int bamsplit(libmaus2::util::ArgInfo const & arginfo)
{
if ( isatty(STDIN_FILENO) )
{
::libmaus2::exception::LibMausException se;
se.getStream() << "Refusing read binary data from terminal, please redirect standard input to pipe or file." << std::endl;
se.finish();
throw se;
}
int const level = libmaus2::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
uint64_t const n = arginfo.getValue<int>("n",getDefaultN());
std::string const prefix = arginfo.getUnparsedValue("prefix",getDefaultFilePrefix(arginfo));
libmaus2::bambam::BamDecoder bamdec(std::cin);
libmaus2::bambam::BamAlignment const & algn = bamdec.getAlignment();
libmaus2::bambam::BamHeader const & header = bamdec.getHeader();
::libmaus2::bambam::BamHeader::unique_ptr_type uphead(updateHeader(arginfo,header));
libmaus2::aio::OutputStreamInstance::unique_ptr_type COS;
libmaus2::bambam::BamWriter::unique_ptr_type writer;
uint64_t c = 0;
uint64_t f = 0;
while ( bamdec.readAlignment() )
{
if ( c++ % n == 0 )
{
writer.reset();
if ( COS )
COS->flush();
COS.reset();
std::ostringstream fnostr;
fnostr << prefix << "_" << std::setw(6) << std::setfill('0') << f++ << std::setw(0) << ".bam";
std::string const fn = fnostr.str();
libmaus2::aio::OutputStreamInstance::unique_ptr_type tCOS(new libmaus2::aio::OutputStreamInstance(fn));
COS = UNIQUE_PTR_MOVE(tCOS);
libmaus2::bambam::BamWriter::unique_ptr_type twriter(new libmaus2::bambam::BamWriter(*COS,*uphead,level));
writer = UNIQUE_PTR_MOVE(twriter);
if ( verbose )
std::cerr << "[V] opened file " << fn << std::endl;
}
algn.serialise(writer->getStream());
}
writer.reset();
if ( COS )
COS->flush();
COS.reset();
return EXIT_SUCCESS;
}
::libmaus::bambam::BamHeader::unique_ptr_type updateHeader(
::libmaus::util::ArgInfo const & arginfo,
::libmaus::bambam::BamHeader const & header
)
{
std::string const headertext(header.text);
// add PG line to header
std::string const upheadtext = ::libmaus::bambam::ProgramHeaderLineSet::addProgramLine(
headertext,
"testbammergecoordinate", // ID
"testbammergecoordinate", // PN
arginfo.commandline, // CL
::libmaus::bambam::ProgramHeaderLineSet(headertext).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
::libmaus::bambam::BamHeader::unique_ptr_type uphead(new ::libmaus::bambam::BamHeader(upheadtext));
return UNIQUE_PTR_MOVE(uphead);
}
int main(int argc, char * argv[])
{
try
{
libmaus::util::ArgInfo const arginfo(argc,argv);
std::vector<std::string> const & inputfilenames = arginfo.restargs;
libmaus::bambam::BamMergeCoordinate bamdec(inputfilenames /* ,true */);
libmaus::bambam::BamAlignment const & algn = bamdec.getAlignment();
libmaus::bambam::BamHeader const & header = bamdec.getHeader();
::libmaus::bambam::BamHeader::unique_ptr_type uphead(updateHeader(arginfo,header));
libmaus::bambam::BamWriter writer(std::cout,*uphead);
libmaus::bambam::BamWriter::stream_type & bamoutstr = writer.getStream();
while ( bamdec.readAlignment() )
algn.serialise(bamoutstr);
}
catch(std::exception const & ex)
{
std::cerr << ex.what() << std::endl;
return EXIT_FAILURE;
}
}
int bamfilter(libmaus::util::ArgInfo const & arginfo)
{
uint64_t const minmapped = arginfo.getValue<uint64_t>("minmapped",getDefaultMinMapped());
uint64_t const maxmapped = arginfo.getValue<uint64_t>("maxmapped",getDefaultMaxMapped());
uint64_t const minlen = arginfo.getValue<uint64_t>("minlen",getDefaultMinLen());
int const level = libmaus::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
::libmaus::bambam::BamDecoder BD(std::cin);
::libmaus::bambam::BamHeader const & bamheader = BD.getHeader();
::libmaus::bambam::BamAlignment & alignment = BD.getAlignment();
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
::libmaus::bambam::BamHeader::unique_ptr_type uphead(libmaus::bambam::BamHeaderUpdate::updateHeader(arginfo,bamheader,"bamfilter",std::string(PACKAGE_VERSION)));
::libmaus::bambam::BamWriter::unique_ptr_type writer(new ::libmaus::bambam::BamWriter(std::cout,*uphead,level,Pcbs));
while ( BD.readAlignment() )
{
bool const a_1_mapped = !(alignment.getFlags() & ::libmaus::bambam::BamFlagBase::LIBMAUS_BAMBAM_FUNMAP);
bool const a_2_mapped = !(alignment.getFlags() & ::libmaus::bambam::BamFlagBase::LIBMAUS_BAMBAM_FMUNMAP);
bool const proper = (alignment.getFlags() & ::libmaus::bambam::BamFlagBase::LIBMAUS_BAMBAM_FPROPER_PAIR);
uint64_t const nummapped = (a_1_mapped?1:0)+(a_2_mapped?1:0)+(proper?1:0);
if (
nummapped >= minmapped &&
nummapped <= maxmapped &&
alignment.getLseq() >= static_cast<int64_t>(minlen)
)
alignment.serialise(writer->getStream());
}
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bamfixmatecoordinatesnamesorted(::libmaus::util::ArgInfo const & arginfo)
{
bool const verbose = arginfo.getValue<unsigned int>("verbose",getDefaultVerbose());
::libmaus::timing::RealTimeClock rtc; rtc.start();
// gzip compression level for output
int const level = libmaus::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
::libmaus::bambam::BamDecoder bamfile(std::cin);
std::string const headertext(bamfile.getHeader().text);
// add PG line to header
std::string const upheadtext = ::libmaus::bambam::ProgramHeaderLineSet::addProgramLine(
headertext,
"bamfixmatecoordinatesnamesorted", // ID
"bamfixmatecoordinatesnamesorted", // PN
arginfo.commandline, // CL
::libmaus::bambam::ProgramHeaderLineSet(headertext).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
::libmaus::bambam::BamHeader uphead(upheadtext);
if ( uphead.getSortOrder() != "queryname" )
uphead.changeSortOrder("unknown");
std::string const & finalheadtext = uphead.text;
::libmaus::bambam::BamHeader finalheader(finalheadtext);
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
::libmaus::bambam::BamWriter::unique_ptr_type writer(new ::libmaus::bambam::BamWriter(std::cout,finalheader,level,Pcbs));
std::pair< std::pair< ::libmaus::bambam::BamAlignment::shared_ptr_type, bool> , std::pair< ::libmaus::bambam::BamAlignment::shared_ptr_type, bool> >
P(std::pair< ::libmaus::bambam::BamAlignment::shared_ptr_type, bool>(::libmaus::bambam::BamAlignment::shared_ptr_type(),false),std::pair< ::libmaus::bambam::BamAlignment::shared_ptr_type, bool>(::libmaus::bambam::BamAlignment::shared_ptr_type(),false));
// try to read two alignments
P.first.second = bamfile.readAlignment();
if ( P.first.second )
{
P.first.first = bamfile.salignment();
P.second.second = P.first.second && bamfile.readAlignment();
P.second.first = bamfile.salignment();
}
uint64_t single = 0, pairs = 0;
uint64_t proc = 0;
uint64_t lastproc = 0;
uint64_t const mod = 1024*1024;
// while we have two alignments
while ( P.first.second && P.second.second )
{
uint32_t const aflags = P.first.first->getFlags();
uint32_t const bflags = P.second.first->getFlags();
// same name?
if (
(aflags & ::libmaus::bambam::BamFlagBase::LIBMAUS_BAMBAM_FPAIRED)
&&
(bflags & ::libmaus::bambam::BamFlagBase::LIBMAUS_BAMBAM_FPAIRED)
&&
(! strcmp(P.first.first->getName(),P.second.first->getName()))
)
{
unsigned int const amap = (aflags & ::libmaus::bambam::BamFlagBase::LIBMAUS_BAMBAM_FUNMAP) ? 0 : 1;
unsigned int const bmap = (bflags & ::libmaus::bambam::BamFlagBase::LIBMAUS_BAMBAM_FUNMAP) ? 0 : 1;
// std::cerr << "Pair " << bam1_qname(P.first.first->alignment) << " amap=" << amap << " bmap=" << bmap << std::endl;
// if exactly one of the two is mapped
if ( amap + bmap == 1 )
{
::libmaus::bambam::BamAlignment::shared_ptr_type mapped = amap ? P.first.first : P.second.first;
int64_t const tid = mapped->getRefID();
int64_t const pos = mapped->getPos();
// std::cerr << "tid=" << tid << " pos=" << pos << std::endl;
// set all tid and pos values
P.first.first->putRefId(tid);
P.first.first->putPos(pos);
P.first.first->putNextRefId(tid);
P.first.first->putNextPos(pos);
P.second.first->putRefId(tid);
P.second.first->putPos(pos);
P.second.first->putNextRefId(tid);
P.second.first->putNextPos(pos);
}
// write alignments
P.first.first->serialise(writer->getStream());
P.second.first->serialise(writer->getStream());
// read new alignments
P.first.second = bamfile.readAlignment();
if ( P.first.second )
{
P.first.first = bamfile.salignment();
P.second.second = bamfile.readAlignment();
P.second.first = bamfile.salignment();
}
pairs++;
proc += 2;
}
// different names
else
{
// write first alignment
P.first.first->serialise(writer->getStream());
// move second to first
std::swap(P.first,P.second);
// read new second
P.second.second = P.first.second && bamfile.readAlignment();
if ( P.second.second )
P.second.first = bamfile.salignment();
single++;
proc += 1;
}
if ( verbose && (proc/mod != lastproc/mod) )
{
std::cerr << proc << "\t" << single << "\t" << pairs << "\t" <<
proc/rtc.getElapsedSeconds() << "al/s"
<< std::endl;
lastproc = proc;
}
}
if ( P.first.second )
{
P.first.first->serialise(writer->getStream());
single++;
proc += 1;
}
if ( verbose )
std::cerr << proc << "\t" << single << "\t" << pairs << "\t" <<
proc/rtc.getElapsedSeconds() << "al/s"
<< std::endl;
assert ( ! P.second.second );
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bamsplitmod(libmaus::util::ArgInfo const & arginfo)
{
if ( isatty(STDIN_FILENO) )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Refusing read binary data from terminal, please redirect standard input to pipe or file." << std::endl;
se.finish();
throw se;
}
int const level = libmaus::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
uint64_t const div = arginfo.getValue<int>("div",getDefaultDiv());
std::string const prefix = arginfo.getUnparsedValue("prefix",getDefaultFilePrefix(arginfo));
if ( ! div )
{
::libmaus::exception::LibMausException se;
se.getStream() << "div cannot be 0." << std::endl;
se.finish();
throw se;
}
libmaus::bambam::BamDecoder bamdec(std::cin);
libmaus::bambam::BamAlignment const & algn = bamdec.getAlignment();
libmaus::bambam::BamHeader const & header = bamdec.getHeader();
::libmaus::bambam::BamHeader::unique_ptr_type uphead(updateHeader(arginfo,header));
libmaus::autoarray::AutoArray<libmaus::aio::CheckedOutputStream::unique_ptr_type> COS(div);
libmaus::autoarray::AutoArray<libmaus::bambam::BamWriter::unique_ptr_type> writers(div);
std::vector < std::string > filenames;
for ( uint64_t i = 0; i < div; ++i )
{
std::ostringstream ostr;
ostr << prefix << "_" << std::setw(6) << std::setfill('0') << i << std::setw(0) << ".bam";
libmaus::aio::CheckedOutputStream::unique_ptr_type tCOS(new libmaus::aio::CheckedOutputStream(ostr.str()));
COS[i] = UNIQUE_PTR_MOVE(tCOS);
libmaus::bambam::BamWriter::unique_ptr_type twriter(new libmaus::bambam::BamWriter(*COS[i],*uphead,level));
writers[i] = UNIQUE_PTR_MOVE(twriter);
}
uint64_t c = 0;
if ( verbose )
{
while ( bamdec.readAlignment() )
{
algn.serialise ( writers [ (c++) % div ] -> getStream() );
if ( ((c) & ((1ull<<20)-1)) == 0 )
std::cerr << "[V] " << c << std::endl;
}
std::cerr << "[V] " << c << std::endl;
}
else
{
while ( bamdec.readAlignment() )
algn.serialise ( writers [ (c++) % div ] -> getStream() );
}
for ( uint64_t i = 0; i < div; ++i )
{
writers[i].reset();
COS[i]->flush();
COS[i].reset();
}
return EXIT_SUCCESS;
}
int bam12auxmerge(::libmaus2::util::ArgInfo const & arginfo)
{
if ( isatty(STDIN_FILENO) )
{
::libmaus2::exception::LibMausException se;
se.getStream() << "Refusing to read binary data from terminal, please redirect standard input to pipe or file." << std::endl;
se.finish();
throw se;
}
if ( isatty(STDOUT_FILENO) )
{
::libmaus2::exception::LibMausException se;
se.getStream() << "Refusing write binary data to terminal, please redirect standard output to pipe or file." << std::endl;
se.finish();
throw se;
}
std::string const prefilename = arginfo.getRestArg<std::string>(0);
libmaus2::bambam::BamDecoder bampredec(prefilename);
int const level = libmaus2::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
int const ranksplit = arginfo.getValue<int>("ranksplit",getDefaultRankSplit());
int const rankstrip = arginfo.getValue<int>("rankstrip",getDefaultRankSplit());
int const clipreinsert = arginfo.getValue<int>("clipreinsert",getDefaultClipReinsert());
int const zztoname = arginfo.getValue<int>("zztoname",getDefaultZZToName());
int const sanity = arginfo.getValue<int>("sanity",getDefaultSanity());
uint64_t const mod = arginfo.getValue<int>("mod",getDefaultMod());
uint64_t const bmod = libmaus2::math::nextTwoPow(mod);
uint64_t const bmask = bmod-1;
libmaus2::autoarray::AutoArray<char> Aread;
::libmaus2::bambam::BamDecoder bamdec(std::cin,false);
::libmaus2::bambam::BamHeader const & header = bamdec.getHeader();
::libmaus2::bambam::BamHeader const & preheader = bampredec.getHeader();
std::string const headertext(header.text);
std::string const preheadertext(libmaus2::bambam::HeaderLine::removeSequenceLines(preheader.text));
libmaus2::bambam::ProgramHeaderLineSet headerlines(headertext);
libmaus2::bambam::ProgramHeaderLineSet preheaderlines(preheadertext);
std::vector<libmaus2::bambam::HeaderLine> allheaderlines = libmaus2::bambam::HeaderLine::extractLines(headertext);
std::string const lastid = preheaderlines.getLastIdInChain();
std::stack < std::pair<uint64_t,std::string> > pgtodo;
for ( uint64_t i = 0; i < headerlines.roots.size(); ++i )
pgtodo.push(std::pair<uint64_t,std::string>(headerlines.roots[i],lastid));
std::string upheadtext = preheadertext;
while ( pgtodo.size() )
{
uint64_t const hid = pgtodo.top().first;
std::string const PP = pgtodo.top().second;
pgtodo.pop();
libmaus2::bambam::HeaderLine const & line = headerlines.lines[hid];
// ID, PP, PN, CL, VN
std::string ID = (line.M.find("ID") != line.M.end()) ? line.M.find("ID")->second : "";
std::string const PN = (line.M.find("PN") != line.M.end()) ? line.M.find("PN")->second : "";
std::string const CL = (line.M.find("CL") != line.M.end()) ? line.M.find("CL")->second : "";
std::string const VN = (line.M.find("VN") != line.M.end()) ? line.M.find("VN")->second : "";
upheadtext = ::libmaus2::bambam::ProgramHeaderLineSet::addProgramLineRef(
upheadtext,
ID,
PN,
CL,
PP,
VN
);
if ( headerlines.edges.find(hid) != headerlines.edges.end() )
{
std::vector<uint64_t> const & children = headerlines.edges.find(hid)->second;
for ( uint64_t j = 0; j < children.size(); ++j )
pgtodo.push(std::pair<uint64_t,std::string>(children[j],ID));
}
}
/* copy SQ lines */
std::ostringstream sqconcstr;
sqconcstr << upheadtext;
for ( uint64_t i = 0; i < allheaderlines.size(); ++i )
if ( allheaderlines[i].type == "SQ" )
sqconcstr << allheaderlines[i].line << "\n";
upheadtext = sqconcstr.str();
::libmaus2::bambam::BamHeader uphead(upheadtext);
uphead.changeSortOrder("unknown");
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus2::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus2::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
::libmaus2::bambam::BamWriter::unique_ptr_type writer(new ::libmaus2::bambam::BamWriter(std::cout,uphead,level,Pcbs));
::libmaus2::bambam::BamAlignment & algn = bamdec.getAlignment();
::libmaus2::bambam::BamAlignment & prealgn = bampredec.getAlignment();
int64_t curid = -1;
libmaus2::autoarray::AutoArray< std::pair<uint8_t,uint8_t> > auxpre;
libmaus2::autoarray::AutoArray< std::pair<uint8_t,uint8_t> > auxnew;
libmaus2::bambam::BamAuxFilterVector auxfilter;
// helpers for clipReinsert
libmaus2::autoarray::AutoArray < std::pair<uint8_t,uint8_t> > auxtags;
libmaus2::autoarray::AutoArray<libmaus2::bambam::cigar_operation> cigop;
std::stack < libmaus2::bambam::cigar_operation > hardstack;
libmaus2::bambam::BamAlignment::D_array_type Tcigar;
libmaus2::bambam::BamAuxFilterVector bafv;
libmaus2::bambam::BamAuxFilterVector auxfilterout;
auxfilterout.set('q','s');
auxfilterout.set('q','q');
// helpers for zztoname
libmaus2::bambam::BamAuxFilterVector zzbafv;
zzbafv.set('z','z');
// tag filters for secondary/supplementary reads
libmaus2::bambam::BamAuxFilterVector auxfiltersec;
auxfiltersec.set('q','s');
auxfiltersec.set('q','q');
auxfiltersec.set('a','s');
auxfiltersec.set('a','h');
auxfiltersec.set('a','a');
auxfiltersec.set('a','f');
auxfiltersec.set('a','r');
auxfiltersec.set('a','3');
// loop over aligned BAM file
while ( bamdec.readAlignment() )
{
if ( ranksplit )
split12(algn);
// extract rank
char const * name = algn.getName();
char const * u1 = name;
bool ok = true;
uint64_t rank = 0;
while ( *u1 && *u1 != '_' )
{
rank *= 10;
rank += (*u1-'0');
ok = ok && isdigit(*u1);
++u1;
}
// unable to find rank? write out as is and continue
if ( ! ok )
{
algn.serialise(writer->getStream());
continue;
}
// loop over unaligned BAM file
while ( curid != static_cast<int64_t>(rank) )
{
bool const a_ok = bampredec.readAlignment();
if ( ! a_ok )
{
libmaus2::exception::LibMausException se;
se.getStream() << "Found unexpected EOF on file " << prefilename << std::endl;
se.finish();
throw se;
}
assert ( a_ok );
++curid;
if ( verbose && (! (curid & bmask)) )
std::cerr << "[V] " << (curid / bmod) << std::endl;
}
if ( verbose > 1 )
std::cerr << "Merging:\n" << algn.formatAlignment(header) << "\n" << prealgn.formatAlignment(preheader) << std::endl;
uint64_t pretagnum = prealgn.enumerateAuxTags(auxpre);
uint64_t newtagnum = algn.enumerateAuxTags(auxnew);
// do some sanity checking
if ( sanity )
{
// first do a name check
char const * prename = prealgn.getName();
u1++; // put on the first letter of readname
if ( verbose > 1 )
std::cerr << "Sanity: comparing " << name << " and " << prename << std::endl;
if ( !is_suffix(prename, u1) ) // names do not match
{
libmaus2::exception::LibMausException se;
se.getStream() << "Sanity check failed on read names, found " << name << " and " << prename << std::endl;
se.finish();
throw se;
}
// now the names match so try the flags
if ( !(algn.isPaired() == prealgn.isPaired() &&
algn.isRead1() == prealgn.isRead1() &&
algn.isRead2() == prealgn.isRead2()) )
{
libmaus2::exception::LibMausException se;
se.getStream() << "Sanity check failed on flags, " << std::endl
<< "Aligned " << name << " paired " << algn.isPaired() << " first " << algn.isRead1() << " last " << algn.isRead2() << std::endl
<< "Unaligned " << prename << " paired " << prealgn.isPaired() << " first " << prealgn.isRead1() << " last " << prealgn.isRead2() << std::endl;
se.finish();
throw se;
}
if ( verbose > 1 )
std::cerr << "Sanity check on flags: " << std::endl
<< "Aligned " << name << " paired " << algn.isPaired() << " first " << algn.isRead1() << " last " << algn.isRead2() << std::endl
<< "Unaligned " << prename << " paired " << prealgn.isPaired() << " first " << prealgn.isRead1() << " last " << prealgn.isRead2() << std::endl;
}
std::sort(auxpre.begin(),auxpre.begin()+pretagnum);
std::sort(auxnew.begin(),auxnew.begin()+newtagnum);
if ( verbose > 1 )
std::cerr << "pretagnum=" << pretagnum << " newtagnum=" << newtagnum << std::endl;
std::pair<uint8_t,uint8_t> * prec = auxpre.begin();
std::pair<uint8_t,uint8_t> * pree = prec + pretagnum;
std::pair<uint8_t,uint8_t> * preo = prec;
std::pair<uint8_t,uint8_t> * newc = auxnew.begin();
std::pair<uint8_t,uint8_t> * newe = newc + newtagnum;
std::pair<uint8_t,uint8_t> * newo = newc;
while ( prec != pree && newc != newe )
{
// pre which is not in new
if ( *prec < *newc )
{
*(preo++) = *(prec++);
}
// tag in both, drop pre
else if ( *prec == *newc )
{
*(newo++) = *(newc++);
prec++;
}
// new not in pre
else
{
*(newo++) = *(newc++);
}
}
while ( prec != pree )
*(preo++) = *(prec++);
while ( newc != newe )
*(newo++) = *(newc++);
pretagnum = preo-auxpre.begin();
newtagnum = newo-auxnew.begin();
for ( uint64_t i = 0; i < pretagnum; ++i )
auxfilter.set(auxpre[i].first,auxpre[i].second);
algn.copyAuxTags(prealgn, auxfilter);
for ( uint64_t i = 0; i < pretagnum; ++i )
auxfilter.clear(auxpre[i].first,auxpre[i].second);
if ( verbose > 1 )
{
std::cerr << "pretagnum=" << pretagnum << " newtagnum=" << newtagnum << std::endl;
std::cerr << "result: " << algn.formatAlignment(header) << std::endl;
}
if ( algn.isSecondary() || algn.isSupplementary() )
{
// adding adapter clip data to secondary/supplementary reads
// can lead to incorrect clip reinserts so remove these tags
algn.filterOutAux(auxfiltersec);
}
// copy QC fail flag from original file to aligner output
if ( prealgn.isQCFail() )
algn.putFlags( algn.getFlags() | libmaus2::bambam::BamFlagBase::LIBMAUS2_BAMBAM_FQCFAIL );
if ( rankstrip )
strip12(algn);
if ( clipreinsert )
clipReinsert(algn,auxtags,bafv,cigop,Tcigar,hardstack,auxfilterout);
if ( zztoname )
zzToRank(algn,zzbafv);
algn.serialise(writer->getStream());
}
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bamcat(libmaus::util::ArgInfo const & arginfo)
{
if ( isatty(STDOUT_FILENO) )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Refusing write binary data to terminal, please redirect standard output to pipe or file." << std::endl;
se.finish();
throw se;
}
int const level = arginfo.getValue<int>("level",getDefaultLevel());
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
switch ( level )
{
case Z_NO_COMPRESSION:
case Z_BEST_SPEED:
case Z_BEST_COMPRESSION:
case Z_DEFAULT_COMPRESSION:
break;
default:
{
::libmaus::exception::LibMausException se;
se.getStream()
<< "Unknown compression level, please use"
<< " level=" << Z_DEFAULT_COMPRESSION << " (default) or"
<< " level=" << Z_BEST_SPEED << " (fast) or"
<< " level=" << Z_BEST_COMPRESSION << " (best) or"
<< " level=" << Z_NO_COMPRESSION << " (no compression)" << std::endl;
se.finish();
throw se;
}
break;
}
std::vector<std::string> inputfilenames = arginfo.getPairValues("I");
for ( uint64_t i = 0; i < arginfo.restargs.size(); ++i )
inputfilenames.push_back(arginfo.restargs[i]);
libmaus::bambam::BamCat bamdec(inputfilenames /* ,true */);
libmaus::bambam::BamAlignment const & algn = bamdec.getAlignment();
libmaus::bambam::BamHeader const & header = bamdec.getHeader();
::libmaus::bambam::BamHeader::unique_ptr_type uphead(updateHeader(arginfo,header));
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
::libmaus::bambam::BamWriter::unique_ptr_type writer(new ::libmaus::bambam::BamWriter(std::cout,*uphead,level,Pcbs));
libmaus::bambam::BamWriter::stream_type & bamoutstr = writer->getStream();
if ( verbose )
{
uint64_t c = 0;
while ( bamdec.readAlignment() )
{
algn.serialise(bamoutstr);
if ( ((++c) & ((1ull<<20)-1)) == 0 )
std::cerr << "[V] " << c << std::endl;
}
std::cerr << "[V] " << c << std::endl;
}
else
while ( bamdec.readAlignment() )
algn.serialise(bamoutstr);
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bamsort(::libmaus::util::ArgInfo const & arginfo)
{
::libmaus::util::TempFileRemovalContainer::setup();
bool const inputisstdin = (!arginfo.hasArg("I")) || (arginfo.getUnparsedValue("I","-") == "-");
bool const outputisstdout = (!arginfo.hasArg("O")) || (arginfo.getUnparsedValue("O","-") == "-");
if ( isatty(STDIN_FILENO) && inputisstdin && (arginfo.getValue<std::string>("inputformat","bam") != "sam") )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Refusing to read binary data from terminal, please redirect standard input to pipe or file." << std::endl;
se.finish();
throw se;
}
if ( isatty(STDOUT_FILENO) && outputisstdout && (arginfo.getValue<std::string>("outputformat","bam") != "sam") )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Refusing write binary data to terminal, please redirect standard output to pipe or file." << std::endl;
se.finish();
throw se;
}
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
bool const disablevalidation = arginfo.getValue<int>("disablevalidation",getDefaultDisableValidation());
std::string const inputformat = arginfo.getUnparsedValue("inputformat",getDefaultInputFormat());
int const level = arginfo.getValue<int>("level",getDefaultLevel());
switch ( level )
{
case Z_NO_COMPRESSION:
case Z_BEST_SPEED:
case Z_BEST_COMPRESSION:
case Z_DEFAULT_COMPRESSION:
break;
default:
{
::libmaus::exception::LibMausException se;
se.getStream()
<< "Unknown compression level, please use"
<< " level=" << Z_DEFAULT_COMPRESSION << " (default) or"
<< " level=" << Z_BEST_SPEED << " (fast) or"
<< " level=" << Z_BEST_COMPRESSION << " (best) or"
<< " level=" << Z_NO_COMPRESSION << " (no compression)" << std::endl;
se.finish();
throw se;
}
break;
}
// prefix for tmp files
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfilenameout = tmpfilenamebase + "_bamsort";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfilenameout);
uint64_t blockmem = arginfo.getValue<uint64_t>("blockmb",getDefaultBlockSize())*1024*1024;
std::string const sortorder = arginfo.getValue<std::string>("SO","coordinate");
bool const fixmates = arginfo.getValue<int>("fixmates",getDefaultFixMates());
uint64_t sortthreads = arginfo.getValue<uint64_t>("sortthreads",getDefaultSortThreads());
// input decoder wrapper
libmaus::bambam::BamAlignmentDecoderWrapper::unique_ptr_type decwrapper(
libmaus::bambam::BamMultiAlignmentDecoderFactory::construct(
arginfo,false // put rank
)
);
::libmaus::bambam::BamAlignmentDecoder * ppdec = &(decwrapper->getDecoder());
::libmaus::bambam::BamAlignmentDecoder & dec = *ppdec;
if ( disablevalidation )
dec.disableValidation();
::libmaus::bambam::BamHeader const & header = dec.getHeader();
std::string const headertext(header.text);
// add PG line to header
std::string const upheadtext = ::libmaus::bambam::ProgramHeaderLineSet::addProgramLine(
headertext,
"bamsort", // ID
"bamsort", // PN
arginfo.commandline, // CL
::libmaus::bambam::ProgramHeaderLineSet(headertext).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
::libmaus::bambam::BamHeader uphead(upheadtext);
/*
* start index/md5 callbacks
*/
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
if ( sortorder != "queryname" )
uphead.changeSortOrder("coordinate");
else
uphead.changeSortOrder("queryname");
libmaus::bambam::BamBlockWriterBase::unique_ptr_type Pout ( libmaus::bambam::BamBlockWriterBaseFactory::construct(uphead, arginfo, Pcbs) );
if ( fixmates )
{
if ( sortorder != "queryname" )
{
::libmaus::bambam::BamEntryContainer< ::libmaus::bambam::BamAlignmentPosComparator >
BEC(blockmem,tmpfilenameout,sortthreads);
if ( verbose )
std::cerr << "[V] Reading alignments from source." << std::endl;
uint64_t incnt = 0;
// current alignment
libmaus::bambam::BamAlignment & curalgn = dec.getAlignment();
// previous alignment
libmaus::bambam::BamAlignment prevalgn;
// previous alignment valid
bool prevalgnvalid = false;
// MQ field filter
libmaus::bambam::BamAuxFilterVector MQfilter;
MQfilter.set("MQ");
while ( dec.readAlignment() )
{
if ( curalgn.isSecondary() || curalgn.isSupplementary() )
{
BEC.putAlignment(curalgn);
}
else if ( prevalgnvalid )
{
// different name
if ( strcmp(curalgn.getName(),prevalgn.getName()) )
{
BEC.putAlignment(prevalgn);
curalgn.swap(prevalgn);
}
// same name
else
{
libmaus::bambam::BamAlignment::fixMateInformation(prevalgn,curalgn,MQfilter);
BEC.putAlignment(prevalgn);
BEC.putAlignment(curalgn);
prevalgnvalid = false;
}
}
else
{
prevalgn.swap(curalgn);
prevalgnvalid = true;
}
if ( verbose && ( ( ++incnt & ((1ull<<20)-1) ) == 0 ) )
std::cerr << "[V] " << incnt << std::endl;
}
if ( prevalgnvalid )
{
BEC.putAlignment(prevalgn);
prevalgnvalid = false;
}
if ( verbose )
std::cerr << "[V] read " << incnt << " alignments" << std::endl;
// BEC.createOutput(std::cout, uphead, level, verbose, Pcbs);
BEC.createOutput(*Pout, verbose);
}
else
{
::libmaus::bambam::BamEntryContainer< ::libmaus::bambam::BamAlignmentNameComparator >
BEC(blockmem,tmpfilenameout,sortthreads);
if ( verbose )
std::cerr << "[V] Reading alignments from source." << std::endl;
uint64_t incnt = 0;
// current alignment
libmaus::bambam::BamAlignment & curalgn = dec.getAlignment();
// previous alignment
libmaus::bambam::BamAlignment prevalgn;
// previous alignment valid
bool prevalgnvalid = false;
// MQ field filter
libmaus::bambam::BamAuxFilterVector MQfilter;
MQfilter.set("MQ");
while ( dec.readAlignment() )
{
if ( curalgn.isSecondary() || curalgn.isSupplementary() )
{
BEC.putAlignment(curalgn);
}
else if ( prevalgnvalid )
{
// different name
if ( strcmp(curalgn.getName(),prevalgn.getName()) )
{
BEC.putAlignment(prevalgn);
curalgn.swap(prevalgn);
}
// same name
else
{
libmaus::bambam::BamAlignment::fixMateInformation(prevalgn,curalgn,MQfilter);
BEC.putAlignment(prevalgn);
BEC.putAlignment(curalgn);
prevalgnvalid = false;
}
}
else
{
prevalgn.swap(curalgn);
prevalgnvalid = true;
}
if ( verbose && ( ( ++incnt & ((1ull<<20)-1) ) == 0 ) )
std::cerr << "[V] " << incnt << std::endl;
}
if ( prevalgnvalid )
{
BEC.putAlignment(prevalgn);
prevalgnvalid = false;
}
if ( verbose )
std::cerr << "[V] read " << incnt << " alignments" << std::endl;
// BEC.createOutput(std::cout, uphead, level, verbose, Pcbs);
BEC.createOutput(*Pout, verbose);
}
}
else
{
if ( sortorder != "queryname" )
{
::libmaus::bambam::BamEntryContainer< ::libmaus::bambam::BamAlignmentPosComparator > BEC(blockmem,tmpfilenameout,sortthreads);
if ( verbose )
std::cerr << "[V] Reading alignments from source." << std::endl;
uint64_t incnt = 0;
while ( dec.readAlignment() )
{
BEC.putAlignment(dec.getAlignment());
incnt++;
if ( verbose && (incnt % (1024*1024) == 0) )
std::cerr << "[V] " << incnt/(1024*1024) << "M" << std::endl;
}
if ( verbose )
std::cerr << "[V] read " << incnt << " alignments" << std::endl;
// BEC.createOutput(std::cout, uphead, level, verbose, Pcbs);
BEC.createOutput(*Pout, verbose);
}
else
{
::libmaus::bambam::BamEntryContainer< ::libmaus::bambam::BamAlignmentNameComparator > BEC(blockmem,tmpfilenameout,sortthreads);
if ( verbose )
std::cerr << "[V] Reading alignments from source." << std::endl;
uint64_t incnt = 0;
while ( dec.readAlignment() )
{
BEC.putAlignment(dec.getAlignment());
incnt++;
if ( verbose && (incnt % (1024*1024) == 0) )
std::cerr << "[V] " << incnt/(1024*1024) << "M" << std::endl;
}
if ( verbose )
std::cerr << "[V] read " << incnt << " alignments" << std::endl;
// BEC.createOutput(std::cout, uphead, level, verbose, Pcbs);
BEC.createOutput(*Pout, verbose);
}
}
// flush encoder so callbacks see all output data
Pout.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bamreset(::libmaus2::util::ArgInfo const & arginfo)
{
if ( isatty(STDIN_FILENO) )
{
::libmaus2::exception::LibMausException se;
se.getStream() << "Refusing to read binary data from terminal, please redirect standard input to pipe or file." << std::endl;
se.finish();
throw se;
}
if ( isatty(STDOUT_FILENO) )
{
::libmaus2::exception::LibMausException se;
se.getStream() << "Refusing write binary data to terminal, please redirect standard output to pipe or file." << std::endl;
se.finish();
throw se;
}
int const level = libmaus2::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
int const resetsortorder = arginfo.getValue<int>("resetsortorder",getDefaultResetSortOrder());
::libmaus2::bambam::BamDecoder dec(std::cin,false);
::libmaus2::bambam::BamHeader const & header = dec.getHeader();
std::string headertext = header.text;
// no replacement header file given
if ( ! arginfo.hasArg("resetheadertext") )
{
// remove SQ lines
std::vector<libmaus2::bambam::HeaderLine> allheaderlines = libmaus2::bambam::HeaderLine::extractLines(headertext);
std::ostringstream upheadstr;
for ( uint64_t i = 0; i < allheaderlines.size(); ++i )
if ( allheaderlines[i].type != "SQ" )
upheadstr << allheaderlines[i].line << std::endl;
headertext = upheadstr.str();
}
// replace header given in file
else
{
std::string const headerfilename = arginfo.getUnparsedValue("resetheadertext","");
uint64_t const headerlen = libmaus2::util::GetFileSize::getFileSize(headerfilename);
libmaus2::aio::CheckedInputStream CIS(headerfilename);
libmaus2::autoarray::AutoArray<char> ctext(headerlen,false);
CIS.read(ctext.begin(),headerlen);
headertext = std::string(ctext.begin(),ctext.end());
}
// add PG line to header
headertext = libmaus2::bambam::ProgramHeaderLineSet::addProgramLine(
headertext,
"bamreset", // ID
"bamreset", // PN
arginfo.commandline, // CL
::libmaus2::bambam::ProgramHeaderLineSet(headertext).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
libmaus2::bambam::BamHeader uphead(headertext);
if ( resetsortorder )
uphead.changeSortOrder("unknown");
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus2::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
uint32_t const excludeflags = libmaus2::bambam::BamFlagBase::stringToFlags(
arginfo.getValue<std::string>("exclude",getDefaultExcludeFlags()));
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus2::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
::libmaus2::bambam::BamWriter::unique_ptr_type writer(new ::libmaus2::bambam::BamWriter(std::cout,uphead,level,Pcbs));
libmaus2::timing::RealTimeClock rtc; rtc.start();
libmaus2::bambam::BamAlignment & algn = dec.getAlignment();
uint64_t c = 0;
bool const resetaux = arginfo.getValue<int>("resetaux",getDefaultResetAux());
libmaus2::bambam::BamAuxFilterVector::unique_ptr_type const prgfilter(libmaus2::bambam::BamAuxFilterVector::parseAuxFilterList(arginfo));
libmaus2::bambam::BamAuxFilterVector const * rgfilter = prgfilter.get();
while ( dec.readAlignment() )
{
bool const keep = resetAlignment(algn,resetaux /* reset aux */,excludeflags,rgfilter);
if ( keep )
algn.serialise(writer->getStream());
if ( verbose && (++c & (1024*1024-1)) == 0 )
std::cerr << "[V] " << c/(1024*1024) << " " << (c / rtc.getElapsedSeconds()) << std::endl;
}
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bammaskflags(::libmaus2::util::ArgInfo const & arginfo)
{
uint64_t const maskpos = arginfo.getValue<uint64_t>("maskpos",0xFFFFUL);
uint64_t const maskneg = arginfo.getValue<uint64_t>("maskneg",getDefaultMaskNeg());
uint64_t const mask = maskpos & (~maskneg);
if ( mask )
{
std::cerr << "Keeping flags ";
for ( uint64_t i = 1; i <= ::libmaus2::bambam::BamFlagBase::LIBMAUS2_BAMBAM_FSUPPLEMENTARY; i <<= 1 )
if ( mask & i )
std::cerr << static_cast< ::libmaus2::bambam::BamFlagBase::bam_flags >(i) << ";";
std::cerr << std::endl;
std::cerr << "Erasing flags ";
for ( uint64_t i = 1; i <= ::libmaus2::bambam::BamFlagBase::LIBMAUS2_BAMBAM_FSUPPLEMENTARY; i <<= 1 )
if ( !(mask & i) )
std::cerr << static_cast< ::libmaus2::bambam::BamFlagBase::bam_flags >(i) << ";";
std::cerr << std::endl;
}
else
{
std::cerr << "Erasing all flags." << std::endl;
}
int const level = libmaus2::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
int const resetmatecoord = arginfo.getValue<int>("resetmatecoord",0);
::libmaus2::bambam::BamDecoder BD(std::cin);
::libmaus2::bambam::BamHeader const & bamheader = BD.getHeader();
std::string const headertext(bamheader.text);
// add PG line to header
std::string const upheadtext = ::libmaus2::bambam::ProgramHeaderLineSet::addProgramLine(
headertext,
"bammaskflags", // ID
"bammaskflags", // PN
arginfo.commandline, // CL
::libmaus2::bambam::ProgramHeaderLineSet(headertext).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
::libmaus2::bambam::BamHeader uphead(upheadtext);
::libmaus2::bambam::BamAlignment & alignment = BD.getAlignment();
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus2::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus2::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
::libmaus2::bambam::BamWriter::unique_ptr_type writer(new ::libmaus2::bambam::BamWriter(std::cout,uphead,level,Pcbs));
while ( BD.readAlignment() )
{
alignment.putFlags(alignment.getFlags() & mask);
if ( resetmatecoord )
{
alignment.putNextRefId(-1);
alignment.putNextPos(-1);
}
alignment.serialise(writer->getStream());
}
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
uint64_t bamheaderfilter(libmaus2::util::ArgInfo const & arginfo)
{
std::string const inputfilename = arginfo.getUnparsedValue("I","");
if ( ! inputfilename.size() || inputfilename == "-" )
{
::libmaus2::exception::LibMausException se;
se.getStream() << "No input filename given, please set the I key appropriately." << std::endl;
se.finish();
throw se;
}
libmaus2::bitio::IndexedBitVector::unique_ptr_type usedrefseq;
libmaus2::bitio::IndexedBitVector::unique_ptr_type usedrg;
libmaus2::bambam::BamHeader::unique_ptr_type uheader;
getUsedRefSeqs(arginfo,usedrefseq,usedrg,uheader);
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus2::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus2::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
/*
* end md5/index callbacks
*/
std::string headertext(uheader->text);
std::vector<libmaus2::bambam::HeaderLine> hl = libmaus2::bambam::HeaderLine::extractLines(headertext);
std::ostringstream headertextostr;
uint64_t rscnt = 0;
uint64_t rgcnt = 0;
for ( uint64_t i = 0; i < hl.size(); ++i )
{
if ( hl[i].type == "SQ" )
{
if ( usedrefseq->get(rscnt) )
headertextostr << hl[i].line << std::endl;
rscnt += 1;
}
else if ( hl[i].type == "RG" )
{
if ( usedrg->get(rgcnt) )
headertextostr << hl[i].line << std::endl;
rgcnt += 1;
}
else
{
headertextostr << hl[i].line << std::endl;
}
}
headertext = headertextostr.str();
// add PG line to header
std::string const upheadtext = ::libmaus2::bambam::ProgramHeaderLineSet::addProgramLine(
headertext,
"bamheaderfilter", // ID
"bamheaderfilter", // PN
arginfo.commandline, // CL
::libmaus2::bambam::ProgramHeaderLineSet(headertext).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
::libmaus2::bambam::BamHeader uphead(upheadtext);
libmaus2::bambam::BamBlockWriterBase::unique_ptr_type Pout ( libmaus2::bambam::BamBlockWriterBaseFactory::construct(uphead, arginfo, &cbs) );
// input decoder wrapper
libmaus2::bambam::BamAlignmentDecoderWrapper::unique_ptr_type decwrapper(
libmaus2::bambam::BamMultiAlignmentDecoderFactory::construct(
arginfo,false // put rank
)
);
::libmaus2::bambam::BamAlignmentDecoder * ppdec = &(decwrapper->getDecoder());
::libmaus2::bambam::BamAlignmentDecoder & dec = *ppdec;
::libmaus2::bambam::BamAlignment & algn = dec.getAlignment();
while ( dec.readAlignment() )
{
if ( (!algn.isPaired()) && algn.isMapped() )
{
assert ( algn.getRefID() >= 0 );
assert ( algn.getRefID() < static_cast<int64_t>(usedrefseq->size()) );
assert ( usedrefseq->get(algn.getRefID()) );
assert ( usedrefseq->rank1(algn.getRefID())-1 < uphead.getNumRef() );
algn.putRefId(usedrefseq->rank1(algn.getRefID())-1);
}
if ( algn.isPaired() && algn.isMapped() )
{
assert ( algn.getRefID() >= 0 );
assert ( algn.getRefID() < static_cast<int64_t>(usedrefseq->size()) );
assert ( usedrefseq->get(algn.getRefID()) );
assert ( usedrefseq->rank1(algn.getRefID())-1 < uphead.getNumRef() );
algn.putRefId(usedrefseq->rank1(algn.getRefID())-1);
}
if ( algn.isPaired() && algn.isMateMapped() )
{
assert ( algn.getNextRefID() >= 0 );
assert ( algn.getNextRefID() < static_cast<int64_t>(usedrefseq->size()) );
assert ( usedrefseq->get(algn.getNextRefID()) );
assert ( usedrefseq->rank1(algn.getNextRefID())-1 < uphead.getNumRef() );
algn.putNextRefId(usedrefseq->rank1(algn.getNextRefID())-1);
}
// erase unmapped refid and pos
if ( algn.isUnmap() )
{
algn.putRefId(-1);
algn.putPos(-1);
}
if ( algn.isMateUnmap() )
{
algn.putNextRefId(-1);
algn.putNextPos(-1);
}
Pout->writeAlignment(algn);
}
Pout.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return 0;
}
int bamvalidateTemplate(::libmaus2::util::ArgInfo const & arginfo)
{
libmaus2::timing::RealTimeClock rtc; rtc.start();
bool const verbose = arginfo.getValue("verbose",getDefaultVerbose());
bool const basequalhist = arginfo.getValue("basequalhist",getDefaultBaseQualHist());
libmaus2::bambam::BamAlignmentDecoderWrapper::unique_ptr_type decwrapper(
libmaus2::bambam::BamMultiAlignmentDecoderFactory::construct(
arginfo,false // put rank
)
);
::libmaus2::bambam::BamAlignmentDecoder * ppdec = &(decwrapper->getDecoder());
::libmaus2::bambam::BamAlignmentDecoder & dec = *ppdec;
::libmaus2::bambam::BamHeader const & header = dec.getHeader();
::libmaus2::bambam::BamAlignment const & algn = dec.getAlignment();
// add PG line to header
std::string const upheadtext = ::libmaus2::bambam::ProgramHeaderLineSet::addProgramLine(
header.text,
"bamvalidate", // ID
"bamvalidate", // PN
arginfo.commandline, // CL
::libmaus2::bambam::ProgramHeaderLineSet(header.text).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
::libmaus2::bambam::BamHeader uphead(upheadtext);
/*
* start index/md5 callbacks and alignment writer
*/
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
libmaus2::bambam::BamBlockWriterBase::unique_ptr_type Pout;
if ( passthrough )
{
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus2::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus2::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
libmaus2::bambam::BamBlockWriterBase::unique_ptr_type Tout (
libmaus2::bambam::BamBlockWriterBaseFactory::construct(uphead, arginfo, Pcbs)
);
Pout = UNIQUE_PTR_MOVE(Tout);
}
libmaus2::autoarray::AutoArray<char> lastvalidname(256); // max valid read name is 255 bytes
uint64_t alsok = 0;
::libmaus2::autoarray::AutoArray<char> qual;
libmaus2::autoarray::AutoArray<uint64_t> H(static_cast<uint64_t>(std::numeric_limits<uint8_t>::max())+1);
std::fill(H.begin(),H.end(),0ull);
try
{
while ( dec.readAlignment() )
{
if ( passthrough )
Pout->writeAlignment(algn);
if ( basequalhist )
{
uint64_t const l = algn.getLseq();
uint8_t const * Qc = libmaus2::bambam::BamAlignmentDecoderBase::getQual(algn.D.begin());
uint8_t const * const Qe = Qc + l;
while ( Qc != Qe )
H[*(Qc++)]++;
}
uint64_t const lname = algn.getLReadName();
char const * name = algn.getName();
std::copy(name,name+lname+1,lastvalidname.begin());
alsok += 1;
}
}
catch(std::exception const & ex)
{
std::cerr << "[E] name of last valid alignment was " << lastvalidname.begin() << std::endl;
std::cerr << "[E] read " << alsok << " valid alignments" << std::endl;
throw;
}
Pout.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
if ( verbose )
std::cerr << "[V] checked " << alsok << " alignments in " << rtc.formatTime(rtc.getElapsedSeconds())
<< " (" << alsok / rtc.getElapsedSeconds() << " al/s)" << std::endl;
if ( basequalhist )
{
uint64_t const s = std::accumulate(H.begin(),H.end(),0ull);
uint64_t a = 0;
uint64_t minq = std::numeric_limits<uint64_t>::max();
uint64_t maxq = 0;
for ( uint64_t i = 0; i < H.size(); ++i )
if ( H[i] )
{
minq = std::min(minq,i);
maxq = std::max(maxq,i);
a += H[i];
std::cerr
<< "[H]\t" << i << "\t";
if ( ( static_cast<uint64_t>(i+33) < static_cast<uint64_t>(std::numeric_limits<char>::max()) && isprint(i+33)) )
std::cerr << static_cast<char>(i+33);
std::cerr << "\t"
<< H[i] << "\t"
<< (H[i] / static_cast<double>(s)) << "\t"
<< (a / static_cast<double>(s))
<< std::endl;
}
if ( s )
{
std::cerr << "[H]\tmin\t" << minq << "\t";
if ( ( static_cast<uint64_t>(minq+33) < static_cast<uint64_t>(std::numeric_limits<char>::max()) && isprint(minq+33)) )
std::cerr << static_cast<char>(minq+33);
std::cerr << std::endl;
std::cerr << "[H]\tmax\t" << maxq << "\t";
if ( ( static_cast<uint64_t>(maxq+33) < static_cast<uint64_t>(std::numeric_limits<char>::max()) && isprint(maxq+33)) )
std::cerr << static_cast<char>(maxq+33);
std::cerr << std::endl;
}
}
return EXIT_SUCCESS;
}
int bamrecalculatecigar(libmaus2::util::ArgInfo const & arginfo)
{
if ( isatty(STDOUT_FILENO) )
{
::libmaus2::exception::LibMausException se;
se.getStream() << "Refusing write binary data to terminal, please redirect standard output to pipe or file." << std::endl;
se.finish();
throw se;
}
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
// input decoder wrapper
libmaus2::bambam::BamAlignmentDecoderWrapper::unique_ptr_type decwrapper(
libmaus2::bambam::BamMultiAlignmentDecoderFactory::construct(
arginfo,false // put rank
)
);
libmaus2::bambam::BamAlignmentDecoder & bamdec = decwrapper->getDecoder();
libmaus2::bambam::BamAlignment & algn = bamdec.getAlignment();
libmaus2::bambam::BamHeader const & header = bamdec.getHeader();
::libmaus2::bambam::BamHeader::unique_ptr_type uphead(updateHeader(arginfo,header));
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus2::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus2::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
libmaus2::bambam::BamBlockWriterBase::unique_ptr_type writer(
libmaus2::bambam::BamBlockWriterBaseFactory::construct(*uphead, arginfo, Pcbs)
);
libmaus2::autoarray::AutoArray<libmaus2::bambam::cigar_operation> cigopin;
libmaus2::autoarray::AutoArray<char> readdata;
libmaus2::bambam::BamAlignment::D_array_type T;
if ( ! arginfo.hasArg("reference") )
{
libmaus2::exception::LibMausException se;
se.getStream() << "reference key is missing." << std::endl;
se.finish();
throw se;
}
std::string const reference = arginfo.getUnparsedValue("reference","");
if ( ! libmaus2::util::GetFileSize::fileExists(reference) )
{
libmaus2::exception::LibMausException se;
se.getStream() << "file " << reference << " does not exist." << std::endl;
se.finish();
throw se;
}
libmaus2::fastx::FastAIndex::unique_ptr_type FAindex(libmaus2::fastx::FastAIndex::load(reference + ".fai"));
libmaus2::aio::InputStreamInstance FAISI(reference);
uint64_t c = 0;
libmaus2::autoarray::AutoArray<char> ref;
int64_t refloaded = -1;
while ( bamdec.readAlignment() )
{
if ( algn.isMapped() )
{
assert ( algn.getRefID() >= 0 );
if ( algn.getRefID() != refloaded )
{
if ( algn.getRefID() < refloaded )
{
libmaus2::exception::LibMausException lme;
lme.getStream() << "bamrecalculatecigar: file is not sorted by coordinate" << std::endl;
lme.finish();
throw lme;
}
ref = FAindex->readSequence(FAISI,algn.getRefID());
refloaded = algn.getRefID();
}
uint64_t const numcig = libmaus2::bambam::BamAlignmentDecoderBase::recalculateCigar(
algn.D.begin(),
ref.begin() + algn.getPos(),
cigopin,
readdata
);
algn.replaceCigarString(cigopin,numcig,T);
}
writer->writeAlignment(algn);
if ( ((++c) & ((1ull<<20)-1)) == 0 && verbose )
std::cerr << "[V] " << c << std::endl;
}
if ( verbose )
std::cerr << "[V] " << c << std::endl;
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bamfilterflags(::libmaus2::util::ArgInfo const & arginfo)
{
uint32_t const excludeflags = libmaus2::bambam::BamFlagBase::stringToFlags(arginfo.getValue<std::string>("exclude",""));
std::cerr << "[V] excluding " << excludeflags << std::endl;
int const level = libmaus2::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",Z_DEFAULT_COMPRESSION));
uint64_t const numthreads = arginfo.getValue<uint64_t>("numthreads",1);
uint64_t cnt = 0;
uint64_t kept = 0;
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus2::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus2::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus2::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus2::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus2::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
if ( numthreads == 1 )
{
::libmaus2::bambam::BamDecoder BD(std::cin);
::libmaus2::bambam::BamHeader const & bamheader = BD.getHeader();
::libmaus2::bambam::BamHeader::unique_ptr_type uphead(libmaus2::bambam::BamHeaderUpdate::updateHeader(arginfo,bamheader,"bamfilterflags",std::string(PACKAGE_VERSION)));
::libmaus2::bambam::BamAlignment & alignment = BD.getAlignment();
::libmaus2::bambam::BamWriter::unique_ptr_type writer(new ::libmaus2::bambam::BamWriter(std::cout,*uphead,level,Pcbs));
for ( ; BD.readAlignment(); ++cnt )
{
if ( cnt % (1024*1024) == 0 )
std::cerr << "[V] processed " << cnt << " kept " << kept << " removed " << (cnt-kept) << std::endl;
if ( ! (alignment.getFlags() & excludeflags) )
{
alignment.serialise(writer->getStream());
++kept;
}
}
std::cerr << "[V] " << cnt << std::endl;
}
else
{
::libmaus2::bambam::BamHeaderUpdate UH(arginfo,"bamfilterflags",std::string(PACKAGE_VERSION));
libmaus2::bambam::BamParallelRewrite BPR(std::cin,UH,std::cout,Z_DEFAULT_COMPRESSION,numthreads,4 /* blocks per thread */,Pcbs);
libmaus2::bambam::BamAlignmentDecoder & dec = BPR.getDecoder();
libmaus2::bambam::BamParallelRewrite::writer_type & writer = BPR.getWriter();
libmaus2::bambam::BamAlignment const & algn = dec.getAlignment();
for ( ; dec.readAlignment(); ++cnt )
{
if ( cnt % (1024*1024) == 0 )
std::cerr << "[V] processed " << cnt << " kept " << kept << " removed " << (cnt-kept) << std::endl;
if ( ! (algn.getFlags() & excludeflags) )
{
algn.serialise(writer.getStream());
++kept;
}
}
std::cerr << "[V] " << cnt << std::endl;
}
std::cerr << "[V] kept " << kept << " removed " << cnt-kept << std::endl;
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bamfixmatecoordinates(::libmaus::util::ArgInfo const & arginfo)
{
::libmaus::util::TempFileRemovalContainer::setup();
::libmaus::timing::RealTimeClock rtc; rtc.start();
bool const verbose = arginfo.getValue<unsigned int>("verbose",getDefaultVerbose());
unsigned int const colhashbits = arginfo.getValue<unsigned int>("colhashbits",getDefaultColHashBits());
unsigned int const collistsize = arginfo.getValue<unsigned int>("collistsize",getDefaultColListSize());
int const level = arginfo.getValue<int>("level",getDefaultLevel());
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
switch ( level )
{
case Z_NO_COMPRESSION:
case Z_BEST_SPEED:
case Z_BEST_COMPRESSION:
case Z_DEFAULT_COMPRESSION:
break;
default:
{
::libmaus::exception::LibMausException se;
se.getStream()
<< "Unknown compression level, please use"
<< " level=" << Z_DEFAULT_COMPRESSION << " (default) or"
<< " level=" << Z_BEST_SPEED << " (fast) or"
<< " level=" << Z_BEST_COMPRESSION << " (best) or"
<< " level=" << Z_NO_COMPRESSION << " (no compression)" << std::endl;
se.finish();
throw se;
}
break;
}
std::string const tmpfilename = tmpfilenamebase + "_bamcollate";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfilename);
::libmaus::bambam::CollatingBamDecoder CBD(std::cin,tmpfilename,false /* put rank */,colhashbits/*hash bits*/,collistsize/*size of output list*/);
::libmaus::bambam::BamFormatAuxiliary auxdata;
::libmaus::bambam::BamHeader const & bamheader = CBD.getHeader();
// add PG line to header
std::string const upheadtext = ::libmaus::bambam::ProgramHeaderLineSet::addProgramLine(
bamheader.text,
"bamfixmatecoordinates", // ID
"bamfixmatecoordinates", // PN
arginfo.commandline, // CL
::libmaus::bambam::ProgramHeaderLineSet(bamheader.text).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
::libmaus::bambam::BamHeader uphead(upheadtext);
if ( uphead.getSortOrder() != "queryname" )
uphead.changeSortOrder("unknown");
/*
* start index/md5 callbacks
*/
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
// setup bam writer
::libmaus::bambam::BamWriter::unique_ptr_type writer(new ::libmaus::bambam::BamWriter(std::cout,uphead,level,Pcbs));
#if 0
::libmaus::bambam::ProgramHeaderLineSet PHLS(bamheader.text);
std::cerr << "Last id in PG chain: " << PHLS.getLastIdInChain() << std::endl;
#endif
// std::cout << bamheader.text;
typedef ::libmaus::bambam::CollatingBamDecoder::alignment_ptr_type alignment_ptr_type;
std::pair<alignment_ptr_type,alignment_ptr_type> P;
uint64_t const mod = 1024*1024;
uint64_t proc = 0;
uint64_t lastproc = 0;
uint64_t paircnt = 0;
while ( CBD.tryPair(P) )
{
uint64_t const mapcnt = getMapCnt(P.first) + getMapCnt(P.second);
if ( mapcnt == 1 )
{
int32_t refid = -1;
int32_t pos = -1;
if ( P.first )
{
refid = P.first->getRefID();
pos = P.first->getPos();
}
else
{
assert ( P.second );
refid = P.second->getRefID();
pos = P.second->getPos();
}
P.first->putRefId(refid);
P.first->putPos(pos);
P.first->putNextRefId(refid);
P.first->putNextPos(pos);
P.second->putRefId(refid);
P.second->putPos(pos);
P.second->putNextRefId(refid);
P.second->putNextPos(pos);
}
if ( P.first )
{
P.first->serialise(writer->getStream());
++proc;
}
if ( P.second )
{
P.second->serialise(writer->getStream());
++proc;
}
if ( P.first && P.second )
{
paircnt++;
}
if ( verbose && (proc/mod != lastproc/mod) )
{
std::cerr
<< "Processed " << proc << " fragments, " << paircnt << " pairs, "
<< proc/rtc.getElapsedSeconds() << " al/s"
<< std::endl;
lastproc = proc;
}
}
if ( verbose )
std::cerr
<< "Processed " << proc << " fragments, " << paircnt << " pairs, "
<< proc/rtc.getElapsedSeconds() << " al/s"
<< std::endl;
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
int bam12split(::libmaus::util::ArgInfo const & arginfo)
{
::libmaus::util::TempFileRemovalContainer::setup();
if ( isatty(STDIN_FILENO) )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Refusing to read binary data from terminal, please redirect standard input to pipe or file." << std::endl;
se.finish();
throw se;
}
if ( isatty(STDOUT_FILENO) )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Refusing write binary data to terminal, please redirect standard output to pipe or file." << std::endl;
se.finish();
throw se;
}
if ( arginfo.hasArg("keep") && arginfo.hasArg("remove") )
{
::libmaus::exception::LibMausException se;
se.getStream() << "The keep and remove keys are mutually exclusive." << std::endl;
se.finish();
throw se;
}
int const level = libmaus::bambam::BamBlockWriterBaseFactory::checkCompressionLevel(arginfo.getValue<int>("level",getDefaultLevel()));
int const verbose = arginfo.getValue<int>("verbose",getDefaultVerbose());
::libmaus::bambam::BamDecoder dec(std::cin,false);
::libmaus::bambam::BamHeader const & header = dec.getHeader();
std::string const headertext(header.text);
// add PG line to header
std::string const upheadtext = ::libmaus::bambam::ProgramHeaderLineSet::addProgramLine(
headertext,
"bam12split", // ID
"bam12split", // PN
arginfo.commandline, // CL
::libmaus::bambam::ProgramHeaderLineSet(headertext).getLastIdInChain(), // PP
std::string(PACKAGE_VERSION) // VN
);
// construct new header
libmaus::bambam::BamHeader uphead(upheadtext);
uphead.changeSortOrder("unknown");
/*
* start index/md5 callbacks
*/
std::string const tmpfilenamebase = arginfo.getValue<std::string>("tmpfile",arginfo.getDefaultTmpFileName());
std::string const tmpfileindex = tmpfilenamebase + "_index";
::libmaus::util::TempFileRemovalContainer::addTempFile(tmpfileindex);
std::string md5filename;
std::string indexfilename;
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > cbs;
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Pmd5cb;
if ( arginfo.getValue<unsigned int>("md5",getDefaultMD5()) )
{
if ( arginfo.hasArg("md5filename") && arginfo.getUnparsedValue("md5filename","") != "" )
md5filename = arginfo.getUnparsedValue("md5filename","");
else
std::cerr << "[V] no filename for md5 given, not creating hash" << std::endl;
if ( md5filename.size() )
{
::libmaus::lz::BgzfDeflateOutputCallbackMD5::unique_ptr_type Tmd5cb(new ::libmaus::lz::BgzfDeflateOutputCallbackMD5);
Pmd5cb = UNIQUE_PTR_MOVE(Tmd5cb);
cbs.push_back(Pmd5cb.get());
}
}
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Pindex;
if ( arginfo.getValue<unsigned int>("index",getDefaultIndex()) )
{
if ( arginfo.hasArg("indexfilename") && arginfo.getUnparsedValue("indexfilename","") != "" )
indexfilename = arginfo.getUnparsedValue("indexfilename","");
else
std::cerr << "[V] no filename for index given, not creating index" << std::endl;
if ( indexfilename.size() )
{
libmaus::bambam::BgzfDeflateOutputCallbackBamIndex::unique_ptr_type Tindex(new libmaus::bambam::BgzfDeflateOutputCallbackBamIndex(tmpfileindex));
Pindex = UNIQUE_PTR_MOVE(Tindex);
cbs.push_back(Pindex.get());
}
}
std::vector< ::libmaus::lz::BgzfDeflateOutputCallback * > * Pcbs = 0;
if ( cbs.size() )
Pcbs = &cbs;
/*
* end md5/index callbacks
*/
::libmaus::bambam::BamWriter::unique_ptr_type writer(new ::libmaus::bambam::BamWriter(std::cout,uphead,level,Pcbs));
libmaus::bambam::BamAlignment & algn = dec.getAlignment();
uint64_t c = 0;
while ( dec.readAlignment() )
{
bool const ok = split12(algn);
if ( ok )
algn.serialise(writer->getStream());
if ( verbose && (++c & (1024*1024-1)) == 0 )
std::cerr << "[V] " << c/(1024*1024) << std::endl;
}
writer.reset();
if ( Pmd5cb )
{
Pmd5cb->saveDigestAsFile(md5filename);
}
if ( Pindex )
{
Pindex->flush(std::string(indexfilename));
}
return EXIT_SUCCESS;
}
