mergeLHE Namespace Reference

Classes
class	BaseLHEMerger
class	DefaultLHEMerger
class	ExternalCppLHEMerger
class	MG5LOLHEMerger

Functions
def	main

Function Documentation

def mergeLHE.main

(

argv = None

)

Main routine of the script.

Arguments:
- `argv`: arguments passed to the main routine

Definition at line 325 of file mergeLHE.py.

References submitPVResolutionJobs.count, and join().

def main(argv = None):
    """Main routine of the script.

    Arguments:
    - `argv`: arguments passed to the main routine
    """

    if argv == None:
        argv = sys.argv[1:]

    parser = argparse.ArgumentParser(
        description=("A universal script that merges multiple LHE files for all possible conditions and in the most "
                    "natural way.\n"
                    "A detailed description of the merging step (in the default mode):\n"
                    "  1. Header:\n"
                    "    a. assert consistency of the headers (allow difference for the info of e.g. #event, seed);\n"
                    "    b. if not MG LO LHEs, will simply use the header from the first LHE; otherwise, reset the "
                    "<MGGenerationInfo> from the headers by merging the #event & xsec info;\n"
                    "  2. Init block: if all <init> blocks are the same, use the same as output; otherwise (the MG LO "
                    "case), merge them by recalculating the # of subprocess (LRPUP) and XSECUP, XERRUP, XMAXUP per "
                    "each subprocess.\n"
                    "  3. Event block: concatenate all event blocks. If for MG LO LHEs, recalculate the per-event "
                    "XWGTUP and all <wgt> tags based on the new XSECUP, #event, and 'event_norm' read from the MG "
                    "run card.\n"
                    "For further development of this script please always validate the merging result on the test "
                    "routines: https://github.com/colizz/mergelhe_validate\n"
                    "Example usage:\n"
                    "    mergeLHE.py -i 'thread*/*.lhe,another_file/another.lhe' -o output.lhe"),
        formatter_class=argparse.RawTextHelpFormatter)
    parser.add_argument("-i", "--input-files", type=str,
                        help="Input LHE file paths separated by commas. Shell-type wildcards are supported.")
    parser.add_argument("-o", "--output-file",
                        default='output.lhe', type=str,
                        help="Output LHE file path.")
    parser.add_argument("--force-mglo-merger", action='store_true',
                        help=("Force to use the merger script dedicated for MG5 LO LHEs, as introduced in "
                             "https://github.com/cms-sw/genproductions/blob/master/bin/MadGraph5_aMCatNLO/Utilities/merge.pl"))
    parser.add_argument("--force-cpp-merger", action='store_true',
                        help=("Force to use the external mergeLheFiles.cpp file to merge LHE files, as introduced in "
                             "https://twiki.cern.ch/twiki/bin/view/CMSPublic/SWGuideSubgroupMC#1_2_Using_pLHE_campaigns"))
    parser.add_argument("-b", "--bypass-check", action='store_true',
                        help=("Bypass the compatibility check for the headers. If true, the header and init block "
                             "will be just a duplicate from the first input file, and events are concatenated without "
                             "modification."))
    parser.add_argument("--debug", action='store_true',
                        help="Use the debug mode.")
    args = parser.parse_args(argv)

    logging.basicConfig(
        format='[%(levelname)s] %(message)s',
        level=logging.INFO if not args.debug else DEBUG)
    logging.info('>>> launch mergeLHE.py in %s' % os.path.abspath(os.getcwd()))

    # Extract input LHE files from the path
    assert len(args.input_files), \
        ('Please specify your input LHE files by -i/--input-files. '
        'Run \'mergeLHE.py -h\' for details.')
    input_files = [] # each individual input file
    for path in args.input_files.split(','):
        find_files = glob.glob(path)
        if len(find_files) == 0:
            logging.info('Warning: cannot find files in %s' % path)
        input_files += find_files
    input_files.sort()
    logging.info('>>> Merge %d files: [%s]' % (len(input_files), ', '.join(input_files)))
    logging.info('>>> Write to output: %s ' % args.output_file)

    if not os.path.exists(os.path.dirname(os.path.realpath(args.output_file))):
        os.makedirs(os.path.dirname(os.path.realpath(args.output_file)))

    # Check arguments
    assert len(input_files) > 0, 'Input LHE files should be more than 0.'
    if len(input_files) == 1:
        logging.warning('Input LHE only has 1 file. Will copy this file to the destination.')
        import shutil
        shutil.copy(input_files[0], args.output_file)
        return
    assert [args.force_mglo_merger, args.force_cpp_merger].count(True) <= 1, \
        "Can only specify at most one from --force-mglo-merger or --force-cpp-merger."

    # Determine the merging scheme
    if args.force_mglo_merger:
        lhe_merger = MG5LOLHEMerger(input_files, args.output_file)
    elif args.force_cpp_merger:
        lhe_merger = ExternalCppLHEMerger(input_files, args.output_file)
    else:
        lhe_merger = DefaultLHEMerger(input_files, args.output_file, bypass_check=args.bypass_check)

    # Do merging
    lhe_merger.merge()