def get_data_path(record_id=None, inspire_id=None):
"""Gets the file path where data files for the given record are stored."""
if record_id:
# Check record exists
hepsubmission = get_latest_hepsubmission(publication_recid=record_id)
if hepsubmission is None:
click.echo("No record with id %s" % record_id)
return
elif inspire_id:
hepsubmission = get_latest_hepsubmission(inspire_id=inspire_id)
if hepsubmission is None:
click.echo("No record with inspire id %s" % inspire_id)
return
else:
record_id = hepsubmission.publication_recid
click.echo("Inspire ID %s maps to record id %s" %
(inspire_id, record_id))
else:
click.echo("Please provide either record-id or inspire-id.")
return
click.echo("Files for record %s are at:\t\t %s" %
(record_id, data_files.get_data_path_for_record(record_id)))
click.echo("Converted files for record %s are at:\t %s" %
(record_id, data_files.get_converted_directory_path(record_id)))
def send_cookie_email(submission_participant,
record_information,
message=None,
version=1):
hepsubmission = get_latest_hepsubmission(
publication_recid=record_information['recid'])
coordinator = User.query.get(hepsubmission.coordinator)
collaboration = _get_collaboration(hepsubmission.coordinator)
message_body = render_template(
'hepdata_theme/email/invite.html',
name=submission_participant.full_name,
role=submission_participant.role,
title=record_information['title'],
site_url=current_app.config.get('SITE_URL', 'https://www.hepdata.net'),
user_account=submission_participant.user_account,
invite_token=submission_participant.invitation_cookie,
status=submission_participant.status,
recid=submission_participant.publication_recid,
version=version,
email=submission_participant.email,
coordinator_email=coordinator.email,
collaboration=collaboration,
message=message)
create_send_email_task(
submission_participant.email,
"[HEPData] Invitation to be {0} {1} of record {2} in HEPData".format(
"an" if submission_participant.role == "uploader" else "a",
submission_participant.role.capitalize(),
submission_participant.publication_recid), message_body)
def get_pending_invitations_for_user(user):
"""
Returns pending invites for upload or review of records.
:param user: User object
:return: array of pending invites
"""
pending_invites = SubmissionParticipant.query.filter(
func.lower(SubmissionParticipant.email) == func.lower(user.email),
or_(SubmissionParticipant.role == 'reviewer',
SubmissionParticipant.role == 'uploader'),
SubmissionParticipant.status == 'primary',
SubmissionParticipant.user_account == None).all()
result = []
for invite in pending_invites:
publication_record = get_record_by_id(invite.publication_recid)
hepsubmission = get_latest_hepsubmission(
publication_recid=invite.publication_recid)
coordinator = get_user_from_id(hepsubmission.coordinator)
result.append({
'title': decode_string(publication_record['title']),
'invitation_cookie': invite.invitation_cookie,
'role': invite.role,
'coordinator': coordinator
})
return result
def send_question_email(question):
reply_to = current_user.email
submission = get_latest_hepsubmission(
publication_recid=question.publication_recid)
submission_participants = get_submission_participants_for_record(
question.publication_recid)
if submission:
destinations = [current_app.config['ADMIN_EMAIL']]
for submission_participant in submission_participants:
destinations.append(submission_participant.email)
coordinator = User.query.get(submission.coordinator)
if coordinator.id > 1:
destinations.append(coordinator.email)
if len(destinations) > 0:
message_body = render_template('hepdata_theme/email/question.html',
inspire_id=submission.inspire_id,
user_email=reply_to,
site_url=current_app.config.get(
'SITE_URL',
'https://www.hepdata.net'),
message=question.question)
create_send_email_task(
destination=','.join(set(destinations)),
subject="[HEPData] Question for record ins{0}".format(
submission.inspire_id),
message=message_body,
reply_to_address=reply_to)
def add_participant(recid):
"""
Adds a participant to a record
:param recid:
:return:
"""
try:
submission_record = get_latest_hepsubmission(publication_recid=recid)
full_name = request.form['name']
email = request.form['email']
participant_type = request.form['type']
new_record = SubmissionParticipant(publication_recid=recid,
full_name=full_name,
email=email, role=participant_type)
submission_record.participants.append(new_record)
db.session.commit()
return json.dumps(
{"success": True, "recid": recid,
"message": "{0} {1} added.".format(full_name, participant_type)})
except Exception as e:
return json.dumps(
{"success": False, "recid": recid,
"message": 'Unable to add participant.'})
def download_submission_with_inspire_id(*args, **kwargs):
"""
Gets the submission file and either serves it back directly from YAML, or converts it
for other formats.
:param inspire_id: inspire id
:param version: version of submission to export. If -1, returns the latest.
:param file_format: yaml, csv, ROOT, or YODA
:return:
"""
inspire_id = kwargs.pop('inspire_id')
if 'ins' in inspire_id:
inspire_id = inspire_id.replace('ins', '')
if 'version' in kwargs:
submission = HEPSubmission.query.filter_by(inspire_id=inspire_id, version=kwargs.pop('version')).first()
else:
submission = get_latest_hepsubmission(inspire_id=inspire_id)
if not submission:
return display_error(
title="No submission found",
description="A submission with INSPIRE id {0} does not exist".format(inspire_id)
)
return download_submission(submission, kwargs.pop('file_format'))
def download_data_table_by_recid(*args, **kwargs):
"""
Record ID download
Downloads the latest data file given the url /download/submission/1231/Table 1/yaml or
by a particular version given /download/submission/1231/Table 1/1/yaml
:param args:
:param kwargs: inspire_id, table_name, version (optional), and file_format
:return:
"""
recid = kwargs.pop('recid')
table_name = kwargs.pop('table_name')
# Allow space in table_name to be omitted from URL.
if ' ' not in table_name:
table_name = table_name.replace('Table', 'Table ')
if 'version' not in kwargs:
version = get_latest_hepsubmission(publication_recid=recid).version
else:
version = kwargs.pop('version')
datasubmission = DataSubmission.query.filter_by(publication_recid=recid,
version=version,
name=table_name).one()
return download_datatable(datasubmission,
kwargs.pop('file_format'),
submission_id='{0}'.format(recid),
table_name=table_name)
def send_question_email(question):
reply_to = current_user.email
submission = get_latest_hepsubmission(publication_recid=question.publication_recid)
submission_participants = get_submission_participants_for_record(question.publication_recid)
if submission:
destinations = [current_app.config['ADMIN_EMAIL']]
for submission_participant in submission_participants:
destinations.append(submission_participant.email)
if submission.coordinator > 1:
destinations.append(submission.coordinator.email)
if len(destinations) > 0:
message_body = render_template(
'hepdata_theme/email/question.html',
inspire_id=submission.inspire_id,
user_email=reply_to,
site_url=current_app.config.get('SITE_URL', 'https://www.hepdata.net'),
message=question.question)
create_send_email_task(destination=','.join(destinations),
subject="[HEPData] Question for record ins{0}".format(submission.inspire_id),
message=message_body, reply_to_address=reply_to)
def download_data_table_by_recid(*args, **kwargs):
"""
Record ID download
Downloads the latest data file given the url /download/submission/1231/Table 1/yaml or
by a particular version given /download/submission/1231/Table 1/yaml
:param args:
:param kwargs: inspire_id, table_name, version (optional), and file_format
:return:
"""
recid = kwargs.pop('recid')
table_name = kwargs.pop('table_name')
if 'version' not in kwargs:
version = get_latest_hepsubmission(publication_recid=recid).version
else:
version = kwargs.pop('version')
datasubmission = db.session.query(DataResource) \
.join(DataSubmission) \
.filter(DataSubmission.publication_recid == recid, version=version,
name=table_name) \
.one()
return download_datatable(datasubmission, kwargs.pop('file_format'),
submission_id='{0}'.format(recid), table_name=table_name)
def download_data_table_by_inspire_id(*args, **kwargs):
"""
Downloads the latest data file given the url /download/submission/ins1283842/Table 1/yaml or
by a particular version given /download/submission/ins1283842/Table 1/yaml
:param args:
:param kwargs: inspire_od, table_name, version (optional), and file_format
:return:
"""
inspire_id = kwargs.pop('inspire_id')
table_name = kwargs.pop('table_name')
if 'ins' in inspire_id:
inspire_id = inspire_id.replace('ins', '')
if 'version' not in kwargs:
version = get_latest_hepsubmission(inspire_id=inspire_id).version
else:
version = kwargs.pop('version')
datasubmission = db.session.query(DataResource) \
.join(DataSubmission) \
.filter(DataSubmission.publication_inspire_id == inspire_id, DataSubmission.version == version,
DataSubmission.name == table_name) \
.one()
return download_datatable(datasubmission, kwargs.pop('file_format'),
submission_id='ins{0}'.format(inspire_id), table_name=table_name)
def get_resources(recid, version):
"""
Gets a list of resources for a publication, relevant to all data records.
:param recid:
:return: json
"""
result = {'submission_items': []}
common_resources = {'name': 'Common Resources', 'type': 'submission', 'version': version, 'id': recid,
'resources': []}
submission = get_latest_hepsubmission(publication_recid=recid, version=version)
if submission:
for reference in submission.resources:
common_resources['resources'].append(process_resource(reference))
result['submission_items'].append(common_resources)
datasubmissions = DataSubmission.query.filter_by(publication_recid=recid, version=version).\
order_by(DataSubmission.id.asc()).all()
for datasubmission in datasubmissions:
submission_item = {'name': datasubmission.name, 'type': 'data', 'id': datasubmission.id, 'resources': [],
'version': datasubmission.version}
for reference in datasubmission.resources:
submission_item['resources'].append(process_resource(reference))
result['submission_items'].append(submission_item)
return json.dumps(result)
def create_new_version(recid, user, notify_uploader=True, uploader_message=None):
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
if hepsubmission.overall_status == 'finished':
# Reopen the submission to allow for revisions,
# by creating 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)
db.session.commit()
if notify_uploader:
uploaders = SubmissionParticipant.query.filter_by(
role='uploader', publication_recid=recid, status='primary'
)
record_information = get_record_by_id(recid)
for uploader in uploaders:
send_cookie_email(uploader,
record_information,
message=uploader_message,
version=_rev_hepsubmission.version)
return jsonify({'success': True, 'version': _rev_hepsubmission.version})
else:
return jsonify({"message": f"Rec id {recid} is not finished so cannot create a new version"}), 400
def add_participant(recid):
"""
Adds a participant to a record.
:param recid:
:return:
"""
try:
submission_record = get_latest_hepsubmission(publication_recid=recid)
full_name = request.form['name']
email = request.form['email']
participant_type = request.form['type']
new_record = SubmissionParticipant(publication_recid=recid,
full_name=full_name,
email=email,
role=participant_type)
submission_record.participants.append(new_record)
db.session.commit()
return json.dumps({
"success":
True,
"recid":
recid,
"message":
"{0} {1} added.".format(full_name, participant_type)
})
except Exception as e:
return json.dumps({
"success": False,
"recid": recid,
"message": 'Unable to add participant.'
})
def get_resources(recid, version):
"""
Gets a list of resources for a publication, relevant to all data records
:param recid:
:return: json
"""
result = OrderedDict()
result["Common Resources"] = {"type": "submission", "version": version, "id": recid, "resources": []}
submission = get_latest_hepsubmission(publication_recid=recid, version=version)
if submission:
for reference in submission.resources:
result["Common Resources"]["resources"].append(process_reference(reference))
datasubmissions = DataSubmission.query.filter_by(publication_recid=recid, version=version).all()
for datasubmission in datasubmissions:
result[datasubmission.name] = {
"type": "data",
"id": datasubmission.id,
"resources": [],
"version": datasubmission.version,
}
for reference in datasubmission.resources:
result[datasubmission.name]["resources"].append(process_reference(reference))
return json.dumps(result)
def promote_or_demote_participant(recid, action, demote_or_promote,
participant_id):
"""
Can promote or demote a participant to/from primary reviewer/uploader.
:param recid: record id that the user will be promoted or demoted for
:param action: upload or review
:param demote_or_promote: demote or promote
:param participant_id: id of user from the SubmissionParticipant table.
:return:
"""
try:
participant = SubmissionParticipant.query.filter_by(
id=participant_id).one()
status = 'reserve'
if demote_or_promote == 'promote':
status = 'primary'
participant.status = status
db.session.add(participant)
db.session.commit()
record = get_record_by_id(recid)
# now send the email telling the user of their new status!
if status == 'primary':
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
send_cookie_email(participant,
record,
version=hepsubmission.version)
return json.dumps({"success": True, "recid": recid})
except Exception as e:
return json.dumps({"success": False, "recid": recid, "error": str(e)})
def download_submission_with_inspire_id(*args, **kwargs):
"""
Gets the submission file and either serves it back directly from YAML, or converts it
for other formats.
:param inspire_id: inspire id
:param version: version of submission to export. If absent, returns the latest.
:param file_format: yaml, csv, root, or yoda
:return:
"""
inspire_id = kwargs.pop('inspire_id')
if 'ins' in inspire_id:
inspire_id = inspire_id.replace('ins', '')
if 'version' in kwargs:
submission = HEPSubmission.query.filter_by(
inspire_id=inspire_id, version=kwargs.pop('version')).first()
else:
submission = get_latest_hepsubmission(inspire_id=inspire_id)
if not submission:
return display_error(
title="No submission found",
description="A submission with Inspire ID {0} does not exist".
format(inspire_id))
return download_submission(submission, kwargs.pop('file_format'))
def unload_submission(record_id, version=1):
submission = get_latest_hepsubmission(publication_recid=record_id)
if not submission:
print('Record {0} not found'.format(record_id))
return
if version == submission.version:
print('Unloading record {0} version {1}...'.format(record_id, version))
remove_submission(record_id, version)
else:
print('Not unloading record {0} version {1} (latest version {2})...'.format(record_id, version, submission.version))
return
if version == 1:
data_records = get_records_matching_field("related_publication", record_id)
for record in data_records["hits"]["hits"]:
print("\t Removed data table {0} from index".format(record["_id"]))
try:
delete_item_from_index(doc_type=CFG_DATA_TYPE, id=record["_id"], parent=record_id)
except Exception as e:
logging.error("Unable to remove {0} from index. {1}".format(record["_id"], e))
try:
delete_item_from_index(doc_type=CFG_PUB_TYPE, id=record_id)
print("Removed publication {0} from index".format(record_id))
except NotFoundError as nfe:
print(nfe)
print('Finished unloading record {0} version {1}.'.format(record_id, version))
def get_all_review_messages(publication_recid):
"""
Gets the review messages for a publication id.
:param publication_recid:
:return:
"""
messages = OrderedDict()
latest_submission = get_latest_hepsubmission(publication_recid=publication_recid)
datareview_query = DataReview.query.filter_by(
publication_recid=publication_recid, version=latest_submission.version).order_by(
DataReview.id.asc())
if datareview_query.count() > 0:
reviews = datareview_query.all()
for data_review in reviews:
data_submission_query = DataSubmission.query.filter_by(
id=data_review.data_recid)
data_submission_record = data_submission_query.one()
if data_review.data_recid not in messages:
if data_submission_query.count() > 0:
messages[data_submission_record.name] = []
query_messages_for_data_review(data_review, messages[
data_submission_record.name])
return json.dumps(messages, default=default_time)
def delete_submission(recid):
"""
Submissions can only be removed if they are not finalised,
meaning they should never be in the index.
Only delete the latest version of a submission.
Delete indexed information only if version = 1.
:param recid:
:return:
"""
if has_role(current_user, 'admin') or has_role(current_user, 'coordinator') \
or check_is_sandbox_record(recid):
submission = get_latest_hepsubmission(publication_recid=recid)
unload_submission(recid, submission.version)
if submission.version == 1:
admin_idx = AdminIndexer()
admin_idx.find_and_delete('recid', recid)
return json.dumps({"success": True,
"recid": recid,
"errors": [
"Record successfully removed!"]})
else:
return json.dumps(
{"success": False, "recid": recid,
"errors": [
"You do not have permission to delete this submission. "
"Only coordinators can do that."]})
def render_record(recid, record, version, output_format, light_mode=False):
if user_allowed_to_perform_action(recid):
version_count = HEPSubmission.query.filter_by(
publication_recid=recid).count()
else:
version_count = HEPSubmission.query.filter_by(
publication_recid=recid, overall_status='finished').count()
if version == -1:
version = version_count
hepdata_submission = get_latest_hepsubmission(publication_recid=recid, version=version)
if hepdata_submission is not None:
ctx = format_submission(recid, record, version, version_count, hepdata_submission)
increment(recid)
if output_format == "json":
ctx = process_ctx(ctx, light_mode)
return jsonify(ctx)
else:
return render_template('hepdata_records/publication_record.html',
ctx=ctx)
else: # this happens when we access an id of a data record
# in which case, we find the related publication, and
# make the front end focus on the relevant data table.
try:
publication_recid = int(record['related_publication'])
publication_record = get_record_contents(publication_recid)
hepdata_submission = get_latest_hepsubmission(recid=publication_recid)
ctx = format_submission(publication_recid, publication_record,
hepdata_submission.version, 1, hepdata_submission,
data_table=record['title'])
ctx['related_publication_id'] = publication_recid
ctx['table_name'] = record['title']
if output_format == "json":
ctx = process_ctx(ctx, light_mode)
return jsonify(ctx)
else:
return render_template('hepdata_records/data_record.html', ctx=ctx)
except Exception:
return render_template('hepdata_theme/404.html')
def create_record_for_dashboard(record_id,
submissions,
current_user,
coordinator=None,
user_role=None,
status="todo"):
if user_role is None:
user_role = ["coordinator"]
publication_record = get_record_by_id(int(record_id))
if publication_record is not None:
if record_id not in submissions:
hepdata_submission_record = get_latest_hepsubmission(
publication_recid=record_id)
submissions[record_id] = {}
submissions[record_id]["metadata"] = {
"recid": record_id,
"role": user_role,
"start_date": publication_record.created
}
submissions[record_id]["metadata"][
"versions"] = hepdata_submission_record.version
submissions[record_id]["status"] = status
submissions[record_id]["stats"] = {
"passed": 0,
"attention": 0,
"todo": 0
}
if coordinator:
submissions[record_id]["metadata"]["coordinator"] = {
'id': coordinator.id,
'name': coordinator.email,
'email': coordinator.email
}
submissions[record_id]["metadata"]["show_coord_view"] = int(
current_user.get_id()) == coordinator.id
else:
submissions[record_id]["metadata"]["coordinator"] = {
'name': 'No coordinator'
}
if "title" in publication_record:
submissions[record_id]["metadata"]["title"] = \
publication_record['title']
if "inspire_id" not in publication_record or publication_record[
"inspire_id"] is None:
submissions[record_id]["metadata"][
"requires_inspire_id"] = True
else:
# if it is, it's because the user has two roles for that
# submission. So we should show them!
if user_role not in submissions[record_id]["metadata"]["role"]:
submissions[record_id]["metadata"]["role"].append(user_role)
def update_analyses():
endpoints = current_app.config["ANALYSES_ENDPOINTS"]
for analysis_endpoint in endpoints:
if "endpoint_url" in endpoints[analysis_endpoint]:
log.info("Updating analyses from {0}...".format(analysis_endpoint))
response = requests.get(
endpoints[analysis_endpoint]["endpoint_url"])
if response:
analyses = response.json()
for record in analyses:
submission = get_latest_hepsubmission(
inspire_id=record, overall_status='finished')
if submission:
num_new_resources = 0
for analysis in analyses[record]:
_resource_url = endpoints[analysis_endpoint][
"url_template"].format(analysis)
if not is_resource_added_to_submission(
submission.publication_recid,
submission.version, _resource_url):
print(
'Adding {} analysis to ins{} with URL {}'.
format(analysis_endpoint, record,
_resource_url))
new_resource = DataResource(
file_location=_resource_url,
file_type=analysis_endpoint)
submission.resources.append(new_resource)
num_new_resources += 1
if num_new_resources:
try:
db.session.add(submission)
db.session.commit()
index_record_ids(
[submission.publication_recid])
except Exception as e:
db.session.rollback()
log.error(e)
else:
log.debug(
"An analysis is available in {0} but with no equivalent in HEPData (ins{1})."
.format(analysis_endpoint, record))
else:
log.debug(
"No endpoint url configured for {0}".format(analysis_endpoint))
def test_update_record_info(app):
"""Test update of publication information from INSPIRE."""
assert update_record_info(
None) == 'Inspire ID is None' # case where Inspire ID is None
for inspire_id in (
'1311487',
'19999999'): # check both a valid and invalid Inspire ID
assert update_record_info(
inspire_id
) == 'No HEPData submission' # before creation of HEPSubmission object
submission = process_submission_payload(inspire_id=inspire_id,
submitter_id=1,
reviewer={
'name': 'Reviewer',
'email':
'*****@*****.**'
},
uploader={
'name': 'Uploader',
'email':
'*****@*****.**'
},
send_upload_email=False)
# Process the files to create DataSubmission tables in the DB.
base_dir = os.path.dirname(os.path.realpath(__file__))
directory = os.path.join(base_dir, 'test_data/test_submission')
tmp_path = os.path.join(tempfile.mkdtemp(dir=CFG_TMPDIR),
'test_submission')
shutil.copytree(directory, tmp_path)
process_submission_directory(tmp_path,
os.path.join(tmp_path, 'submission.yaml'),
submission.publication_recid)
do_finalise(submission.publication_recid,
force_finalise=True,
convert=False)
if inspire_id == '19999999':
assert update_record_info(inspire_id) == 'Invalid Inspire ID'
else:
# First change the publication information to that of a different record.
different_inspire_record_information, status = get_inspire_record_information(
'1650066')
assert status == 'success'
hep_submission = get_latest_hepsubmission(inspire_id=inspire_id)
assert hep_submission is not None
update_record(hep_submission.publication_recid,
different_inspire_record_information)
# Then can check that the update works and that a further update is not required.
assert update_record_info(inspire_id, send_email=True) == 'Success'
assert update_record_info(
inspire_id
) == 'No update needed' # check case where information already current
unload_submission(submission.publication_recid)
def send_email(inspireids):
"""Send finalised email announcing that records have been added or revised (in case it wasn't done automatically)."""
processed_inspireids = parse_inspireids_from_string(inspireids)
for inspireid in processed_inspireids:
_cleaned_id = inspireid.replace("ins", "")
submission = get_latest_hepsubmission(inspire_id=_cleaned_id, overall_status='finished')
if submission:
send_finalised_email(submission)
else:
print("No records found for Inspire ID {}".format(inspireid))
def process_payload(recid, file, redirect_url, synchronous=False):
"""Process an uploaded file
:param recid: int
The id of the record to update
:param file: file
The file to process
:param redirect_url: string
Redirect URL to record, for use if the upload fails or in synchronous mode
:param synchronous: bool
Whether to process asynchronously via celery (default) or immediately (only recommended for tests)
:return: JSONResponse either containing 'url' (for success cases) or
'message' (for error cases, which will give a 400 error).
"""
if file and (allowed_file(file.filename)):
file_path = save_zip_file(file, recid)
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
if hepsubmission.overall_status == 'finished':
# If it is finished and we receive an update,
# then we need to reopen the submission to allow for revisions,
# by creating 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
previous_status = hepsubmission.overall_status
hepsubmission.overall_status = 'sandbox_processing' if previous_status == 'sandbox' else 'processing'
db.session.add(hepsubmission)
db.session.commit()
if synchronous:
process_saved_file(file_path, recid, current_user.get_id(),
redirect_url, previous_status)
else:
process_saved_file.delay(file_path, recid, current_user.get_id(),
redirect_url, previous_status)
flash(
'File saved. You will receive an email when the file has been processed.',
'info')
return jsonify({'url': redirect_url.format(recid)})
else:
return jsonify({
"message":
"You must upload a .zip, .tar, .tar.gz or .tgz file" +
" (or a .oldhepdata or single .yaml or .yaml.gz file)."
}), 400
def download_submission_with_inspire_id(*args, **kwargs):
"""
Gets the submission file and either serves it back directly from YAML, or converts it
for other formats. Routes:\n
``/submission/<inspire_id>/<file_format>``\n
``/submission/<inspire_id>/<int:version>/<file_format>``\n
``/submission/<inspire_id>/<int:version>/<file_format>/<rivet>``
:param inspire_id: inspire id
:param version: version of submission to export. If absent, returns the latest.
:param file_format: yaml, csv, root, or yoda
:param rivet: Rivet analysis name to override default written in YODA export
:return: download_submission
"""
inspire_id = kwargs.pop('inspire_id')
if 'ins' in inspire_id:
inspire_id = inspire_id.replace('ins', '')
submission = get_latest_hepsubmission(inspire_id=inspire_id)
if not submission:
return display_error(
title="No submission found",
description="A submission with Inspire ID {0} does not exist".
format(inspire_id))
recid = submission.publication_recid
version_count, version_count_all = get_version_count(recid)
if 'version' in kwargs:
version = kwargs.pop('version')
else:
# If version not given explicitly, take to be latest allowed version (or 1 if there are no allowed versions).
version = version_count if version_count else 1
if version_count < version_count_all and version == version_count_all:
# Check for a user trying to access a version of a publication record where they don't have permissions.
abort(403)
elif version < version_count_all:
submission = HEPSubmission.query.filter_by(inspire_id=inspire_id,
version=version).first()
if not submission:
return display_error(
title="No submission found",
description=
"A submission with Inspire ID {0} and version {1} does not exist".
format(inspire_id, version))
return download_submission(submission,
kwargs.pop('file_format'),
rivet_analysis_name=kwargs.pop('rivet', ''))
def has_upload_permissions(recid, user, is_sandbox=False):
if has_role(user, 'admin'):
return True
if is_sandbox:
hepsubmission_record = get_latest_hepsubmission(publication_recid=recid, overall_status='sandbox')
return hepsubmission_record is not None and hepsubmission_record.coordinator == user.id
participant = SubmissionParticipant.query.filter_by(user_account=user.id,
role='uploader', publication_recid=recid, status='primary').first()
if participant:
return True
def test_create_new_version(app, load_default_data, identifiers, mocker):
hepsubmission = get_latest_hepsubmission(publication_recid=1)
assert hepsubmission.version == 1
# Add an uploader
uploader = SubmissionParticipant(publication_recid=1,
email='*****@*****.**',
role='uploader',
status='primary')
db.session.add(uploader)
db.session.commit()
user = User.query.first()
# Mock `send_cookie_email` method
send_cookie_mock = mocker.patch(
'hepdata.modules.records.api.send_cookie_email')
# Create new version of valid finished record
result = create_new_version(1, user, uploader_message="Hello!")
assert result.status_code == 200
assert result.json == {'success': True, 'version': 2}
# get_latest_hepsubmission should now return version 2
hepsubmission = get_latest_hepsubmission(publication_recid=1)
assert hepsubmission.version == 2
assert hepsubmission.overall_status == 'todo'
# Should have attempted to send uploader email
send_cookie_mock.assert_called_with(uploader,
get_record_by_id(1),
message="Hello!",
version=2)
# Try creating a new version - should not work as status of most recent is 'todo'
result, status_code = create_new_version(1, user)
assert status_code == 400
assert result.json == {
'message': 'Rec id 1 is not finished so cannot create a new version'
}
def send_tweet(inspireids):
"""Send tweet announcing that records have been added or revised (in case it wasn't done automatically)."""
processed_inspireids = parse_inspireids_from_string(inspireids)
for inspireid in processed_inspireids:
_cleaned_id = inspireid.replace("ins", "")
submission = get_latest_hepsubmission(inspire_id=_cleaned_id, overall_status='finished')
if submission:
record = get_record_by_id(submission.publication_recid)
site_url = current_app.config.get('SITE_URL', 'https://www.hepdata.net')
url = site_url + '/record/ins{0}'.format(record['inspire_id'])
tweet(record['title'], record['collaborations'], url, record['version'])
else:
print("No records found for Inspire ID {}".format(inspireid))
def test_tables(app, live_server, env_browser):
"""E2E test to tables in a record."""
browser = env_browser
# Import record with non-default table names
import_default_data(app, [{'hepdata_id': 'ins1206352'}])
try:
browser.get(flask.url_for('hepdata_theme.index', _external=True))
assert (flask.url_for('hepdata_theme.index', _external=True)
in browser.current_url)
latest_item = browser.find_element_by_css_selector(
'.latest-record .title')
actions = ActionChains(browser)
actions.move_to_element(latest_item).perform()
latest_item.click()
# Check current table name
assert (browser.find_element_by_id('table_name').text ==
'Figure 8 panel (a)')
# Check switching tables works as expected
new_table = browser.find_elements_by_css_selector(
'#table-list li h4')[2]
assert (new_table.text == "Figure 8 panel (c)")
new_table.click()
_check_table_links(browser, "Figure 8 panel (c)")
# Get link to table from table page
table_link = browser.find_element_by_css_selector('#data_link_container button') \
.get_attribute('data-clipboard-text')
assert (table_link.endswith('table=Figure%208%20panel%20(c)'))
_check_table_links(browser, "Figure 8 panel (c)", url=table_link)
# Check a link to a table name with spaces removed
short_table_link = table_link.replace('%20', '')
_check_table_links(browser, "Figure 8 panel (c)", url=short_table_link)
# Check a link to an invalid table
invalid_table_link = table_link.replace('Figure%208%20panel%20(c)',
'NotARealTable')
_check_table_links(browser,
"Figure 8 panel (a)",
url=invalid_table_link)
finally:
# Delete record and reindex so added record doesn't affect other tests
submission = get_latest_hepsubmission(inspire_id='1206352')
unload_submission(submission.publication_recid)
reindex_all(recreate=True)
def convert_and_store(inspire_id, file_format, force):
"""
Converts a submission to a given file format, and stores
on the file system to be retrieved later by users
:param inspire_id:
:param file_format:
:param force:
:return:
"""
submission = get_latest_hepsubmission(inspire_id=inspire_id)
if submission:
print("Creating {0} conversion for ins{1}".format(file_format, inspire_id))
download_submission(submission, file_format, offline=True, force=force)
else:
print("Unable to find a matching submission for {0}".format(inspire_id))
def create_record_for_dashboard(record_id, submissions, primary_uploader=None,
primary_reviewer=None, coordinator=None,
user_role=None,
status="todo"):
if user_role is None:
user_role = ["coordinator"]
publication_record = get_record_by_id(int(record_id))
if publication_record is not None:
if record_id not in submissions:
hepdata_submission_record = get_latest_hepsubmission(publication_recid=record_id)
submissions[record_id] = {}
submissions[record_id]["metadata"] = {"recid": record_id,
"role": user_role,
"start_date": publication_record.created}
submissions[record_id]["metadata"][
"versions"] = hepdata_submission_record.version
submissions[record_id]["status"] = status
submissions[record_id]["stats"] = {"passed": 0, "attention": 0,
"todo": 0}
if coordinator:
submissions[record_id]["metadata"]["coordinator"] = {
'id': coordinator.id, 'name': coordinator.email,
'email': coordinator.email}
submissions[record_id]["metadata"][
"show_coord_view"] = int(current_user.get_id()) == coordinator.id
else:
submissions[record_id]["metadata"]["coordinator"] = {
'name': 'No coordinator'}
if "title" in publication_record:
submissions[record_id]["metadata"]["title"] = \
publication_record['title']
if "inspire_id" not in publication_record or publication_record["inspire_id"] is None:
submissions[record_id]["metadata"][
"requires_inspire_id"] = True
else:
# if it is, it's because the user has two roles for that
# submission. So we should show them!
if user_role not in submissions[record_id]["metadata"]["role"]:
submissions[record_id]["metadata"]["role"].append(user_role)
def convert_and_store(inspire_id, file_format, force):
"""
Converts a submission to a given file format, and stores
on the file system to be retrieved later by users
:param inspire_id:
:param file_format:
:param force:
:return:
"""
submission = get_latest_hepsubmission(inspire_id=inspire_id)
if submission:
print("Creating {0} conversion for ins{1}".format(
file_format, inspire_id))
download_submission(submission, file_format, offline=True, force=force)
else:
print(
"Unable to find a matching submission for {0}".format(inspire_id))
def add_analyses(doc):
"""
TODO: Generalise for badges other than Rivet
:param doc:
:return:
"""
latest_submission = get_latest_hepsubmission(
publication_recid=doc['recid'])
if latest_submission:
doc["analyses"] = []
for reference in latest_submission.resources:
if reference.file_type in current_app.config['ANALYSES_ENDPOINTS']:
doc["analyses"].append({
'type': reference.file_type,
'analysis': reference.file_location
})
def send_tweet(inspireids):
"""
Send tweet announcing that records have been added or revised (in case it wasn't done automatically).
:param inspireids: list of record IDs to tweet
"""
processed_inspireids = parse_inspireids_from_string(inspireids)
for inspireid in processed_inspireids:
_cleaned_id = inspireid.replace("ins", "")
submission = get_latest_hepsubmission(inspire_id=_cleaned_id)
if submission:
record = get_record_by_id(submission.publication_recid)
url = "http://www.hepdata.net/record/ins{0}".format(
record['inspire_id'])
tweet(record['title'], record['collaborations'], url,
record['version'])
else:
print("No records found for Inspire ID {}".format(inspireid))
def process_general_submission_info(basepath, submission_info_document, recid):
"""
Processes the top level information about a submission,
extracting the information about the data abstract,
additional resources for the submission (files, links,
and html inserts) and historical modification information.
:param basepath: the path the submission has been loaded to
:param submission_info_document: the data document
:param recid:
:return:
"""
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
if "comment" in submission_info_document:
hepsubmission.data_abstract = encode_string(
submission_info_document['comment'])
if "dateupdated" in submission_info_document:
try:
hepsubmission.last_updated = parse(
submission_info_document['dateupdated'], dayfirst=True)
except ValueError:
hepsubmission.last_updated = datetime.now()
else:
hepsubmission.last_updated = datetime.now()
if "modifications" in submission_info_document:
parse_modifications(hepsubmission, recid, submission_info_document)
if 'additional_resources' in submission_info_document:
for reference in hepsubmission.resources:
db.session.delete(reference)
resources = parse_additional_resources(basepath, recid,
submission_info_document)
for resource in resources:
hepsubmission.resources.append(resource)
db.session.add(hepsubmission)
db.session.commit()
def add_analyses(doc):
"""
TODO: Generalise for other badges other than rivit
:param doc:
:return:
"""
# do lookup from http://rivet.hepforge.org/list_of_analyses.json
# for HEPforge. But only one lookup.
# look up once per day, and cache the result in REDIS.
latest_submission = get_latest_hepsubmission(publication_recid=doc['recid'])
if latest_submission:
doc["analyses"] = []
for reference in latest_submission.resources:
if reference.file_type in current_app.config['ANALYSES_ENDPOINTS']:
doc["analyses"].append({'type': reference.file_type, 'analysis': reference.file_location})
def delete_all_files(rec_id, check_old_data_paths=True):
"""
Deletes all data files across ALL versions of a record.
"""
record_data_paths = [get_data_path_for_record(rec_id)]
if check_old_data_paths:
record_data_paths.append(get_old_data_path_for_record(rec_id))
hepsubmission = get_latest_hepsubmission(publication_recid=rec_id)
if hepsubmission and hepsubmission.inspire_id is not None:
record_data_paths.append(
get_old_data_path_for_record('ins%s' %
hepsubmission.inspire_id))
for record_data_path in record_data_paths:
log.debug("Scanning directory: %s" % record_data_path)
if os.path.isdir(record_data_path):
log.debug("Removing %s" % record_data_path)
shutil.rmtree(record_data_path)
def get_coordinator_view(recid):
# there should only ever be one rev
hepsubmission_record = get_latest_hepsubmission(publication_recid=recid)
participants = {"reviewer": {"reserve": [], "primary": []}, "uploader": {"reserve": [], "primary": []}}
for participant in hepsubmission_record.participants:
participants[participant.role][participant.status].append(
{"full_name": participant.full_name, "email": participant.email, "id": participant.id}
)
return json.dumps(
{
"recid": recid,
"primary-reviewers": participants["reviewer"]["primary"],
"reserve-reviewers": participants["reviewer"]["reserve"],
"primary-uploaders": participants["uploader"]["primary"],
"reserve-uploaders": participants["uploader"]["reserve"],
}
)
def send_coordinator_notification_email(recid, version, user, message=None):
"""
:param recid:
:param user: user object
:param message: message to send
:return:
"""
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
coordinator = get_user_from_id(hepsubmission.coordinator)
if not coordinator:
raise NoParticipantsException()
site_url = current_app.config.get('SITE_URL', 'https://www.hepdata.net')
record = get_record_by_id(recid)
name = coordinator.email
coordinator_profile = UserProfile.get_by_userid(hepsubmission.coordinator)
if coordinator_profile:
name = coordinator_profile.full_name
collaboration = _get_collaboration(hepsubmission.coordinator)
message_body = render_template('hepdata_theme/email/passed_review.html',
name=name,
actor=user.email,
collaboration=collaboration,
article=recid,
version=version,
message=message,
title=record['title'],
site_url=site_url,
link=site_url + "/record/{0}".format(recid),
dashboard_link=site_url + "/dashboard")
create_send_email_task(
coordinator.email,
'[HEPData] Submission {0} is ready to be finalised'.format(recid),
message_body)
def get_pending_invitations_for_user(user):
pending_invites = SubmissionParticipant.query.filter(
SubmissionParticipant.email == user.email,
or_(SubmissionParticipant.role == 'reviewer',
SubmissionParticipant.role == 'uploader'),
SubmissionParticipant.user_account == None
).all()
result = []
for invite in pending_invites:
publication_record = get_record_by_id(invite.publication_recid)
hepsubmission = get_latest_hepsubmission(publication_recid=invite.publication_recid)
coordinator = get_user_from_id(hepsubmission.coordinator)
result.append(
{'title': encode_string(publication_record['title'], 'utf-8'),
'invitation_cookie': invite.invitation_cookie,
'role': invite.role, 'coordinator': coordinator})
return result
def process_general_submission_info(basepath, submission_info_document, recid):
"""
Processes the top level information about a submission,
extracting the information about the data abstract,
additional resources for the submission (files, links,
and html inserts) and historical modification information.
:param submission_info_document: the data document
:param recid:
:return:
"""
if 'comment' in submission_info_document \
or 'modifications' in submission_info_document \
or 'record_ids' in submission_info_document:
hepsubmission = get_latest_hepsubmission(publication_recid=recid)
hepsubmission.data_abstract = encode_string(
submission_info_document['comment'])
if "dateupdated" in submission_info_document:
try:
hepsubmission.last_updated = parse(submission_info_document['dateupdated'], dayfirst=True)
except ValueError as ve:
hepsubmission.last_updated = datetime.now()
if "modifications" in submission_info_document:
parse_modifications(hepsubmission, recid, submission_info_document)
if 'additional_resources' in submission_info_document:
for reference in hepsubmission.resources:
db.session.delete(reference)
resources = parse_additional_resources(basepath,
recid, hepsubmission.version, submission_info_document)
for resource in resources:
hepsubmission.resources.append(resource)
db.session.add(hepsubmission)
db.session.commit()
def update_analyses():
endpoints = current_app.config['ANALYSES_ENDPOINTS']
for analysis_endpoint in endpoints:
if 'endpoint_url' in endpoints[analysis_endpoint]:
log.info('Updating analyses from {0}...'.format(analysis_endpoint))
response = requests.get(endpoints[analysis_endpoint]['endpoint_url'])
if response:
analyses = response.json()
for record in analyses:
submission = get_latest_hepsubmission(inspire_id=record)
if submission:
for analysis in analyses[record]:
_resource_url = endpoints[analysis_endpoint]['url_template'].format(analysis)
if not is_resource_added_to_submission(submission.publication_recid, submission.version,
_resource_url):
new_resource = DataResource(
file_location=_resource_url,
file_type=analysis_endpoint)
submission.resources.append(new_resource)
try:
db.session.add(submission)
db.session.commit()
except Exception as e:
db.session.rollback()
log.error(e)
else:
log.debug('An analysis is available in {0} but with no equivalent in HEPData (ins{1}).'.format(analysis_endpoint, record))
else:
log.debug('No endpoint url configured for {0}'.format(analysis_endpoint))
def test_load_file(app, migrator):
with app.app_context():
success = migrator.load_file('ins1487726')
assert (success)
hepsubmission = get_latest_hepsubmission(inspire_id='1487726')
assert (hepsubmission is not None)
assert (hepsubmission.inspire_id == '1487726')
index_records = get_records_matching_field('inspire_id',
hepsubmission.inspire_id,
doc_type='publication')
assert (len(index_records['hits']['hits']) == 1)
publication_record = get_record_contents(
hepsubmission.publication_recid)
assert (publication_record is not None)
data_submissions = DataSubmission.query.filter_by(
publication_recid=hepsubmission.publication_recid).count()
assert (data_submissions == 5)
def download_submission_with_recid(*args, **kwargs):
"""
Gets the submission file and either serves it back directly from YAML, or converts it
for other formats.
:param recid: submissions recid
:param version: version of submission to export. If -1, returns the latest.
:param file_format: yaml, csv, ROOT, or YODA
:return:
"""
recid = kwargs.pop('recid')
if 'version' in kwargs:
submission = HEPSubmission.query.filter_by(publication_recid=recid, version=kwargs.pop('version')) \
.first()
else:
submission = get_latest_hepsubmission(publication_recid=recid)
if not submission:
return display_error(
title="No submission found",
description="A submission with record id {0} does not exist".format(kwargs.pop('recid'))
)
return download_submission(submission, kwargs.pop('file_format'))
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
