int main (int argc, char * argv[])
{
vector<string> inputFilenames;
string combinedOutFilename, alignmentsOutFilename;
try
{
TCLAP::CmdLine cmd("Program description", ' ', VERSION);
TCLAP::ValueArg<string> combinedOutputArg("o", "out",
"Combined output filename (BAM format)", true, "", "combined.bam", cmd);
TCLAP::ValueArg<int> minInsertArg("n", "min-insert",
"Minimum insert size", false, DEFAULT_MIN_GAP, "min insert size", cmd);
TCLAP::ValueArg<int> maxInsertArg("x", "max-insert",
"Maximum insert size", false, DEFAULT_MAX_GAP, "max insert size", cmd);
TCLAP::MultiArg<string> inputArgs("b", "bam",
"Input BAM file", true,
"input.bam", cmd);
cmd.parse(argc, argv);
combinedOutFilename = combinedOutputArg.getValue();
MIN_GAP = minInsertArg.getValue();
MAX_GAP = maxInsertArg.getValue();
inputFilenames = inputArgs.getValue();
} catch (TCLAP::ArgException &e) {
cerr << "Error: " << e.error() << " " << e.argId() << endl;
}
// TODO require that alignments are sorted by name
BamMultiReader reader;
reader.Open(inputFilenames);
if (!ValidOut.Open(combinedOutFilename, reader.GetHeader(),
reader.GetReferenceData()))
{
cerr << ValidOut.GetErrorString() << endl;
return 1;
}
string current, prev;
char mateID;
Group group;
set<string> references;
Alignment a;
while (reader.GetNextAlignment(a))
{
parseID(a.Name, current, mateID);
if (current.compare(prev) && prev.size() > 0)
{
processGroup(group, references);
group.clear();
references.clear();
}
references.insert(a.RefName);
GroupKey key;
key.refID = a.RefName;
key.mateID = mateID;
key.rev = a.IsReverseStrand();
group.insert( std::make_pair( key, a ) );
prev = current;
}
processGroup(group, references);
}
int FileReader::runInternal()
{
ogeNameThread("am_FileReader");
if(!format_specified)
format = deduceFileFormat();
if(format == FORMAT_BAM)
{
BamMultiReader reader;
if(!reader.Open(filenames)) {
cerr << "Error opening BAM files." << endl;
reader.Close();
return -1;
}
header = reader.GetHeader();
references = reader.GetReferenceData();
open = true;
BamAlignment * al;
while(true)
{
if(load_string_data)
al = reader.GetNextAlignment();
else
al = reader.GetNextAlignmentCore();
if(!al)
break;
putOutputAlignment(al);
}
reader.Close();
} else if(format == FORMAT_SAM) {
vector<SamReader> readers;
SamHeader first_header;
// before doing any reading, open the files to
// verify they are the right format, etc.
for(int i = 0; i < filenames.size(); i++) {
SamReader reader;
if(!reader.Open(filenames[i])) {
cerr << "Error opening SAM file: " << filenames[i] << endl;
return -1;
}
if(filenames.size() > 1 && i == 0)
first_header = header;
// TODO: We can probably find a better way to deal with multiple SAM file headers,
// but for now we should disallow different headers to avoid issues.
if(i > 0 && header.ToString() != first_header.ToString())
cerr << "Warning! SAM input files have different headers." << endl;
reader.Close();
}
for(int i = 0; i < filenames.size(); i++) {
SamReader reader;
if(!reader.Open(filenames[i])) {
cerr << "Error opening SAM file: " << filenames[i] << endl;
return -1;
}
header = reader.GetHeader();
references = reader.GetReferenceData();
open = true;
if(filenames.size() > 1 && i == 0)
first_header = header;
BamAlignment * al = NULL;
while(true)
{
al = reader.GetNextAlignment();
if(NULL == al)
break;
putOutputAlignment(al);
}
reader.Close();
}
} else {
cerr << "FileReader couldn't detect file format. Aborting." << endl;
exit(-1);
return -1;
}
return 0;
}