def _get_input_files(samples, base_dir, tx_out_dir):
"""Retrieve input files, keyed by sample and QC method name.
Stages files into the work directory to ensure correct names for
MultiQC sample assessment when running with CWL.
"""
in_files = collections.defaultdict(list)
for data in samples:
sum_qc = tz.get_in(["summary", "qc"], data, {})
if sum_qc in [None, "None"]:
sum_qc = {}
elif isinstance(sum_qc, six.string_types):
sum_qc = {dd.get_algorithm_qc(data)[0]: sum_qc}
elif not isinstance(sum_qc, dict):
raise ValueError("Unexpected summary qc: %s" % sum_qc)
for program, pfiles in sum_qc.items():
if isinstance(pfiles, dict):
pfiles = [pfiles["base"]] + pfiles.get("secondary", [])
# CWL: presents output files as single file plus associated secondary files
elif isinstance(pfiles, six.string_types):
if os.path.exists(pfiles):
pfiles = [
os.path.join(basedir, f)
for basedir, subdir, filenames in os.walk(
os.path.dirname(pfiles)) for f in filenames
]
else:
pfiles = []
in_files[(dd.get_sample_name(data), program)].extend(pfiles)
staged_files = []
for (sample, program), files in in_files.items():
cur_dir = utils.safe_makedir(
os.path.join(base_dir, "inputs", sample, program))
for f in files:
if _check_multiqc_input(f) and _is_good_file_for_multiqc(f):
if _in_temp_directory(f) or any(
[cwlutils.is_cwl_run(d) for d in samples]):
staged_f = os.path.join(cur_dir, os.path.basename(f))
shutil.copy(f, staged_f)
staged_files.append(staged_f)
else:
staged_files.append(f)
staged_files.extend(get_qsig_multiqc_files(samples))
# Back compatible -- to migrate to explicit specifications in input YAML
if not any([cwlutils.is_cwl_run(d) for d in samples]):
staged_files += ["trimmed", "htseq-count/*summary"]
# Add in created target_info file
if os.path.isfile(
os.path.join(base_dir, "report", "metrics",
"target_info.yaml")):
staged_files += [
os.path.join(base_dir, "report", "metrics", "target_info.yaml")
]
return sorted(list(set(staged_files)))
def _get_input_files(samples, base_dir, tx_out_dir):
"""Retrieve input files, keyed by sample and QC method name.
Stages files into the work directory to ensure correct names for
MultiQC sample assessment when running with CWL.
"""
in_files = collections.defaultdict(list)
for data in samples:
sum_qc = tz.get_in(["summary", "qc"], data, {})
if sum_qc in [None, "None"]:
sum_qc = {}
elif isinstance(sum_qc, six.string_types):
sum_qc = {dd.get_algorithm_qc(data)[0]: sum_qc}
elif not isinstance(sum_qc, dict):
raise ValueError("Unexpected summary qc: %s" % sum_qc)
for program, pfiles in sum_qc.items():
if isinstance(pfiles, dict):
pfiles = [pfiles["base"]] + pfiles.get("secondary", [])
# CWL: presents output files as single file plus associated secondary files
elif isinstance(pfiles, six.string_types):
if os.path.exists(pfiles):
pfiles = [os.path.join(basedir, f) for basedir, subdir, filenames in os.walk(os.path.dirname(pfiles)) for f in filenames]
else:
pfiles = []
in_files[(dd.get_sample_name(data), program)].extend(pfiles)
staged_files = []
for (sample, program), files in in_files.items():
cur_dir = utils.safe_makedir(os.path.join(base_dir, "inputs", sample, program))
for f in files:
if _check_multiqc_input(f) and _is_good_file_for_multiqc(f):
if _in_temp_directory(f) or any([cwlutils.is_cwl_run(d) for d in samples]):
staged_f = os.path.join(cur_dir, os.path.basename(f))
shutil.copy(f, staged_f)
staged_files.append(staged_f)
else:
staged_files.append(f)
staged_files.extend(get_qsig_multiqc_files(samples))
# Back compatible -- to migrate to explicit specifications in input YAML
if not any([cwlutils.is_cwl_run(d) for d in samples]):
staged_files += ["trimmed", "htseq-count/*summary"]
# Add in created target_info file
if os.path.isfile(os.path.join(base_dir, "report", "metrics", "target_info.yaml")):
staged_files += [os.path.join(base_dir, "report", "metrics", "target_info.yaml")]
return sorted(list(set(staged_files)))