/**
* FASTQ format specification: http://maq.sourceforge.net/fastq.shtml
*/
static void load(IOAdapter* io, const U2DbiRef& dbiRef, const QVariantMap& hints, const GUrl& docUrl, QList<GObject*>& objects, U2OpStatus& os,
int gapSize, int predictedSize, QString& writeLockReason) {
DbiOperationsBlock opBlock(dbiRef, os);
CHECK_OP(os, );
Q_UNUSED(opBlock);
writeLockReason.clear();
bool merge = gapSize!=-1;
QByteArray sequence;
QByteArray qualityScores;
QStringList headers;
QSet<QString> uniqueNames;
QVector<U2Region> mergedMapping;
QByteArray gapSequence((merge ? gapSize : 0), 0);
sequence.reserve(predictedSize);
qualityScores.reserve(predictedSize);
// for lower case annotations
GObjectReference sequenceRef;
qint64 sequenceStart = 0;
U2SequenceImporter seqImporter(hints, true);
const QString folder = hints.value(DocumentFormat::DBI_FOLDER_HINT, U2ObjectDbi::ROOT_FOLDER).toString();
int seqNumber = 0;
int progressUpNum = 0;
const int objectsCountLimit = hints.contains(DocumentReadingMode_MaxObjectsInDoc) ? hints[DocumentReadingMode_MaxObjectsInDoc].toInt() : -1;
const bool settingsMakeUniqueName = !hints.value(DocumentReadingMode_DontMakeUniqueNames, false).toBool();
while (!os.isCoR()) {
//read header
QString sequenceName = readSequenceName(os, io, '@');
// check for eof while trying to read another FASTQ block
if (io->isEof()) {
break;
}
CHECK_OP_BREAK(os);
if(sequenceName.isEmpty()){
sequenceName = "Sequence";
}
if ((merge == false) || (seqNumber == 0)) {
QString objName = sequenceName;
if (settingsMakeUniqueName) {
objName = (merge) ? "Sequence" : TextUtils::variate(sequenceName, "_", uniqueNames);
objName.squeeze();
uniqueNames.insert(objName);
}
seqImporter.startSequence(dbiRef, folder, objName, false, os);
CHECK_OP_BREAK(os);
}
//read sequence
if (merge && sequence.length() > 0) {
seqImporter.addDefaultSymbolsBlock(gapSize,os);
sequenceStart += sequence.length();
sequenceStart+=gapSize;
CHECK_OP_BREAK(os);
}
sequence.clear();
readSequence(os, io, sequence);
MemoryLocker lSequence(os, qCeil(sequence.size()/(1000*1000)));
CHECK_OP_BREAK(os);
Q_UNUSED(lSequence);
seqImporter.addBlock(sequence.data(),sequence.length(),os);
CHECK_OP_BREAK(os);
QString qualSequenceName = readSequenceName(os, io, '+');
if (!qualSequenceName.isEmpty()) {
static const QString err = U2::FastqFormat::tr("Not a valid FASTQ file: %1, sequence name differs from quality scores name: %2 and %3");
CHECK_EXT_BREAK(sequenceName == qualSequenceName,
os.setError(err.arg(docUrl.getURLString()).arg(sequenceName).arg(qualSequenceName)));
}
// read qualities
qualityScores.clear();
readQuality(os, io, qualityScores, sequence.size());
CHECK_OP_BREAK(os);
static const QString err = U2::FastqFormat::tr("Not a valid FASTQ file: %1. Bad quality scores: inconsistent size.").arg(docUrl.getURLString());
CHECK_EXT_BREAK(sequence.length() == qualityScores.length(), os.setError(err));
seqNumber++;
progressUpNum++;
if (merge) {
headers.append(sequenceName);
mergedMapping.append(U2Region(sequenceStart, sequence.length() ));
}
else {
if (objectsCountLimit > 0 && objects.size() >= objectsCountLimit) {
os.setError(FastqFormat::tr("File \"%1\" contains too many sequences to be displayed. "
"However, you can process these data using instruments from the menu <i>Tools -> NGS data analysis</i> "
"or pipelines built with Workflow Designer.")
.arg(io->getURL().getURLString()));
break;
}
U2Sequence u2seq = seqImporter.finalizeSequenceAndValidate(os);
CHECK_OP_BREAK(os);
sequenceRef = GObjectReference(io->getURL().getURLString(), u2seq.visualName, GObjectTypes::SEQUENCE, U2EntityRef(dbiRef, u2seq.id));
U2SequenceObject* seqObj = new U2SequenceObject(u2seq.visualName, U2EntityRef(dbiRef, u2seq.id));
CHECK_EXT_BREAK(seqObj != NULL, os.setError("U2SequenceObject is NULL"));
seqObj->setQuality(DNAQuality(qualityScores));
objects << seqObj;
U1AnnotationUtils::addAnnotations(objects, seqImporter.getCaseAnnotations(), sequenceRef, NULL, hints);
}
if (PROGRESS_UPDATE_STEP == progressUpNum) {
progressUpNum = 0;
os.setProgress(io->getProgress());
}
}
CHECK_OP_EXT(os, qDeleteAll(objects); objects.clear(), );
bool emptyObjects = objects.isEmpty();
CHECK_EXT(!emptyObjects || merge, os.setError(Document::tr("Document is empty.")), );
SAFE_POINT(headers.size() == mergedMapping.size(), "headers <-> regions mapping failed!", );
if (!merge) {
return;
}
U2Sequence u2seq = seqImporter.finalizeSequenceAndValidate(os);
CHECK_OP(os,);
sequenceRef = GObjectReference(io->getURL().getURLString(), u2seq.visualName, GObjectTypes::SEQUENCE, U2EntityRef(dbiRef, u2seq.id));
U1AnnotationUtils::addAnnotations(objects, seqImporter.getCaseAnnotations(), sequenceRef, NULL, hints);
objects << new U2SequenceObject(u2seq.visualName, U2EntityRef(dbiRef, u2seq.id));
objects << DocumentFormatUtils::addAnnotationsForMergedU2Sequence(sequenceRef, dbiRef, headers, mergedMapping, hints);
if (headers.size() > 1) {
writeLockReason = DocumentFormat::MERGED_SEQ_LOCK;
}
}
