int main(int argv, char* argc[]){
int error = 0;
while(!eof && readBuff() > 0){
if(findEndOfString() == -1){
error = 1;
clean();
} else{
int i;
while((i = findEndOfString()) != -1){
if(error == 1){
error = 0;
} else {
printReverse(0, i);
}
normalize(i + 1);
size -= i + 1;
}
}
}
return 0;
}
void PDBFormat::PDBParser::parseBioStruct3D(BioStruct3D& biostruct, U2OpStatus& ti) {
QByteArray readBuff(DocumentFormat::READ_BUFF_SIZE + 1, 0);
char* buf = readBuff.data();
qint64 len = 0;
bool firstCompndLine = true;
while (!ti.isCoR()) {
bool lineOk = true;
len = io->readUntil(buf, DocumentFormat::READ_BUFF_SIZE, TextUtils::LINE_BREAKS, IOAdapter::Term_Include, &lineOk);
if (len == 0) {
break;
}
// there could be no terminator if this is end of file, so we have to check for this
if (!lineOk && !io->isEof()) {
ti.setError(U2::PDBFormat::tr("Line is too long"));
return;
}
currentPDBLine = QString(QByteArray(buf, len));
ti.setProgress(io->getProgress() * 0.8);
if (currentPDBLine.startsWith("HEADER")) {
parseHeader(biostruct, ti);
continue;
}
if (currentPDBLine.startsWith("COMPND")) {
parseMacromolecularContent(firstCompndLine, ti);
firstCompndLine = false;
continue;
}
if (currentPDBLine.startsWith("SEQRES")) {
parseSequence(biostruct, ti);
continue;
}
if (currentPDBLine.startsWith("HELIX ") || currentPDBLine.startsWith("SHEET ")
|| currentPDBLine.startsWith("TURN ")) {
parseSecondaryStructure(biostruct, ti);
continue;
}
if (currentPDBLine.startsWith("ATOM ") || currentPDBLine.startsWith("HETATM")) {
parseAtom(biostruct, ti);
continue;
}
if (currentPDBLine.startsWith("TER")) {
++currentChainIndex;
continue;
}
if (currentPDBLine.startsWith("SPLIT ")) {
parseSplitSection(ti);
continue;
}
if (currentPDBLine.startsWith("MODEL")) {
currentChainIndex = 1;
parseModel(biostruct, ti);
continue;
}
if (currentPDBLine.startsWith("ENDMDL")) {
flagMultipleModels = true;
++currentModelIndex;
continue;
}
}
CHECK_OP(ti,);
if (!flagAtomRecordPresent) {
ti.setError(U2::PDBFormat::tr("Some mandatory records are absent"));
}
updateSecStructChainIndexes(biostruct);
}
static void load(IOAdapter* io, const U2DbiRef& dbiRef, const QVariantMap& fs, QList<GObject*>& objects,
int gapSize, QString& writeLockReason, U2OpStatus& os)
{
DbiOperationsBlock opBlock(dbiRef, os);
CHECK_OP(os, );
Q_UNUSED(opBlock);
static char fastaCommentStartChar = FastaFormat::FASTA_COMMENT_START_SYMBOL;
MemoryLocker memoryLocker(os, 1);
CHECK_OP(os, );
writeLockReason.clear();
QByteArray readBuff(DocumentFormat::READ_BUFF_SIZE + 1, 0);
char* buff = readBuff.data();
qint64 len = 0;
bool merge = gapSize != -1;
QStringList headers;
QSet<QString> uniqueNames;
QVector<U2Region> mergedMapping;
// for lower case annotations
GObjectReference sequenceRef;
//skip leading whites if present
bool lineOk = true;
static QBitArray nonWhites = ~TextUtils::WHITES;
io->readUntil(buff, DocumentFormat::READ_BUFF_SIZE, nonWhites, IOAdapter::Term_Exclude, &lineOk);
CHECK_EXT(!io->hasError(), os.setError(io->errorString()), );
U2SequenceImporter seqImporter(fs, true);
const QString folder = fs.value(DocumentFormat::DBI_FOLDER_HINT, U2ObjectDbi::ROOT_FOLDER).toString();
qint64 sequenceStart = 0;
int sequenceNumber = 0;
DbiConnection con(dbiRef, os);
bool headerReaded = false;
QStringList emptySeqNames;
const int objectsCountLimit = fs.contains(DocumentReadingMode_MaxObjectsInDoc) ? fs[DocumentReadingMode_MaxObjectsInDoc].toInt() : -1;
const bool settingsMakeUniqueName = !fs.value(DocumentReadingMode_DontMakeUniqueNames, false).toBool();
while (!os.isCoR()) {
//skip start comments and read header
if(!headerReaded){
do{
len = io->readLine(buff, DocumentFormat::READ_BUFF_SIZE);
CHECK_EXT(!io->hasError(), os.setError(io->errorString()), );
}while(buff[0] == fastaCommentStartChar && len > 0);
}
if (len == 0 && io->isEof()) { //end if stream
break;
}
CHECK_EXT(!io->hasError(), os.setError(io->errorString()), );
CHECK_EXT_BREAK(lineOk, os.setError(FastaFormat::tr("Line is too long")));
QString headerLine = QString(QByteArray(buff+1, len-1)).trimmed();
CHECK_EXT_BREAK(buff[0] == FastaFormat::FASTA_HEADER_START_SYMBOL, os.setError(FastaFormat::tr("First line is not a FASTA header")));
//read sequence
if (sequenceNumber == 0 || !merge) {
QString objName = headerLine;
if(objName.isEmpty()){
objName = "Sequence";
}
if (settingsMakeUniqueName) {
objName = (merge) ? "Sequence" : TextUtils::variate(objName, "_", uniqueNames);
objName.squeeze();
memoryLocker.tryAcquire(2*objName.size());
CHECK_OP_BREAK(os);
uniqueNames.insert(objName);
}
seqImporter.startSequence(os, dbiRef, folder, objName, false);
CHECK_OP_BREAK(os);
sequenceRef = GObjectReference(io->getURL().getURLString(), objName, GObjectTypes::SEQUENCE);
}
if (sequenceNumber >= 1 && merge) {
seqImporter.addDefaultSymbolsBlock(gapSize, os);
sequenceStart += gapSize;
CHECK_OP_BREAK(os);
}
int sequenceLen = 0;
while (!os.isCoR()) {
do {
len = io->readLine(buff, DocumentFormat::READ_BUFF_SIZE);
CHECK_EXT(!io->hasError(), os.setError(io->errorString()), );
} while (len <= 0 && !io->isEof());
CHECK_EXT(!io->hasError(), os.setError(io->errorString()), );
if (len <= 0 && io->isEof()) {
break;
}
CHECK_EXT(!io->hasError(), os.setError(io->errorString()), );
buff[len] = 0;
if(buff[0] != fastaCommentStartChar && buff[0] != FastaFormat::FASTA_HEADER_START_SYMBOL){
len = TextUtils::remove(buff, len, TextUtils::WHITES);
if(len > 0){
seqImporter.addBlock(buff, len, os);
sequenceLen += len;
}
}else if( buff[0] == FastaFormat::FASTA_HEADER_START_SYMBOL){
headerReaded = true;
break;
}
CHECK_OP_BREAK(os);
os.setProgress(io->getProgress());
}
if (merge) {
memoryLocker.tryAcquire(headerLine.size());
CHECK_OP_BREAK(os);
headers.append(headerLine);
mergedMapping.append(U2Region(sequenceStart, sequenceLen));
} else {
if (objectsCountLimit > 0 && objects.size() >= objectsCountLimit) {
os.setError(FastaFormat::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;
}
memoryLocker.tryAcquire(800);
CHECK_OP_BREAK(os);
U2Sequence seq = seqImporter.finalizeSequenceAndValidate(os);
if (os.hasError() && os.getError() == U2SequenceImporter::EMPTY_SEQUENCE_ERROR) {
os.setError("");
emptySeqNames << headerLine;
continue;
}
sequenceRef.entityRef = U2EntityRef(dbiRef, seq.id);
//TODO parse header
U2StringAttribute attr(seq.id, DNAInfo::FASTA_HDR, headerLine);
con.dbi->getAttributeDbi()->createStringAttribute(attr, os);
CHECK_OP_BREAK(os);
objects << new U2SequenceObject(seq.visualName, U2EntityRef(dbiRef, seq.id));
CHECK_OP_BREAK(os);
U1AnnotationUtils::addAnnotations(objects, seqImporter.getCaseAnnotations(), sequenceRef, NULL, fs);
}
sequenceStart += sequenceLen;
sequenceNumber++;
ioLog.trace(QString("Sequence #%1 is processed").arg(sequenceNumber));
}
CHECK_OP_EXT(os, qDeleteAll(objects); objects.clear(), );
CHECK_EXT(!objects.isEmpty() || merge, os.setError(Document::tr("Document is empty.")), );
SAFE_POINT(headers.size() == mergedMapping.size(), "headers <-> regions mapping failed!", );
ioLog.trace("All sequences are processed");
if (!emptySeqNames.isEmpty()) {
QString warningMessage;
warningMessage.append(FastaFormat::tr("Loaded sequences: %1.\n").arg(sequenceNumber));
warningMessage.append(FastaFormat::tr("Skipped sequences: %1.\n").arg(emptySeqNames.size()));
warningMessage.append(FastaFormat::tr("The following sequences are empty:\n%1").arg(emptySeqNames.join(",\n")));
os.addWarning(warningMessage);
}
if (!merge) {
return;
}
U2Sequence seq = seqImporter.finalizeSequenceAndValidate(os);
CHECK_OP(os, );
sequenceRef.entityRef = U2EntityRef(dbiRef, seq.id);
U1AnnotationUtils::addAnnotations(objects, seqImporter.getCaseAnnotations(), sequenceRef, NULL, fs);
objects << new U2SequenceObject(seq.visualName, U2EntityRef(dbiRef, seq.id));
objects << DocumentFormatUtils::addAnnotationsForMergedU2Sequence( sequenceRef, dbiRef, headers, mergedMapping, fs );
if (headers.size() > 1) {
writeLockReason = QObject::tr("Document sequences were merged");
}
}
