public CommandWorkflowParser(final DataFile file) throws IOException {
this(file.open());
}
@Override
public void merge(final Iterator<DataFile> inFileIterator, DataFile outFile) throws IOException {
final Multiset<String> counts = HashMultiset.create();
final Set<String> emptyCounts = new HashSet<>();
while (inFileIterator.hasNext()) {
// Get input file
final DataFile inFile = inFileIterator.next();
EoulsanLogger.getLogger().info("Merge " + inFile.getName() + " to " + outFile.getName());
boolean first = true;
try (BufferedReader reader = new BufferedReader(new InputStreamReader(inFile.open()))) {
String line = null;
while ((line = reader.readLine()) != null) {
// Do no handle header
if (first) {
first = false;
continue;
}
final int tabPos = line.indexOf('\t');
// Do not handle empty or invalid lines
if (tabPos == -1) {
continue;
}
try {
final String id = line.substring(0, tabPos).trim();
final int count = Integer.parseInt(line.substring(tabPos).trim());
if (count == 0) {
emptyCounts.add(id);
}
counts.add(id, count);
} catch (NumberFormatException e) {
// Do not handle parsing errors
}
}
}
}
// Write the result file
try (Writer writer = new OutputStreamWriter(outFile.create())) {
writer.write(ExpressionSplitter.EXPRESSION_FILE_HEADER);
// Write the non empty counts
for (Multiset.Entry<String> e : counts.entrySet()) {
final String id = e.getElement();
// Remove the id from empty counts
emptyCounts.remove(id);
// Write the entry
writer.write(id + '\t' + e.getCount() + '\n');
}
// Write the empty counts
for (String id : emptyCounts) {
writer.write(id + "\t0\n");
}
}
}
@Override
public TaskResult execute(final TaskContext context, final TaskStatus status) {
try {
final StringBuilder additionalArguments = new StringBuilder();
final Map<String, String> additionalDescription = new HashMap<>();
final List<File> temporaryFiles = new ArrayList<>();
// Search expression parameter is needed
if (this.useExpressionStepParameters) {
searchExpressionStepParameters(context);
}
if (this.gtfFile) {
// Get the annotation data
final Data annotationData =
context.getInputData(this.gtfFormat ? ANNOTATION_GTF : ANNOTATION_GFF);
// Get the annotation DataFile
final DataFile gffFile = annotationData.getDataFile();
final File gffFilePath = uncompressFileIfNecessary(context, temporaryFiles, gffFile);
additionalArguments.append("--sjdbGTFfile");
additionalArguments.append(' ');
additionalArguments.append(gffFilePath.getAbsolutePath());
additionalArguments.append(' ');
additionalDescription.put("sjdbGTFfile", computeMD5SumFile(gffFilePath));
}
if (this.overhang != null) {
additionalArguments.append("--sjdbOverhang");
additionalArguments.append(' ');
additionalArguments.append(this.overhang.toString());
additionalArguments.append(' ');
additionalDescription.put("sjdbOverhang", this.overhang.toString());
}
if (this.gtfTagExonParentTranscript != null) {
additionalArguments.append("--sjdbGTFtagExonParentTranscript");
additionalArguments.append(' ');
additionalArguments.append(this.gtfTagExonParentTranscript);
additionalArguments.append(' ');
additionalDescription.put(
"sjdbGTFtagExonParentTranscript", this.gtfTagExonParentTranscript);
}
if (this.gtfFeatureExon != null) {
additionalArguments.append("--sjdbGTFfeatureExon");
additionalArguments.append(' ');
additionalArguments.append(this.gtfFeatureExon);
additionalArguments.append(' ');
additionalDescription.put("sjdbGTFfeatureExon", this.gtfFeatureExon);
}
if (this.chrStartEndFilename != null) {
DataFile chrStartEndFile = new DataFile(this.chrStartEndFilename);
if (!chrStartEndFile.exists()) {
throw new IOException("Unable to read chromosome startend file: " + chrStartEndFile);
}
final File chrStartEndFilePath =
uncompressFileIfNecessary(context, temporaryFiles, chrStartEndFile);
additionalArguments.append("--sjdbFileChrStartEnd");
additionalArguments.append(' ');
additionalArguments.append(chrStartEndFilePath.getAbsolutePath());
additionalArguments.append(' ');
additionalDescription.put("sjdbFileChrStartEnd", computeMD5SumFile(chrStartEndFilePath));
}
if (this.genomeSAindexNbases != null) {
additionalArguments.append("--genomeSAindexNbases");
additionalArguments.append(' ');
additionalArguments.append(this.genomeSAindexNbases.toString());
additionalArguments.append(' ');
additionalDescription.put("genomeSAindexNbases", this.genomeSAindexNbases.toString());
}
if (this.genomeChrBinNbits != null) {
additionalArguments.append("--genomeChrBinNbits");
additionalArguments.append(' ');
additionalArguments.append(this.genomeChrBinNbits.toString());
additionalArguments.append(' ');
additionalDescription.put("genomeChrBinNbits", this.genomeChrBinNbits.toString());
}
status.setProgressMessage(this.mapper.getMapperName() + " index creation");
// Create the index
GenomeMapperIndexGeneratorModule.execute(
this.mapper,
context,
additionalArguments.toString(),
additionalDescription,
Common.getThreadsNumber(this.localThreads, this.maxLocalThreads));
// Remove temporary files
for (File temporaryFile : temporaryFiles) {
if (!temporaryFile.delete()) {
context.getLogger().warning("Cannot remove temporary file: " + temporaryFile);
}
}
} catch (IOException | EoulsanException e) {
return status.createTaskResult(e);
}
return status.createTaskResult();
}