static std::pair<int64_t,uint64_t> countBinChunks(stream_type & stream)
{
int64_t bin = -1;
int64_t refid = -1;
uint64_t chunks = 0;
::libmaus2::bambam::BamIndexBinChunk BC;
while ( stream.peek() != stream_type::traits_type::eof() )
{
stream.read(reinterpret_cast<char *>(&BC),sizeof(::libmaus2::bambam::BamIndexBinChunk));
if ( refid < 0 )
refid = BC.refid;
if ( bin < 0 )
bin = BC.bin;
if (
refid != static_cast<int64_t>(BC.refid)
||
bin != static_cast<int64_t>(BC.bin)
)
{
stream.clear();
stream.seekg(-static_cast<int64_t>(sizeof(::libmaus2::bambam::BamIndexBinChunk)),std::ios::cur);
stream.clear();
break;
}
chunks++;
}
return std::pair<uint64_t,uint64_t>(bin,chunks);
}
static std::pair<int64_t,int64_t> getLinearMaxChunk(stream_type & stream)
{
int64_t refid = -1;
int64_t maxpos = 0;
::libmaus2::bambam::BamIndexLinearChunk LC;
while ( stream.peek() != stream_type::traits_type::eof() )
{
stream.read(reinterpret_cast<char *>(&LC),sizeof(::libmaus2::bambam::BamIndexLinearChunk));
if ( refid == -1 )
refid = LC.refid;
// put back element, if it has a different refid
if ( static_cast<int64_t>(LC.refid) != refid )
{
stream.clear();
stream.seekg(-static_cast<int64_t>(sizeof(::libmaus2::bambam::BamIndexLinearChunk)),std::ios::cur);
break;
}
maxpos = std::max(maxpos,LC.chunkid);
}
return std::pair<int64_t,int64_t>(refid,maxpos);
}
static std::pair<int64_t,uint64_t> countLinearChunks(stream_type & stream)
{
int64_t refid = -1;
uint64_t cnt = 0;
::libmaus2::bambam::BamIndexLinearChunk LC;
while ( stream.peek() != stream_type::traits_type::eof() )
{
stream.read(reinterpret_cast<char *>(&LC),sizeof(::libmaus2::bambam::BamIndexLinearChunk));
if ( refid == -1 )
refid = LC.refid;
// put back element, if it has a different refid
if ( static_cast<int64_t>(LC.refid) != refid )
{
stream.clear();
stream.seekg(-static_cast<int64_t>(sizeof(::libmaus2::bambam::BamIndexLinearChunk)),std::ios::cur);
break;
}
cnt++;
}
return std::pair<int64_t,uint64_t>(refid,cnt);
}
static bool peekLinearChunk(stream_type & stream, uint64_t const refid, uint64_t const pos, unsigned int const posshift)
{
::libmaus2::bambam::BamIndexLinearChunk LC;
if ( stream.peek() == stream_type::traits_type::eof() )
return false;
stream.read(reinterpret_cast<char *>(&LC),sizeof(::libmaus2::bambam::BamIndexLinearChunk));
stream.clear();
stream.seekg(-static_cast<int64_t>(sizeof(::libmaus2::bambam::BamIndexLinearChunk)),std::ios::cur);
return (LC.refid == refid) && ((LC.pos >> posshift)==(pos>>posshift));
}
static bool peekLinearChunk(stream_type & stream, uint64_t const refid, int64_t const chunkid)
{
::libmaus2::bambam::BamIndexLinearChunk LC;
if ( stream.peek() == stream_type::traits_type::eof() )
return false;
stream.read(reinterpret_cast<char *>(&LC),sizeof(::libmaus2::bambam::BamIndexLinearChunk));
stream.clear();
stream.seekg(-static_cast<int64_t>(sizeof(::libmaus2::bambam::BamIndexLinearChunk)),std::ios::cur);
return LC.refid == refid && LC.chunkid == chunkid;
}
static bool readAlignmentGz(
stream_type & GZ,
::libmaus::bambam::BamAlignment & alignment,
::libmaus::bambam::BamHeader const * bamheader = 0,
bool const validate = true
)
{
/* read alignment block size */
int64_t const bs0 = GZ.get();
int64_t const bs1 = GZ.get();
int64_t const bs2 = GZ.get();
int64_t const bs3 = GZ.get();
if ( bs3 < 0 )
// reached end of file
return false;
/* assemble block size as LE integer */
alignment.blocksize = (bs0 << 0) | (bs1 << 8) | (bs2 << 16) | (bs3 << 24) ;
/* read alignment block */
if ( alignment.blocksize > alignment.D.size() )
alignment.D = ::libmaus::bambam::BamAlignment::D_array_type(alignment.blocksize,false);
GZ.read(reinterpret_cast<char *>(alignment.D.begin()),alignment.blocksize);
if ( static_cast<int64_t>(GZ.gcount()) != static_cast<int64_t>(alignment.blocksize) )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Invalid alignment (EOF in alignment block of length " << alignment.blocksize << ")" << std::endl;
se.finish();
throw se;
}
if ( validate )
{
libmaus_bambam_alignment_validity const validity = bamheader ? alignment.valid(*bamheader) : alignment.valid();
if ( validity != ::libmaus::bambam::libmaus_bambam_alignment_validity_ok )
{
::libmaus::exception::LibMausException se;
se.getStream() << "Invalid alignment: " << validity << std::endl;
se.finish();
throw se;
}
}
return true;
}
static int64_t peekBin(stream_type & stream)
{
::libmaus2::bambam::BamIndexBinChunk BC;
if ( stream.peek() == stream_type::traits_type::eof() )
return -1;
stream.read(
reinterpret_cast<char *>(&BC),
sizeof(::libmaus2::bambam::BamIndexBinChunk)
);
assert ( stream.gcount() == sizeof(::libmaus2::bambam::BamIndexBinChunk) );
stream.clear();
stream.seekg(-static_cast<int64_t>(sizeof(::libmaus2::bambam::BamIndexBinChunk)),std::ios::cur);
return BC.refid;
}
void init(stream_type & stream)
{
char magic[4];
stream.read(&magic[0],sizeof(magic));
if (
! stream
||
stream.gcount() != 4
||
magic[0] != 'B'
||
magic[1] != 'A'
||
magic[2] != 'I'
||
magic[3] != '\1'
)
{
libmaus2::exception::LibMausException ex;
ex.getStream() << "Failed to read BAI magic BAI\\1." << std::endl;
ex.finish();
throw ex;
}
uint32_t const numref = getLEInteger<stream_type,uint32_t,4>(stream);
refs = libmaus2::autoarray::AutoArray<libmaus2::bambam::BamIndexRef>(numref);
for ( uint64_t i = 0; i < numref; ++i )
{
uint32_t const distbins = getLEInteger<stream_type,uint32_t,4>(stream);
#if 0
std::cerr << "chr " << i << " distbins " << distbins << std::endl;
#endif
if ( distbins )
{
refs[i].bin = libmaus2::autoarray::AutoArray<libmaus2::bambam::BamIndexBin>(distbins,false);
libmaus2::autoarray::AutoArray< std::pair<uint64_t,uint64_t> > pi(distbins,false);
libmaus2::autoarray::AutoArray<libmaus2::bambam::BamIndexBin> prebins(distbins,false);
for ( uint64_t j = 0; j < distbins; ++j )
{
uint32_t const bin = getLEInteger<stream_type,uint32_t,4>(stream);
uint32_t const chunks = getLEInteger<stream_type,uint32_t,4>(stream);
// std::cerr << "chr " << i << " bin " << bin << " chunks " << chunks << std::endl;
prebins[j].bin = bin;
prebins[j].chunks = libmaus2::autoarray::AutoArray<libmaus2::bambam::BamIndexBin::Chunk>(chunks,false);
// read chunks
for ( uint64_t k = 0; k < chunks; ++k )
{
prebins[j].chunks[k].first = getLEInteger<stream_type,uint64_t,8>(stream);
prebins[j].chunks[k].second = getLEInteger<stream_type,uint64_t,8>(stream);
}
pi [ j ] = std::pair<uint64_t,uint64_t>(bin,j);
}
// sort by bin
std::sort(pi.begin(),pi.end());
// move
for ( uint64_t j = 0; j < distbins; ++j )
refs[i].bin[j] = prebins[pi[j].second];
}
uint32_t const lins = getLEInteger<stream_type,uint32_t,4>(stream);
if ( lins )
{
refs[i].lin.intervals = libmaus2::autoarray::AutoArray<uint64_t>(lins,false);
for ( uint64_t j = 0; j < lins; ++j )
refs[i].lin.intervals[j] = getLEInteger<stream_type,uint64_t,8>(stream);
}
}
}