Mercurial > repos > devteam > data_manager_twobit_builder
changeset 7:94bffa4fdedc draft
planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/data_managers/data_manager_twobit_builder commit 18d696a933873372ae5c77257fd31eaf2cd4ed13
| author | iuc |
|---|---|
| date | Sun, 16 Apr 2023 08:15:04 +0000 |
| parents | b640d79b49cb |
| children | 4a7740ee2892 |
| files | data_manager/twobit_builder.py data_manager/twobit_builder.xml data_manager_conf.xml |
| diffstat | 3 files changed, 18 insertions(+), 9 deletions(-) [+] |
line wrap: on
line diff
--- a/data_manager/twobit_builder.py Thu Nov 28 20:33:03 2019 +0000 +++ b/data_manager/twobit_builder.py Sun Apr 16 08:15:04 2023 +0000 @@ -2,12 +2,12 @@ # Dan Blankenberg from __future__ import print_function +import json import optparse import os import subprocess import sys import tempfile -from json import dumps, loads CHUNK_SIZE = 2**20 # 1mb @@ -26,11 +26,14 @@ return sequence_id, sequence_name -def build_twobit(data_manager_dict, fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name): +def build_twobit(data_manager_dict, fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name, long): twobit_base_name = "%s.2bit" % (sequence_id) twobit_filename = os.path.join(target_directory, twobit_base_name) - args = ['faToTwoBit', fasta_filename, twobit_filename] + args = ['faToTwoBit'] + if long: + args.append('-long') + args.extend([fasta_filename, twobit_filename]) tmp_stderr = tempfile.NamedTemporaryFile(prefix="tmp-data-manager-twobit-builder-stderr") proc = subprocess.Popen(args=args, shell=False, cwd=target_directory, stderr=tmp_stderr.fileno()) return_code = proc.wait() @@ -42,7 +45,7 @@ chunk = tmp_stderr.read(CHUNK_SIZE) if not chunk: break - sys.stderr.write(chunk) + sys.stderr.write(chunk.decode('utf-8')) sys.exit(return_code) tmp_stderr.close() # lastz_seqs @@ -71,11 +74,14 @@ parser.add_option('-f', '--fasta_filename', dest='fasta_filename', action='store', type="string", default=None, help='fasta_filename') parser.add_option('-d', '--fasta_dbkey', dest='fasta_dbkey', action='store', type="string", default=None, help='fasta_dbkey') parser.add_option('-t', '--fasta_description', dest='fasta_description', action='store', type="string", default=None, help='fasta_description') + parser.add_option('-l', '--long', dest='long', action="store_true", default=False, help='For big genomes you need to pass the -long option.') + (options, args) = parser.parse_args() filename = args[0] - params = loads(open(filename).read()) + with open(filename) as fh: + params = json.load(fh) target_directory = params['output_data'][0]['extra_files_path'] os.mkdir(target_directory) @@ -89,10 +95,11 @@ sequence_id, sequence_name = get_id_name(params, dbkey=dbkey, fasta_description=options.fasta_description) # build the index - build_twobit(data_manager_dict, options.fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name) + build_twobit(data_manager_dict, options.fasta_filename, params, target_directory, dbkey, sequence_id, sequence_name, options.long) # save info to json file - open(filename, 'w').write(dumps(data_manager_dict, sort_keys=True)) + with open(filename, 'w') as fh: + json.dump(data_manager_dict, fh, sort_keys=True) if __name__ == "__main__":
--- a/data_manager/twobit_builder.xml Thu Nov 28 20:33:03 2019 +0000 +++ b/data_manager/twobit_builder.xml Sun Apr 16 08:15:04 2023 +0000 @@ -1,4 +1,4 @@ -<tool id="twobit_builder_data_manager" name="TwoBit" tool_type="manage_data" version="0.0.4" profile="19.05"> +<tool id="twobit_builder_data_manager" name="TwoBit" tool_type="manage_data" version="0.0.5" profile="19.05"> <requirements> <requirement type="package" version="377">ucsc-fatotwobit</requirement> <requirement type="package" version="3.7">python</requirement> @@ -7,6 +7,7 @@ <command detect_errors="exit_code"><![CDATA[ python '$__tool_directory__/twobit_builder.py' '${out_file}' +$long --fasta_filename '${all_fasta_source.fields.path}' --fasta_dbkey '${all_fasta_source.fields.dbkey}' --fasta_description '${all_fasta_source.fields.name}' @@ -17,6 +18,7 @@ </param> <param name="sequence_name" type="text" value="" label="Name of sequence" /> <param name="sequence_id" type="text" value="" label="ID for sequence" /> + <param argument="--long" type="boolean" truevalue="--long" falsevalue="" label="Use the -long option for very big genomes." /> </inputs> <outputs> <data name="out_file" format="data_manager_json"/>
--- a/data_manager_conf.xml Thu Nov 28 20:33:03 2019 +0000 +++ b/data_manager_conf.xml Sun Apr 16 08:15:04 2023 +0000 @@ -1,7 +1,7 @@ <?xml version="1.0"?> <data_managers> - <data_manager tool_file="data_manager/twobit_builder.xml" id="twobit_builder" version="0.0.2"> + <data_manager tool_file="data_manager/twobit_builder.xml" id="twobit_builder"> <data_table name="lastz_seqs"> <output> <column name="value" />