hisat-3n/hisat-3n-build

#!/usr/bin/env python

"""
 Copyright 2015, Daehwan Kim <infphilo@gmail.com>

 This file is part of HISAT 2.

 HISAT 2 is free software: you can redistribute it and/or modify
 it under the terms of the GNU General Public License as published by
 the Free Software Foundation, either version 3 of the License, or
 (at your option) any later version.

 HISAT 2 is distributed in the hope that it will be useful,
 but WITHOUT ANY WARRANTY; without even the implied warranty of
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 GNU General Public License for more details.

 You should have received a copy of the GNU General Public License
 along with HISAT 2.  If not, see <http://www.gnu.org/licenses/>.
"""


import os
import sys
import inspect
import logging
import re



def build_args():
    """
    Parse the wrapper arguments. Returns the options,<programm arguments> tuple.
    """

    parsed_args = {}
    to_remove = []
    argv = sys.argv[:]
    for i, arg in enumerate(argv):
        if arg == '--large-index':
            parsed_args[arg] = ""
            to_remove.append(i)
        elif arg == '--debug':
            parsed_args[arg] = ""
            to_remove.append(i)
        elif arg == '--verbose':
            parsed_args[arg] = ""
            to_remove.append(i)

    for i in reversed(to_remove):
        del argv[i]

    return parsed_args, argv


def main():
    logging.basicConfig(level=logging.ERROR,
                        format='%(levelname)s: %(message)s'
                        )
    delta               = 200
    small_index_max_size= 4 * 1024**3 - delta
    build_bin_name      = "hisat2-build"
    build_bin_s         = "hisat2-build-s"
    build_bin_l         = "hisat2-build-l"
    curr_script         = os.path.realpath(inspect.getsourcefile(main))
    ex_path             = os.path.dirname(curr_script)
    build_bin_spec      = os.path.join(ex_path,build_bin_s)
    #repeat_bin_spec     = os.path.join(ex_path,"hisat2-repeat")

    options, argv = build_args()

    if '--verbose' in options:
        logging.getLogger().setLevel(logging.INFO)
        
    if '--debug' in options:
        build_bin_spec += '-debug'
        build_bin_l += '-debug'

    if '--large-index' in options:
        build_bin_spec = os.path.join(ex_path,build_bin_l)
    elif len(argv) >= 2:
        ref_fnames = argv[-2]
        tot_size = 0
        for fn in ref_fnames.split(','):
            if os.path.exists(fn):
                statinfo = os.stat(fn)
                tot_size += statinfo.st_size
        if tot_size > small_index_max_size:
            build_bin_spec = os.path.join(ex_path, build_bin_l)


    if "--repeat-index" in argv:
        #build repeat index first
        if argv[1].startswith('-'):
            outputName = argv[-1]
        else:
            outputName = argv[2]
        nThread = "1"
        repeatLength = "100-300"
        repeatCount = "5"
        referenceName = ""
        base_change = ""
        removeRepeatLength = False
        for i in range(len(argv)):
            if argv[i] == "--base-change":
                base_change = argv[i+1]
            if argv[i] == "-p":
                nThread = argv[i+1]
            elif argv[i] == "-f":
                referenceName = argv[i+1]
            elif argv[i] == "--repeat-index":
                if ("--repeat-index" != argv[-1]):
                    match = re.match(r'\d+-\d+', argv[i+1])
                    if match:
                        repeatLength = argv[i+1]
                        removeRepeatLength = True
        if removeRepeatLength:
            argv.remove(repeatLength)
        if len(referenceName) == 0:
            if argv[1].startswith('-'):
                referenceName = argv[-2]
            else:
                referenceName = argv[1]
        cwd = sys.path[0]

        repeatArgv = []
        repeatArgv += [cwd + "/hisat2-repeat"]
        repeatArgv += ["-p", nThread]
        repeatArgv += ["--repeat-length", repeatLength]
        repeatArgv += ["--repeat-count", repeatCount]
        repeatArgv += ["--base-change", base_change]
        repeatArgv += ["--3N"]
        repeatArgv += [referenceName, outputName]
        try:
            os.system(" ".join(repeatArgv))
        except ValueError:
            print("Can not automatically generate repeat database for HISAT-3N. Please manually generate repeat database by using hisat2-repeat.")

    argv.append("--3N")

    argv[0] = build_bin_name
    argv.insert(1, 'basic-0')
    argv.insert(1, '--wrapper')
    logging.info('Command: %s %s' % (build_bin_spec, ' '.join(argv[1:])))
    os.execv(build_bin_spec, argv)

if __name__ == "__main__":
    main()