def process_submission_directory(basepath,
submission_file_path,
recid,
update=False,
*args,
**kwargs):
"""
Goes through an entire submission directory and processes the
files within to create DataSubmissions
with the files and related material attached as DataResources.
:param basepath:
:param submission_file_path:
:param recid:
:param update:
:return:
"""
added_file_names = []
errors = {}
if submission_file_path is not None:
submission_file_validator = SubmissionFileValidator()
is_valid_submission_file = submission_file_validator.validate(
file_path=submission_file_path)
if is_valid_submission_file:
submission_file = open(submission_file_path, 'r')
submission_processed = yaml.load_all(submission_file,
Loader=Loader)
# process file, extracting contents, and linking
# the data record with the parent publication
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
if hepsubmission is None:
HEPSubmission(publication_recid=recid,
overall_status='todo',
inspire_id=hepsubmission.inspire_id,
coordinator=kwargs.get('user_id') if 'user_id'
in kwargs else int(current_user.get_id()),
version=hepsubmission.version + 1)
# On a new upload, we reset the flag to notify reviewers
hepsubmission.reviewers_notified = False
# if it is finished and we receive an update,
# then we need to reopen the submission to allow for revisions.
if hepsubmission.overall_status == 'finished' and not update:
# we create a new HEPSubmission object
_rev_hepsubmission = HEPSubmission(
publication_recid=recid,
overall_status='todo',
inspire_id=hepsubmission.inspire_id,
coordinator=hepsubmission.coordinator,
version=hepsubmission.version + 1)
db.session.add(_rev_hepsubmission)
hepsubmission = _rev_hepsubmission
reserve_doi_for_hepsubmission(hepsubmission, update)
no_general_submission_info = True
data_file_validator = DataFileValidator()
# Delete all data records associated with this submission.
# Fixes problems with ordering where the table names are changed between uploads.
# See https://github.com/HEPData/hepdata/issues/112
# Side effect that reviews will be deleted between uploads.
cleanup_submission(recid, hepsubmission.version, added_file_names)
for yaml_document_index, yaml_document in enumerate(
submission_processed):
if not yaml_document:
continue
# Check for presence of local files given as additional_resources.
if 'additional_resources' in yaml_document:
for resource in yaml_document['additional_resources']:
location = os.path.join(basepath, resource['location'])
if not resource['location'].startswith(
('http', '/resource/')):
if not os.path.isfile(location):
errors[resource['location']] = [{
"level":
"error",
"message":
"Missing 'additional_resources' file from uploaded archive."
}]
elif '/' in resource['location']:
errors[resource['location']] = [{
"level":
"error",
"message":
"Location of 'additional_resources' file should not contain '/'."
}]
if not yaml_document_index and 'name' not in yaml_document:
no_general_submission_info = False
process_general_submission_info(basepath, yaml_document,
recid)
elif not all(k in yaml_document
for k in ('name', 'description', 'keywords',
'data_file')):
errors["submission.yaml"] = [{
"level":
"error",
"message":
"YAML document with index {} ".format(
yaml_document_index) +
"missing one or more required keys (name, description, keywords, data_file)."
}]
else:
existing_datasubmission_query = DataSubmission.query \
.filter_by(name=encode_string(yaml_document["name"]),
publication_recid=recid,
version=hepsubmission.version)
added_file_names.append(yaml_document["name"])
try:
if existing_datasubmission_query.count() == 0:
datasubmission = DataSubmission(
publication_recid=recid,
name=encode_string(yaml_document["name"]),
description=encode_string(
yaml_document["description"]),
version=hepsubmission.version)
else:
datasubmission = existing_datasubmission_query.one(
)
datasubmission.description = encode_string(
yaml_document["description"])
db.session.add(datasubmission)
except SQLAlchemyError as sqlex:
errors[yaml_document["data_file"]] = [{
"level":
"error",
"message":
str(sqlex)
}]
db.session.rollback()
continue
main_file_path = os.path.join(basepath,
yaml_document["data_file"])
data, ex = _eos_fix_read_data(main_file_path)
if not data or data is None or ex is not None:
errors[yaml_document["data_file"]] = \
[{"level": "error", "message": "There was a problem parsing the file.\n" + str(ex)}]
elif '/' in yaml_document["data_file"]:
errors[yaml_document["data_file"]] = \
[{"level": "error", "message": "Name of data_file should not contain '/'.\n"}]
else:
if data_file_validator.validate(
file_path=main_file_path, data=data):
try:
process_data_file(recid, hepsubmission.version,
basepath, yaml_document,
datasubmission,
main_file_path)
except SQLAlchemyError as sqlex:
errors[yaml_document["data_file"]] = [{
"level":
"error",
"message":
"There was a problem processing the file.\n"
+ str(sqlex)
}]
db.session.rollback()
else:
errors = process_validation_errors_for_display(
data_file_validator.get_messages())
data_file_validator.clear_messages()
if yaml_document["data_file"] not in errors:
# Check that the length of the 'values' list is consistent
# for each of the independent_variables and dependent_variables.
indep_count = [
len(indep['values'])
for indep in data['independent_variables']
]
dep_count = [
len(dep['values'])
for dep in data['dependent_variables']
]
if len(set(indep_count + dep_count)
) > 1: # if more than one unique count
errors.setdefault(
yaml_document["data_file"], []
).append({
"level":
"error",
"message":
"Inconsistent length of 'values' list:\n" +
"independent_variables{}, dependent_variables{}"
.format(str(indep_count), str(dep_count))
})
submission_file.close()
if no_general_submission_info:
hepsubmission.last_updated = datetime.now()
db.session.add(hepsubmission)
db.session.commit()
# The line below is commented out since it does not preserve the order of tables.
# Delete all tables above instead: side effect of deleting reviews between uploads.
#cleanup_submission(recid, hepsubmission.version, added_file_names)
db.session.commit()
if len(errors) is 0:
errors = package_submission(basepath, recid, hepsubmission)
reserve_dois_for_data_submissions(
publication_recid=recid, version=hepsubmission.version)
admin_indexer = AdminIndexer()
admin_indexer.index_submission(hepsubmission)
else: # delete all tables if errors
cleanup_submission(recid, hepsubmission.version, {})
else:
errors = process_validation_errors_for_display(
submission_file_validator.get_messages())
submission_file_validator.clear_messages()
else:
# return an error
errors = {
"submission.yaml": [{
"level":
"error",
"message":
"No submission.yaml file found in submission."
}]
}
return errors
# we return all the errors collectively.
# This makes more sense that returning errors as
# soon as problems are found on one file.
return errors
def process_submission_directory(basepath,
submission_file_path,
recid,
update=False,
*args,
**kwargs):
"""
Goes through an entire submission directory and processes the
files within to create DataSubmissions
with the files and related material attached as DataResources.
:param basepath:
:param submission_file_path:
:param recid:
:return:
"""
added_file_names = []
errors = {}
if submission_file_path is not None:
submission_file = open(submission_file_path, 'r')
submission_file_validator = SubmissionFileValidator()
is_valid_submission_file = submission_file_validator.validate(
file_path=submission_file_path)
data_file_validator = DataFileValidator()
if is_valid_submission_file:
try:
submission_processed = yaml.load_all(submission_file,
Loader=yaml.CSafeLoader)
except:
submission_processed = yaml.safe_load_all(submission_file)
# process file, extracting contents, and linking
# the data record with the parent publication
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
if hepsubmission is None:
HEPSubmission(publication_recid=recid,
overall_status='todo',
inspire_id=hepsubmission.inspire_id,
coordinator=kwargs.get('user_id') if 'user_id'
in kwargs else int(current_user.get_id()),
version=hepsubmission.version + 1)
# On a new upload, we reset the flag to notify reviewers
hepsubmission.reviewers_notified = False
# if it is finished and we receive an update,
# then we need to reopen the submission to allow for revisions.
if hepsubmission.overall_status == 'finished' and not update:
# we create a new HEPSubmission object
_rev_hepsubmission = HEPSubmission(
publication_recid=recid,
overall_status='todo',
inspire_id=hepsubmission.inspire_id,
coordinator=hepsubmission.coordinator,
version=hepsubmission.version + 1)
db.session.add(_rev_hepsubmission)
hepsubmission = _rev_hepsubmission
reserve_doi_for_hepsubmission(hepsubmission, update)
for yaml_document in submission_processed:
if 'record_ids' in yaml_document or 'comment' in yaml_document or 'modifications' in yaml_document:
# comments are only present in the general submission
# information document.
process_general_submission_info(basepath, yaml_document,
recid)
else:
existing_datasubmission_query = DataSubmission.query \
.filter_by(name=encode_string(yaml_document["name"]),
publication_recid=recid,
version=hepsubmission.version)
added_file_names.append(yaml_document["name"])
if existing_datasubmission_query.count() == 0:
datasubmission = DataSubmission(
publication_recid=recid,
name=encode_string(yaml_document["name"]),
description=encode_string(
yaml_document["description"]),
version=hepsubmission.version)
else:
datasubmission = existing_datasubmission_query.one()
datasubmission.description = encode_string(
yaml_document["description"])
db.session.add(datasubmission)
main_file_path = os.path.join(basepath,
yaml_document["data_file"])
if data_file_validator.validate(file_path=main_file_path):
process_data_file(recid, hepsubmission.version,
basepath, yaml_document,
datasubmission, main_file_path)
else:
errors = process_validation_errors_for_display(
data_file_validator.get_messages())
data_file_validator.clear_messages()
cleanup_submission(recid, hepsubmission.version, added_file_names)
db.session.commit()
if len(errors) is 0:
package_submission(basepath, recid, hepsubmission)
reserve_dois_for_data_submissions(recid, hepsubmission.version)
admin_indexer = AdminIndexer()
admin_indexer.index_submission(hepsubmission)
else:
errors = process_validation_errors_for_display(
submission_file_validator.get_messages())
submission_file_validator.clear_messages()
data_file_validator.clear_messages()
else:
# return an error
errors = {
"submission.yaml": [{
"level":
"error",
"message":
"No submission.yaml file found in submission."
}]
}
return errors
# we return all the errors collectively.
# This makes more sense that returning errors as
# soon as problems are found on one file.
return errors
def process_submission_directory(basepath, submission_file_path, recid, update=False, *args, **kwargs):
"""
Goes through an entire submission directory and processes the
files within to create DataSubmissions
with the files and related material attached as DataResources.
:param basepath:
:param submission_file_path:
:param recid:
:return:
"""
added_file_names = []
errors = {}
if submission_file_path is not None:
submission_file = open(submission_file_path, 'r')
submission_file_validator = SubmissionFileValidator()
is_valid_submission_file = submission_file_validator.validate(
file_path=submission_file_path)
data_file_validator = DataFileValidator()
if is_valid_submission_file:
try:
submission_processed = yaml.load_all(submission_file, Loader=yaml.CSafeLoader)
except:
submission_processed = yaml.safe_load_all(submission_file)
# process file, extracting contents, and linking
# the data record with the parent publication
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
if hepsubmission is None:
HEPSubmission(publication_recid=recid,
overall_status='todo',
inspire_id=hepsubmission.inspire_id,
coordinator=kwargs.get('user_id') if 'user_id' in kwargs else int(current_user.get_id()),
version=hepsubmission.version + 1)
# On a new upload, we reset the flag to notify reviewers
hepsubmission.reviewers_notified = False
# if it is finished and we receive an update,
# then we need to reopen the submission to allow for revisions.
if hepsubmission.overall_status == 'finished' and not update:
# we create a new HEPSubmission object
_rev_hepsubmission = HEPSubmission(publication_recid=recid,
overall_status='todo',
inspire_id=hepsubmission.inspire_id,
coordinator=hepsubmission.coordinator,
version=hepsubmission.version + 1)
db.session.add(_rev_hepsubmission)
hepsubmission = _rev_hepsubmission
reserve_doi_for_hepsubmission(hepsubmission)
for yaml_document in submission_processed:
if 'record_ids' in yaml_document or 'comment' in yaml_document or 'modifications' in yaml_document:
# comments are only present in the general submission
# information document.
process_general_submission_info(basepath, yaml_document, recid)
else:
existing_datasubmission_query = DataSubmission.query \
.filter_by(name=encode_string(yaml_document["name"]),
publication_recid=recid,
version=hepsubmission.version)
added_file_names.append(yaml_document["name"])
if existing_datasubmission_query.count() == 0:
datasubmission = DataSubmission(
publication_recid=recid,
name=encode_string(yaml_document["name"]),
description=encode_string(
yaml_document["description"]),
version=hepsubmission.version)
else:
datasubmission = existing_datasubmission_query.one()
datasubmission.description = encode_string(
yaml_document["description"])
db.session.add(datasubmission)
main_file_path = os.path.join(basepath,
yaml_document["data_file"])
if data_file_validator.validate(file_path=main_file_path):
process_data_file(recid, hepsubmission.version, basepath, yaml_document,
datasubmission, main_file_path)
else:
errors = process_validation_errors_for_display(
data_file_validator.get_messages())
data_file_validator.clear_messages()
cleanup_submission(recid, hepsubmission.version,
added_file_names)
db.session.commit()
if len(errors) is 0:
package_submission(basepath, recid, hepsubmission)
reserve_dois_for_data_submissions(recid, hepsubmission.version)
admin_indexer = AdminIndexer()
admin_indexer.index_submission(hepsubmission)
else:
errors = process_validation_errors_for_display(
submission_file_validator.get_messages())
submission_file_validator.clear_messages()
data_file_validator.clear_messages()
else:
# return an error
errors = {"submission.yaml": [
{"level": "error",
"message": "No submission.yaml file found in submission."}
]}
return errors
# we return all the errors collectively.
# This makes more sense that returning errors as
# soon as problems are found on one file.
return errors
