comparison vsnp_get_snps.py @ 11:6b3b0f5858e6 draft default tip

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/vsnp commit c38fd63f7980c70390d104a73ba4c72b266444c3

10:152716f90b84

import vcf


def get_time_stamp():
    return datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H-%M-%S')


def set_num_cpus(num_files, processes):
    num_cpus = len(os.sched_getaffinity(0))
    if num_files < num_cpus and num_files < processes:
        return num_files
    if num_cpus < processes:
        half_cpus = int(num_cpus / 2)
        if num_files < half_cpus:
            return num_files
        return half_cpus
    return processes


def setup_all_vcfs(vcf_files, vcf_dirs):
    # Create the all_vcfs directory and link
    # all input vcf files into it for processing.

        vcf_files.append(file_path)

    multiprocessing.set_start_method('spawn')
    queue1 = multiprocessing.JoinableQueue()
    num_files = len(vcf_files)
    cpus = set_num_cpus(num_files, args.processes)
    # Set a timeout for get()s in the queue.
    timeout = 0.05

    # Initialize the snp_finder object.
    snp_finder = SnpFinder(num_files, args.dbkey, args.input_excel, args.all_isolates, args.ac, args.min_mq, args.quality_score_n_threshold, args.min_quality_score, args.input_vcf_dir, args.output_json_avg_mq_dir, args.output_json_snps_dir, args.output_snps_dir, args.output_summary)

    # Populate the queue for job splitting.
    for vcf_dir in vcf_dirs:
        queue1.put(vcf_dir)

    # Complete the get_snps task.
    processes = [multiprocessing.Process(target=snp_finder.get_snps, args=(queue1, timeout, )) for _ in range(cpus)]
    for p in processes:
        p.start()
    for p in processes:
        p.join()
    queue1.join()

11:6b3b0f5858e6

import vcf


def get_time_stamp():
    return datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H-%M-%S')


def setup_all_vcfs(vcf_files, vcf_dirs):
    # Create the all_vcfs directory and link
    # all input vcf files into it for processing.

        vcf_files.append(file_path)

    multiprocessing.set_start_method('spawn')
    queue1 = multiprocessing.JoinableQueue()
    num_files = len(vcf_files)
    # Set a timeout for get()s in the queue.
    timeout = 0.05

    # Initialize the snp_finder object.
    snp_finder = SnpFinder(num_files, args.dbkey, args.input_excel, args.all_isolates, args.ac, args.min_mq, args.quality_score_n_threshold, args.min_quality_score, args.input_vcf_dir, args.output_json_avg_mq_dir, args.output_json_snps_dir, args.output_snps_dir, args.output_summary)

    # Populate the queue for job splitting.
    for vcf_dir in vcf_dirs:
        queue1.put(vcf_dir)

    # Complete the get_snps task.
    processes = [multiprocessing.Process(target=snp_finder.get_snps, args=(queue1, timeout, )) for _ in range(args.processes)]
    for p in processes:
        p.start()
    for p in processes:
        p.join()
    queue1.join()