# Set up and run pipeline
pipeline = Pipeline.PipelineRunner(runner,max_concurrent_jobs=options.max_concurrent_jobs,
jobCompletionHandler=JobCleanup,
groupCompletionHandler=lambda group,jobs,
email=options.email_addr: SendReport(email,group,jobs))
for data_dir in data_dirs:
# Get for this directory
print "Collecting data from %s" % data_dir
if options.input_type == "solid":
run_data = Pipeline.GetSolidDataFiles(data_dir,pattern=options.pattern)
elif options.input_type == "solid_paired_end":
run_data = Pipeline.GetSolidPairedEndFiles(data_dir,pattern=options.pattern)
elif options.input_type == "fastq":
run_data = Pipeline.GetFastqFiles(data_dir,pattern=options.pattern)
elif options.input_type == "fastqgz":
run_data = Pipeline.GetFastqGzFiles(data_dir,pattern=options.pattern)
else:
logging.error("Unknown input type: '%s'" % options.input_type)
sys.exit(1)
# Add jobs to pipeline runner (up to limit of max_total_jobs)
for data in run_data:
if options.max_total_jobs > 0 and pipeline.nWaiting() == options.max_total_jobs:
print "Maximum number of jobs queued (%d)" % options.max_total_jobs
break
label = os.path.splitext(os.path.basename(data[0]))[0]
group = os.path.basename(data_dir)
# Set up argument list for script
args = []
if script_args:
for arg in script_args:
args.append(arg)
