Mercurial > repos > devteam > data_manager_twobit_builder

comparison data_manager/twobit_builder.py @ 7:94a50db37d1a draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/data_managers/data_manager_twobit_builder commit 096286097ed5cdf189a1b68c3fc34d10f4142e54
author iuc
date Sun, 16 Apr 2023 08:31:43 +0000
parents 62359a151521
children
comparison
equal deleted inserted replaced
6:62359a151521 7:94a50db37d1a
24 if not sequence_name: 24 if not sequence_name:
25 sequence_name = dbkey 25 sequence_name = dbkey
26 return sequence_id, sequence_name 26 return sequence_id, sequence_name
27 27
28 28
29 def build_twobit(data_manager_dict, fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name, long): 29 def build_twobit(data_manager_dict, fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name):
30 twobit_base_name = "%s.2bit" % (sequence_id) 30 twobit_base_name = "%s.2bit" % (sequence_id)
31 twobit_filename = os.path.join(target_directory, twobit_base_name) 31 twobit_filename = os.path.join(target_directory, twobit_base_name)
32 32
33 args = ['faToTwoBit'] 33 args = ['faToTwoBit', fasta_filename, twobit_filename]
34 if long:
35 args.append('-long')
36 args.extend([fasta_filename, twobit_filename])
37 tmp_stderr = tempfile.NamedTemporaryFile(prefix="tmp-data-manager-twobit-builder-stderr") 34 tmp_stderr = tempfile.NamedTemporaryFile(prefix="tmp-data-manager-twobit-builder-stderr")
38 proc = subprocess.Popen(args=args, shell=False, cwd=target_directory, stderr=tmp_stderr.fileno()) 35 proc = subprocess.Popen(args=args, shell=False, cwd=target_directory, stderr=tmp_stderr.fileno())
39 return_code = proc.wait() 36 return_code = proc.wait()
40 if return_code: 37 if return_code:
41 tmp_stderr.flush() 38 tmp_stderr.flush()
43 print("Error building index:", file=sys.stderr) 40 print("Error building index:", file=sys.stderr)
44 while True: 41 while True:
45 chunk = tmp_stderr.read(CHUNK_SIZE) 42 chunk = tmp_stderr.read(CHUNK_SIZE)
46 if not chunk: 43 if not chunk:
47 break 44 break
48 sys.stderr.write(chunk.decode('utf-8')) 45 sys.stderr.write(chunk)
49 sys.exit(return_code) 46 sys.exit(return_code)
50 tmp_stderr.close() 47 tmp_stderr.close()
51 # lastz_seqs 48 # lastz_seqs
52 data_table_entry = dict(value=sequence_id, name=sequence_name, path=twobit_base_name) 49 data_table_entry = dict(value=sequence_id, name=sequence_name, path=twobit_base_name)
53 50
72 def main(): 69 def main():
73 parser = optparse.OptionParser() 70 parser = optparse.OptionParser()
74 parser.add_option('-f', '--fasta_filename', dest='fasta_filename', action='store', type="string", default=None, help='fasta_filename') 71 parser.add_option('-f', '--fasta_filename', dest='fasta_filename', action='store', type="string", default=None, help='fasta_filename')
75 parser.add_option('-d', '--fasta_dbkey', dest='fasta_dbkey', action='store', type="string", default=None, help='fasta_dbkey') 72 parser.add_option('-d', '--fasta_dbkey', dest='fasta_dbkey', action='store', type="string", default=None, help='fasta_dbkey')
76 parser.add_option('-t', '--fasta_description', dest='fasta_description', action='store', type="string", default=None, help='fasta_description') 73 parser.add_option('-t', '--fasta_description', dest='fasta_description', action='store', type="string", default=None, help='fasta_description')
77 parser.add_option('-l', '--long', dest='long', action="store_true", default=False, help='For big genomes you need to pass the -long option.')
78
79 (options, args) = parser.parse_args() 74 (options, args) = parser.parse_args()
80 75
81 filename = args[0] 76 filename = args[0]
82 77
83 with open(filename) as fh: 78 with open(filename) as fh:
93 raise Exception('"%s" is not a valid dbkey. You must specify a valid dbkey.' % (dbkey)) 88 raise Exception('"%s" is not a valid dbkey. You must specify a valid dbkey.' % (dbkey))
94 89
95 sequence_id, sequence_name = get_id_name(params, dbkey=dbkey, fasta_description=options.fasta_description) 90 sequence_id, sequence_name = get_id_name(params, dbkey=dbkey, fasta_description=options.fasta_description)
96 91
97 # build the index 92 # build the index
98 build_twobit(data_manager_dict, options.fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name, options.long) 93 build_twobit(data_manager_dict, options.fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name)
99 94
100 # save info to json file 95 # save info to json file
101 with open(filename, 'w') as fh: 96 with open(filename, 'w') as fh:
102 json.dump(data_manager_dict, fh, sort_keys=True) 97 json.dump(data_manager_dict, fh, sort_keys=True)
103 98