def script__test_ssi_stamper(sample, sample_yaml, sample_component, log_err):
# Genering error handling to redirect output to stderr file
try:
comps = get_components(sample)
if "detected_species" in sample["properties"]:
species = datahandling.load_species(
sample["properties"]["detected_species"])
else:
species = datahandling.load_species(None)
results, summary, stamp = stamps.ssi_stamp.test(
comps['whats_my_species'], comps['assemblatron'], species, sample)
datadump_dict = datahandling.load_sample_component(sample_component)
datadump_dict["summary"] = summary
datadump_dict["results"] = results
datahandling.save_sample_component(datadump_dict, sample_component)
# Get the _id back
datadump_dict = datahandling.load_sample_component(sample_component)
stamp["_sample_component"] = datadump_dict["_id"]
stamp_dict = sample.get("stamps", {})
stamp_list = stamp_dict.get("stamp_list", [])
stamp_list.append(stamp)
stamp_dict["stamp_list"] = stamp_list
stamp_dict[stamp["name"]] = stamp
sample["stamps"] = stamp_dict
datahandling.save_sample(sample, sample_yaml)
return 0
except Exception:
datahandling.log(log_err, str(traceback.format_exc()))
exit(1)
def passes_check_reads_pipeline(sample, requirements_file, log_err):
"""
Checks if the component is a pipeline. In that case it will require reads to be present
so the component can run.
"""
sample_db = datahandling.load_yaml(sample)
if requirements_file["type"] == "pipeline":
if "reads" not in sample_db:
datahandling.log(
log_err,
"Pipeline component can't run on a sample with no reads. db:{}"
.format(sample_db))
return False
return True
def script__initialization(requirements_file, sample, sample_component,
output_file, log_out, log_err):
set_status_to_running(sample_component)
if requirements_met(requirements_file, sample, log_out, log_err) == True:
datahandling.log(log_out, "{}\n{}\n".format(os.getcwd(), output_file))
with open(str(output_file), "w") as handle:
handle.write("Requirements met")
pass
else:
datahandling.log(log_err, "Requirements not met")
sample_component_entry = datahandling.load_sample_component(
sample_component)
sample_component_entry["status"] = "Requirements not met"
datahandling.save_sample_component(sample_component_entry,
sample_component)
return 0
def requirements_met(requirements_file, sample, log_out, log_err):
requirements_file = datahandling.load_yaml(requirements_file)
sample_name = datahandling.load_yaml(sample)["name"]
if not passes_check_reads_pipeline(sample, requirements_file, log_err):
return False
no_failures = True
if requirements_file.get('requirements', None) is not None:
df = pandas.io.json.json_normalize(
requirements_file.get('requirements')
) # flattens json into key while maintaining values https://stackoverflow.com/a/41801708
requirements_dict = df.to_dict(orient='records')[0]
requirements = []
for key in requirements_dict:
values = key.split('.')
if values[0] == 'components':
file_location = sample_name + "__" + values[1] + ".yaml"
requirement = values[2:] # TODO: add check for no requirements
expected_value = requirements_dict[key]
requirements.append(
[file_location, requirement, expected_value])
elif values[0] == 'sample':
file_location = "sample.yaml"
requirement = values[1:] # TODO: add check for no requirements
expected_value = requirements_dict[key]
requirements.append(
[file_location, requirement, expected_value])
elif values[0] == 'run':
file_location = "run.yaml"
requirement = values[1:] # TODO: add check for no requirements
expected_value = requirements_dict[key]
requirements.append(
[file_location, requirement, expected_value])
else:
datahandling.log(log_err,
"Improper requirement {}".format(key))
for requirement in requirements:
# requirements are in form [file_path, [keys,key2,...], value_of_key (optional otherwise None)]
file_location = requirement[0]
keys = requirement[1]
desired_value = requirement[2]
db = datahandling.load_yaml(file_location)
"""
What this does is run dict.get interatively on the db based on the keys till it can 't go
deeper then returns the value or None if it couldn' t reach that level. While used with dict.get
any function can be passed
https://pymotw.com/3/functools/
"""
actual_value = functools.reduce(dict.get, keys, db)
"""
Check has been adjusted to check for a list to allow multiple potential options to match
"""
if not isinstance(desired_value, list):
desired_value = [desired_value]
# As it should be a list because of the last line.
if actual_value is not None:
# Not sure why desired is [None] instead of None
if desired_value != [None]:
if actual_value in desired_value:
datahandling.log(
log_err,
"Found required entry (value checked) for\ndb: {}\nentry: {}\n"
.format(":".join(keys), db))
else:
datahandling.log(
log_err,
"Requirements not met for\ndb: {}\nentry: {}\ndesired_entry: {}\n"
.format(":".join(keys), db, desired_value))
no_failures = False
else:
datahandling.log(
log_err,
"Found required entry (value not checked) for\ndb: {}\nentry: {}\n"
.format(":".join(keys), db))
else:
datahandling.log(
log_err,
"Requirements not met for\ndb: {}\nentry: {}\n".format(
file_location, ":".join(keys)))
no_failures = False
if no_failures:
return True
else:
return False