Mercurial > repos > devteam > data_manager_twobit_builder

comparison data_manager/twobit_builder.py @ 0:74b09c8e5f6e draft

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-devteam/tree/master/data_managers/data_manager_twobit_builder commit 130cb0c08ad3c5b858ba46b1024dcdccc3cb68c6-dirty
author devteam
date Thu, 11 Jun 2015 16:14:27 -0400
parents
children 9946bc39c834
comparison
equal deleted inserted replaced
-1:000000000000 0:74b09c8e5f6e
1 #!/usr/bin/env python
2 #Dan Blankenberg
3
4 import sys, os, tempfile, optparse, uuid, subprocess
5
6 from json import loads, dumps
7
8
9 CHUNK_SIZE = 2**20 #1mb
10
11 def get_id_name( params, dbkey, fasta_description=None):
12 #TODO: ensure sequence_id is unique and does not already appear in location file
13 sequence_id = params['param_dict']['sequence_id']
14 if not sequence_id:
15 sequence_id = dbkey #uuid.uuid4() generate and use an uuid
16
17 sequence_name = params['param_dict']['sequence_name']
18 if not sequence_name:
19 sequence_name = fasta_description
20 if not sequence_name:
21 sequence_name = dbkey
22 return sequence_id, sequence_name
23
24 def build_twobit( data_manager_dict, fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name ):
25 twobit_base_name = "%s.2bit" % ( sequence_id )
26 twobit_filename = os.path.join( target_directory, twobit_base_name )
27
28 args = [ 'faToTwoBit', fasta_filename, twobit_filename ]
29 tmp_stderr = tempfile.NamedTemporaryFile( prefix = "tmp-data-manager-twobit-builder-stderr" )
30 proc = subprocess.Popen( args=args, shell=False, cwd=target_directory, stderr=tmp_stderr.fileno() )
31 return_code = proc.wait()
32 if return_code:
33 tmp_stderr.flush()
34 tmp_stderr.seek(0)
35 print >> sys.stderr, "Error building index:"
36 while True:
37 chunk = tmp_stderr.read( CHUNK_SIZE )
38 if not chunk:
39 break
40 sys.stderr.write( chunk )
41 sys.exit( return_code )
42 tmp_stderr.close()
43 #lastz_seqs
44 data_table_entry = dict( value=sequence_id, name=sequence_name, path=twobit_base_name )
45
46 _add_data_table_entry( data_manager_dict, "lastz_seqs", data_table_entry )
47 #twobit.loc
48 data_table_entry = dict( value=sequence_id, path=twobit_base_name )
49
50 _add_data_table_entry( data_manager_dict, "twobit", data_table_entry )
51 #alignseq
52 data_table_entry = dict( type="seq", value=sequence_id, path=twobit_base_name )
53
54 _add_data_table_entry( data_manager_dict, "alignseq_seq", data_table_entry )
55
56 def _add_data_table_entry( data_manager_dict, data_table_name, data_table_entry ):
57 data_manager_dict['data_tables'] = data_manager_dict.get( 'data_tables', {} )
58 data_manager_dict['data_tables'][ data_table_name ] = data_manager_dict['data_tables'].get( data_table_name, [] )
59 data_manager_dict['data_tables'][ data_table_name ].append( data_table_entry )
60 return data_manager_dict
61
62 def main():
63 #Parse Command Line
64 parser = optparse.OptionParser()
65 parser.add_option( '-f', '--fasta_filename', dest='fasta_filename', action='store', type="string", default=None, help='fasta_filename' )
66 parser.add_option( '-d', '--fasta_dbkey', dest='fasta_dbkey', action='store', type="string", default=None, help='fasta_dbkey' )
67 parser.add_option( '-t', '--fasta_description', dest='fasta_description', action='store', type="string", default=None, help='fasta_description' )
68 (options, args) = parser.parse_args()
69
70 filename = args[0]
71
72 params = loads( open( filename ).read() )
73
74 target_directory = params[ 'output_data' ][0]['extra_files_path']
75 os.mkdir( target_directory )
76 data_manager_dict = {}
77
78 dbkey = options.fasta_dbkey
79
80 if dbkey in [ None, '', '?' ]:
81 raise Exception( '"%s" is not a valid dbkey. You must specify a valid dbkey.' % ( dbkey ) )
82
83 sequence_id, sequence_name = get_id_name( params, dbkey=dbkey, fasta_description=options.fasta_description )
84
85 #build the index
86 build_twobit( data_manager_dict, options.fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name )
87
88 #save info to json file
89 open( filename, 'wb' ).write( dumps( data_manager_dict ) )
90
91 if __name__ == "__main__": main()
92