@Override
public SAMRecordPair getNextReadPair() {
// insert first read into dictionary by queryname
// insert second read into dictionary
// check if the dictionary length for that entry has both pairs
// if it is does return the read pair
// otherwise continue reading
// this way just return pairs as they are completed
// should be MUCH faster
// make sure to delete the entry after returning so that we dont have a memory leak
if (iterator.hasNext()) {
while (iterator.hasNext()) {
SAMRecord record = iterator.next();
countRead(record);
// skip if the read is unmapped, not properly paired or mate is unmapped
if (record.getReadUnmappedFlag() == true
|| record.getProperPairFlag() == false
|| record.getMateUnmappedFlag() == true) {
continue;
}
String query = record.getReadName();
// check if read mate has been read already
if (readBuffer.containsKey(query)) {
// if it has then return the pair
SAMRecordPair pair = readBuffer.get(query);
pair.addPair(record);
if (pair.bothPairsAligned() && pair.isValidPair()) {
// prevent memory leak by deleting keys that are no longer needed
readBuffer.remove(query);
return pair;
} else {
throw new RuntimeException(query + " is not properly mated");
}
} else {
// otherwise create an entry and store it by its query name
SAMRecordPair pair = new SAMRecordPair();
pair.addPair(record);
readBuffer.put(query, pair);
}
}
} else {
if (readBuffer.size() > 0) {
for (String key : readBuffer.keySet()) {
logger.info("No mate for for " + key);
}
throw new RuntimeException(
"No mates found for some reads please make sure all reads are properly paired");
}
}
return null;
}
@Override
public String next() {
if (sri != null) {
return sri.next().getReadString();
} else if (fri != null) {
return fri.next().getReadString();
} else if (fsi != null) {
return fsi.next();
}
return null;
}
private void scan() {
String chrom = "";
int chromStart = 0;
int width = 80;
int column2genomic[] = new int[width];
for (int i = 0; i < column2genomic.length; ++i) {
column2genomic[i] = chromStart + i;
}
GenomicSequence genomicSequence = new GenomicSequence(indexedFastaSequenceFile, chrom);
Element contig = new Element();
for (int i = 0; i < width; ++i) {}
Root root = new Root();
SAMRecordIterator iter = null;
while (iter.hasNext()) {
SAMRecord rec = iter.next();
}
}
@Override
public int doWork(String[] args) {
boolean compressed = false;
int maxRecordsInRAM = 100000;
long count = -1L;
File fileout = null;
com.github.lindenb.jvarkit.util.cli.GetOpt opt =
new com.github.lindenb.jvarkit.util.cli.GetOpt();
int c;
while ((c = opt.getopt(args, getGetOptDefault() + "o:n:N:T:b")) != -1) {
switch (c) {
case 'b':
compressed = true;
break;
case 'N':
maxRecordsInRAM = Integer.parseInt(opt.getOptArg());
break;
case 'n':
count = Long.parseLong(opt.getOptArg());
break;
case 'o':
fileout = new File(opt.getOptArg());
break;
case 'T':
this.addTmpDirectory(new File(opt.getOptArg()));
break;
default:
{
switch (handleOtherOptions(c, opt, null)) {
case EXIT_FAILURE:
return -1;
case EXIT_SUCCESS:
return 0;
default:
break;
}
}
}
}
if (count < -1L) // -1 == infinite
{
error("Bad count:" + count);
return -1;
}
SamReader samReader = null;
SAMRecordIterator iter = null;
SAMFileWriter samWriter = null;
Random random = new Random();
CloseableIterator<RandSamRecord> iter2 = null;
try {
SamFileReaderFactory.setDefaultValidationStringency(ValidationStringency.SILENT);
if (opt.getOptInd() == args.length) {
info("Reading from stdin");
samReader = SamFileReaderFactory.mewInstance().openStdin();
} else if (opt.getOptInd() + 1 == args.length) {
File filename = new File(args[opt.getOptInd()]);
info("Reading from " + filename);
samReader = SamFileReaderFactory.mewInstance().open(filename);
} else {
error("Illegal number of arguments.");
return -1;
}
SAMFileHeader header = samReader.getFileHeader();
header = header.clone();
header.setSortOrder(SortOrder.unsorted);
header.addComment("Processed with " + getProgramName() + " : " + getProgramCommandLine());
SAMFileWriterFactory sfw = new SAMFileWriterFactory();
sfw.setCreateIndex(false);
sfw.setCreateMd5File(false);
if (fileout == null) {
if (compressed) {
samWriter = sfw.makeBAMWriter(header, true, System.out);
} else {
samWriter = sfw.makeSAMWriter(header, true, System.out);
}
} else {
samWriter = sfw.makeSAMOrBAMWriter(header, true, fileout);
this.addTmpDirectory(fileout);
}
iter = samReader.iterator();
SAMSequenceDictionaryProgress progress =
new SAMSequenceDictionaryProgress(samReader.getFileHeader().getSequenceDictionary());
SortingCollection<RandSamRecord> sorter =
SortingCollection.newInstance(
RandSamRecord.class,
new RandSamRecordCodec(header),
new RandSamRecordComparator(),
maxRecordsInRAM,
getTmpDirectories());
sorter.setDestructiveIteration(true);
while (iter.hasNext()) {
RandSamRecord r = new RandSamRecord();
r.rand_index = random.nextInt();
r.samRecord = progress.watch(iter.next());
sorter.add(r);
}
iter.close();
iter = null;
sorter.doneAdding();
iter2 = sorter.iterator();
if (count == -1) {
while (iter2.hasNext()) {
samWriter.addAlignment(iter2.next().samRecord);
}
} else {
while (iter2.hasNext() && count > 0) {
samWriter.addAlignment(iter2.next().samRecord);
count--;
}
}
iter2.close();
iter2 = null;
sorter.cleanup();
progress.finish();
} catch (Exception e) {
error(e);
return -1;
} finally {
CloserUtil.close(iter);
CloserUtil.close(iter2);
CloserUtil.close(samReader);
CloserUtil.close(samWriter);
}
return 0;
}