def main():
# because of ruffus, have to use some global variables
# global variables: options, config, samples, env, logger, logger_mutex
# minimize the number of global variables as much as possible
global options, config
options = parse_args_for_rp_run()
config = misc.get_config(options.config_file)
global samples
G = PPR.gen_all_samples_from_soft_and_isamp
samples = G(options.soft_files, options.isamp, config)
PPR.init_sample_outdirs(samples, config['LOCAL_TOP_OUTDIR'])
PPR.fetch_sras_info(samples, options.recreate_sras_info)
top_outdir = config['LOCAL_TOP_OUTDIR']
cmd_df = config['LOCAL_CMD_DF']
min_free = misc.ugly_usage(config['LOCAL_MIN_FREE'])
max_usage = misc.ugly_usage(config['LOCAL_MAX_USAGE'])
free_to_use = calc_local_free_space_to_use(top_outdir, cmd_df, min_free,
max_usage)
logger.info('Selecting samples to process based their usage')
samples = PPR.select_gsms_to_process(samples, free_to_use)
if not samples: # when samples == []
logger.info('Cannot find a GSM that fits the disk usage rule')
return
logger.info('GSMs to process:')
for k, gsm in enumerate(samples):
logger.info('\t{0:3d} {1:30s} {2}'.format(k + 1, gsm, gsm.outdir))
if 'gen_qsub_script' in options.target_tasks:
if not options.qsub_template:
raise IOError(
'-t/--qsub_template required when running gen_qsub_script')
R.pipeline_run(
logger=logger,
target_tasks=options.target_tasks,
forcedtorun_tasks=options.forced_tasks,
multiprocess=options.jobs,
verbose=options.verbose,
touch_files_only=options.touch_files_only,
# history_file=os.path.join('log', '.{0}.sqlite'.format(
# '_'.join([_.name for _ in sorted(samples, key=lambda x: x.name)])))
)
def main():
# because of ruffus, have to use some global variables
# global variables: options, config, samples, env, logger, logger_mutex
# minimize the number of global variables as much as possible
global options, config
options = parse_args_for_rp_run()
config = misc.get_config(options.config_file)
global samples
G = PPR.gen_all_samples_from_soft_and_isamp
samples = G(options.soft_files, options.isamp, config)
PPR.init_sample_outdirs(samples, config['LOCAL_TOP_OUTDIR'])
PPR.fetch_sras_info(samples, options.recreate_sras_info)
top_outdir = config['LOCAL_TOP_OUTDIR']
cmd_df = config['LOCAL_CMD_DF']
min_free = misc.ugly_usage(config['LOCAL_MIN_FREE'])
max_usage = misc.ugly_usage(config['LOCAL_MAX_USAGE'])
free_to_use = calc_local_free_space_to_use(
top_outdir, cmd_df, min_free, max_usage)
logger.info('Selecting samples to process based their usage')
samples = PPR.select_gsms_to_process(samples, free_to_use)
if not samples: # when samples == []
logger.info('Cannot find a GSM that fits the disk usage rule')
return
logger.info('GSMs to process:')
for k, gsm in enumerate(samples):
logger.info('\t{0:3d} {1:30s} {2}'.format(k+1, gsm, gsm.outdir))
if 'gen_qsub_script' in options.target_tasks:
if not options.qsub_template:
raise IOError('-t/--qsub_template required when running gen_qsub_script')
R.pipeline_run(
logger=logger,
target_tasks=options.target_tasks,
forcedtorun_tasks=options.forced_tasks,
multiprocess=options.jobs,
verbose=options.verbose,
touch_files_only=options.touch_files_only,
# history_file=os.path.join('log', '.{0}.sqlite'.format(
# '_'.join([_.name for _ in sorted(samples, key=lambda x: x.name)])))
)
def test_select_gsms_to_process_ignore_disk_usage(self, mock_is_processed):
mock_is_processed.return_value = False
samples = [mock.Mock(), mock.Mock()]
self.assertEqual(ppr.select_gsms_to_process(samples, 1024 ** 3, True), samples)
def test_select_gsms_to_process_fit_disk_usage(self, mock_is_processed, mock_estimate_sra2fastq_usage):
mock_is_processed.return_value = False
mock_estimate_sra2fastq_usage.return_value = 513
samples = [mock.Mock(), mock.Mock()]
self.assertEqual(ppr.select_gsms_to_process(samples, 1024, False), [samples[0]])
def test_select_gsms_to_process_all_processed(self, mock_is_processed):
mock_is_processed.return_value = True
samples = [mock.Mock(), mock.Mock()]
self.assertEqual(ppr.select_gsms_to_process(samples, 1024 ** 3, False), [])