build.rna_tools.tools.mq.rna_mq_collect.RunAllDirectory.run()

Complexity

Conditions

24

Size

Total Lines	126
Code Lines	79

Duplication

Lines	15
Ratio	11.9 %

Importance

Changes

0

#!/usr/bin/env python
    def run(self, filenames, csv_path, opt):
        """Open csv (with appropriate headers), run methods, print & save csv

        There are two modes of execution:
         * multiprocessing
         * single

        .. warning:: Works on global variables: ref_seq, ref_ss, methods, lock, c
        """
        global ref_seq, ref_ss, verbose, methods, lock, c

        if opt.seq_ss_filename:

This code seems to be duplicated in your project.

            pdb_id, ref_seq, ref_ss = [x.strip() for x in open(opt.seq_ss_filename).read().strip().split('\n')]
            #sg.phr_text('FASTA SEQ/SS')
            sg.poptions({'AnalyzeGeometry': True, 'SSAgreement' : True})
            sg.poption('pdb_id', pdb_id)
            sg.poption('ref_seq', ref_seq)
            sg.poption('ref_ss', ref_ss)
        else:
            pdb_id, ref_seq, ref_ss = ['', '', '']
            sg.poptions({'SSAgreement' : True})
            # hack
            try: # if it's not on the list
                methods.remove('SSAgreement')
            except ValueError:
                pass

        verbose = opt.verbose

        global csv_file, csv_writer  # hack
        # csv open & add header
        csv_file = open(csv_path, 'a')
        csv_writer = csv.writer(csv_file,  delimiter=',')
        # make header
        headers = ['id', 'fn']
        for m in methods:
            headers += attributes[m]

        if opt.native_pdb_filename:
            headers += ['RMSDALL']
        if opt.mqapscore:
            headers += ['SCORE']
        csv_writer.writerow(headers)
        csv_file.flush()

        # remove ~ and remove .out
        for f in copy.copy(filenames):
            if f.endswith('~'):
                filenames.remove(f)
            if f.endswith('.out'):
                filenames.remove(f)
            if f.find('._')>-1:
                filenames.remove(f)

        files_to_ignore = []
        # or if not provided
        import glob

        if not opt.force:
            opt.ignore_pdb_filename = glob.glob('*' + opt.methods + '*.csv')
            for f in opt.ignore_pdb_filename:  # do it for the list, that's nice!
                fn = open(f)
                for f in fn.read().strip().split('\n'):
                    if 'error' in f:
                        continue  # don't add files with errors, so the program will be re-run for them
                    # if there is an error, this will give error again quickly
                    # but this solves when you kill the job, you get erros, but it's not rally errors
                    # but stopped jobs
                    if f.find('\t') > -1:
                        f = f.split('\t')[1] # id, fn
                    if f.find(',') > -1:
                        f = f.split(',')[1] # id, fn
                    files_to_ignore.append(os.path.basename(f))

        ## files to ignore
        print(' to ignore', len(files_to_ignore), files_to_ignore[:4])

        filenames = []
        for i, f in enumerate(input_files):

The variable input_files does not seem to be defined in case __name__ == '__main__' on line 557 is False. Are you sure this can never be the case?

            # print(i, f)
            if f.startswith('_'):  # skip
                continue
            if os.path.basename(f) not in files_to_ignore:
                filenames.append(f)
        ## for fi in files_to_ignore:
        ##     for fn in copy.copy(filenames):
        ##         if os.path.basename(fn).startswith('._'):
        ##             filenames.remove(fn)
        ##         if os.path.basename(fn).startswith(fi.split('\t')[0]): # # hack,  @todo <- re could be used here!  to ignore ['fn,RASP,SimRNA,FARNA,NAST_pyro\r', '1ykv_1_ba_c.pdb,-0.104705,-504.468933,-306.245,122.7\r', '2esj_1_ba_c.pdb,-0.1522,-1,-266.217,46.7\r', '2quw_1_ba_c.pdb,-0.103789,-729.386726,-419.047,984.0\r
        ##             filenames.remove(fn)
        print(' files to analyze: %s' % len(filenames), filenames[:300])
        ## headers
        methods_to_print = copy.copy(methods)
        if opt.native_pdb_filename:
            methods_to_print += ['RMSDALL']
        if opt.mqapscore:
            methods_to_print += ['SCORE']

        ## if verbose: print ''.ljust(80), ''.join([m[:9].ljust(10) for m in methods_to_print]) ## print headers

        sg.phr()

        lock = Lock()
        
        counter.value = len(files_to_ignore)

        flist = []
        c  = 1
        # two running modes
        global filename_length
        filenames_length = len(filenames) + len(files_to_ignore)

        global bar
        bar = progressbar.ProgressBar(max_value=filenames_length)
        bar.update(len(files_to_ignore))

        fl = []
        for f in filenames:
            fl.append([f,filenames_length])

        if opt.number_processes:
            p = Pool(opt.number_processes)
            p.map(single_run, fl)
        else:
            for filename,x in fl:
                single_run((filename,x))