def create(self, validated_data):
node = validated_data.pop('node', {})
if isinstance(node, dict):
node = Node.objects.create(creator=self.context['request'].user,
**node)
if node.is_deleted:
raise exceptions.ValidationError(
'Cannot create a preprint from a deleted node.')
auth = get_user_auth(self.context['request'])
if not node.has_permission(auth.user, permissions.ADMIN):
raise exceptions.PermissionDenied
provider = validated_data.pop('provider', None)
if not provider:
raise exceptions.ValidationError(
detail='You must specify a valid provider to create a preprint.'
)
node_preprints = node.preprints.filter(provider=provider)
if node_preprints.exists():
raise Conflict(
'Only one preprint per provider can be submitted for a node. Check `meta[existing_resource_id]`.',
meta={'existing_resource_id': node_preprints.first()._id})
preprint = PreprintService(node=node, provider=provider)
preprint.save()
preprint.node._has_abandoned_preprint = True
preprint.node.save()
return self.update(preprint, validated_data)
def create(self, validated_data):
node = validated_data.pop('node', {})
if isinstance(node, dict):
node = Node.objects.create(creator=self.context['request'].user, **node)
if node.is_deleted:
raise exceptions.ValidationError('Cannot create a preprint from a deleted node.')
auth = get_user_auth(self.context['request'])
if not node.has_permission(auth.user, permissions.ADMIN):
raise exceptions.PermissionDenied
provider = validated_data.pop('provider', None)
if not provider:
raise exceptions.ValidationError(detail='You must specify a valid provider to create a preprint.')
node_preprints = node.preprints.filter(provider=provider)
if node_preprints.exists():
raise Conflict('Only one preprint per provider can be submitted for a node. Check `meta[existing_resource_id]`.', meta={'existing_resource_id': node_preprints.first()._id})
preprint = PreprintService(node=node, provider=provider)
preprint.save()
preprint.node._has_abandoned_preprint = True
preprint.node.save()
return self.update(preprint, validated_data)
def create(self, validated_data):
node = validated_data.pop('node', None)
if not node:
raise exceptions.NotFound('Unable to find Node with specified id.')
elif node.is_deleted:
raise exceptions.ValidationError('Cannot create a preprint from a deleted node.')
auth = get_user_auth(self.context['request'])
if not node.has_permission(auth.user, permissions.ADMIN):
raise exceptions.PermissionDenied
primary_file = validated_data.pop('primary_file', None)
if not primary_file:
raise exceptions.ValidationError(detail='You must specify a valid primary_file to create a preprint.')
provider = validated_data.pop('provider', None)
if not provider:
raise exceptions.ValidationError(detail='You must specify a valid provider to create a preprint.')
if PreprintService.find(Q('node', 'eq', node) & Q('provider', 'eq', provider)).count():
conflict = PreprintService.find_one(Q('node', 'eq', node) & Q('provider', 'eq', provider))
raise Conflict('Only one preprint per provider can be submitted for a node. Check `meta[existing_resource_id]`.', meta={'existing_resource_id': conflict._id})
preprint = PreprintService(node=node, provider=provider)
self.set_field(preprint.set_primary_file, primary_file, auth, save=True)
preprint.node._has_abandoned_preprint = True
preprint.node.save()
return self.update(preprint, validated_data)
def on_preprint_updated(preprint_id, update_share=True):
# WARNING: Only perform Read-Only operations in an asynchronous task, until Repeatable Read/Serializable
# transactions are implemented in View and Task application layers.
from osf.models import PreprintService
preprint = PreprintService.load(preprint_id)
if preprint.node:
status = 'public' if preprint.node.is_public else 'unavailable'
try:
update_ezid_metadata_on_change(preprint, status=status)
except HTTPError as err:
sentry.log_exception()
sentry.log_message(err.args[0])
if settings.SHARE_URL and update_share:
if not preprint.provider.access_token:
raise ValueError('No access_token for {}. Unable to send {} to SHARE.'.format(preprint.provider, preprint))
resp = requests.post('{}api/v2/normalizeddata/'.format(settings.SHARE_URL), json={
'data': {
'type': 'NormalizedData',
'attributes': {
'tasks': [],
'raw': None,
'data': {'@graph': format_preprint(preprint)}
}
}
}, headers={'Authorization': 'Bearer {}'.format(preprint.provider.access_token), 'Content-Type': 'application/vnd.api+json'})
logger.debug(resp.content)
resp.raise_for_status()
def on_preprint_updated(preprint_id):
# WARNING: Only perform Read-Only operations in an asynchronous task, until Repeatable Read/Serializable
# transactions are implemented in View and Task application layers.
from osf.models import PreprintService
preprint = PreprintService.load(preprint_id)
if settings.SHARE_URL:
if not preprint.provider.access_token:
raise ValueError(
'No access_token for {}. Unable to send {} to SHARE.'.format(
preprint.provider, preprint))
resp = requests.post(
'{}api/v2/normalizeddata/'.format(settings.SHARE_URL),
json={
'data': {
'type': 'NormalizedData',
'attributes': {
'tasks': [],
'raw': None,
'data': {
'@graph': format_preprint(preprint)
}
}
}
},
headers={
'Authorization':
'Bearer {}'.format(preprint.provider.access_token),
'Content-Type': 'application/vnd.api+json'
})
logger.debug(resp.content)
resp.raise_for_status()
def _async_update_preprint_share(self, preprint_id, old_subjects, share_type):
# Any modifications to this function may need to change _update_preprint_share
# Takes preprint_id to ensure async retries push fresh data
PreprintService = apps.get_model('osf.PreprintService')
preprint = PreprintService.load(preprint_id)
data = serialize_share_preprint_data(preprint, share_type, old_subjects)
resp = send_share_preprint_data(preprint, data)
try:
resp = requests.post('{}api/v2/normalizeddata/'.format(settings.SHARE_URL), json={
'data': {
'type': 'NormalizedData',
'attributes': {
'tasks': [],
'raw': None,
'data': {'@graph': format_preprint(preprint, share_type, old_subjects)}
}
}
}, headers={'Authorization': 'Bearer {}'.format(preprint.provider.access_token), 'Content-Type': 'application/vnd.api+json'})
logger.debug(resp.content)
resp.raise_for_status()
except Exception as e:
if resp.status_code >= 500:
if self.request.retries == self.max_retries:
send_desk_share_preprint_error(preprint, resp, self.request.retries)
raise self.retry(
exc=e,
countdown=(random.random() + 1) * min(60 + settings.CELERY_RETRY_BACKOFF_BASE ** self.request.retries, 60 * 10)
)
else:
send_desk_share_preprint_error(preprint, resp, self.request.retries)
def _async_update_preprint_share(self, preprint_id, old_subjects, share_type):
# Any modifications to this function may need to change _update_preprint_share
# Takes preprint_id to ensure async retries push fresh data
PreprintService = apps.get_model('osf.PreprintService')
preprint = PreprintService.load(preprint_id)
data = serialize_share_preprint_data(preprint, share_type, old_subjects)
resp = send_share_preprint_data(preprint, data)
try:
resp = requests.post('{}api/v2/normalizeddata/'.format(settings.SHARE_URL), json={
'data': {
'type': 'NormalizedData',
'attributes': {
'tasks': [],
'raw': None,
'data': {'@graph': format_preprint(preprint, share_type, old_subjects)}
}
}
}, headers={'Authorization': 'Bearer {}'.format(preprint.provider.access_token), 'Content-Type': 'application/vnd.api+json'})
logger.debug(resp.content)
resp.raise_for_status()
except Exception as e:
if resp.status_code >= 500:
if self.request.retries == self.max_retries:
send_desk_share_preprint_error(preprint, resp, self.request.retries)
raise self.retry(
exc=e,
countdown=(random.random() + 1) * min(60 + settings.CELERY_RETRY_BACKOFF_BASE ** self.request.retries, 60 * 10)
)
else:
send_desk_share_preprint_error(preprint, resp, self.request.retries)
def get_and_set_preprint_identifiers(preprint_id):
from osf.models import PreprintService
preprint = PreprintService.load(preprint_id)
ezid_response = request_identifiers_from_ezid(preprint)
id_dict = parse_identifiers(ezid_response)
preprint.set_identifier_values(doi=id_dict['doi'], ark=id_dict['ark'])
def get_and_set_preprint_identifiers(preprint_id):
PreprintService = apps.get_model('osf.PreprintService')
preprint = PreprintService.load(preprint_id)
ezid_response = request_identifiers_from_ezid(preprint)
if ezid_response is None:
return
id_dict = parse_identifiers(ezid_response)
preprint.set_identifier_values(doi=id_dict['doi'], save=True)
def get_and_set_preprint_identifiers(preprint_id):
PreprintService = apps.get_model('osf.PreprintService')
preprint = PreprintService.load(preprint_id)
ezid_response = request_identifiers_from_ezid(preprint)
if ezid_response is None:
return
id_dict = parse_identifiers(ezid_response)
preprint.set_identifier_values(doi=id_dict['doi'], ark=id_dict['ark'], save=True)
def get_preprint_provider(self, obj):
preprint_id = obj.get('preprint', None)
if preprint_id:
preprint = PreprintService.load(preprint_id)
if preprint:
provider = preprint.provider
return {'url': provider.external_url, 'name': provider.name}
return None
def get_preprint_provider(self, obj):
preprint_id = obj.get('preprint', None)
if preprint_id:
preprint = PreprintService.load(preprint_id)
if preprint:
provider = preprint.provider
return {'url': provider.external_url, 'name': provider.name}
return None
def post(self, request):
crossref_email_content = lxml.etree.fromstring(str(request.POST['body-plain']))
status = crossref_email_content.get('status').lower() # from element doi_batch_diagnostic
record_count = int(crossref_email_content.find('batch_data/record_count').text)
records = crossref_email_content.xpath('//record_diagnostic')
dois_processed = 0
if status == 'completed':
guids = []
# Keep track of errors recieved, ignore those that are handled
unexpected_errors = False
for record in records:
doi = getattr(record.find('doi'), 'text', None)
guid = doi.split('/')[-1] if doi else None
guids.append(guid)
preprint = PreprintService.load(guid) if guid else None
if record.get('status').lower() == 'success' and doi:
msg = record.find('msg').text
created = bool(msg == 'Successfully added')
legacy_doi = preprint.get_identifier(category='legacy_doi')
if created or legacy_doi:
# Sets preprint_doi_created and saves the preprint
preprint.set_identifier_values(doi=doi, save=True)
# Double records returned when possible matching DOI is found in crossref
elif 'possible preprint/vor pair' not in msg.lower():
# Directly updates the identifier
preprint.set_identifier_value(category='doi', value=doi)
dois_processed += 1
# Mark legacy DOIs overwritten by newly batch confirmed crossref DOIs
if legacy_doi:
legacy_doi.remove()
elif record.get('status').lower() == 'failure':
if 'Relation target DOI does not exist' in record.find('msg').text:
logger.warn('Related publication DOI does not exist, sending metadata again without it...')
client = preprint.get_doi_client()
client.create_identifier(preprint, category='doi', include_relation=False)
# This error occurs when a single preprint is being updated several times in a row with the same metadata [#PLAT-944]
elif 'less or equal to previously submitted version' in record.find('msg').text and record_count == 2:
break
else:
unexpected_errors = True
logger.info('Creation success email received from CrossRef for preprints: {}'.format(guids))
if dois_processed != record_count or status != 'completed':
if unexpected_errors:
batch_id = crossref_email_content.find('batch_id').text
mails.send_mail(
to_addr=settings.OSF_SUPPORT_EMAIL,
mail=mails.CROSSREF_ERROR,
batch_id=batch_id,
email_content=request.POST['body-plain'],
)
logger.error('Error submitting metadata for batch_id {} with CrossRef, email sent to help desk'.format(batch_id))
return HttpResponse('Mail received', status=200)
def on_preprint_updated(preprint_id, update_share=True, share_type=None, old_subjects=None, saved_fields=None):
# WARNING: Only perform Read-Only operations in an asynchronous task, until Repeatable Read/Serializable
# transactions are implemented in View and Task application layers.
from osf.models import PreprintService
preprint = PreprintService.load(preprint_id)
if old_subjects is None:
old_subjects = []
if should_update_preprint_identifiers(preprint, old_subjects, saved_fields):
update_or_create_preprint_identifiers(preprint)
if update_share:
update_preprint_share(preprint, old_subjects, share_type)
def create(self, validated_data):
node = validated_data.pop('node', None)
if not node:
raise exceptions.NotFound('Unable to find Node with specified id.')
elif node.is_deleted:
raise exceptions.ValidationError(
'Cannot create a preprint from a deleted node.')
auth = get_user_auth(self.context['request'])
if not node.has_permission(auth.user, permissions.ADMIN):
raise exceptions.PermissionDenied
primary_file = validated_data.pop('primary_file', None)
if not primary_file:
raise exceptions.ValidationError(
detail=
'You must specify a valid primary_file to create a preprint.')
provider = validated_data.pop('provider', None)
if not provider:
raise exceptions.ValidationError(
detail='You must specify a valid provider to create a preprint.'
)
if PreprintService.find(
Q('node', 'eq', node) & Q('provider', 'eq', provider)).count():
conflict = PreprintService.find_one(
Q('node', 'eq', node) & Q('provider', 'eq', provider))
raise Conflict(
'Only one preprint per provider can be submitted for a node. Check `meta[existing_resource_id]`.',
meta={'existing_resource_id': conflict._id})
preprint = PreprintService(node=node, provider=provider)
self.set_field(preprint.set_primary_file,
primary_file,
auth,
save=True)
preprint.node._has_abandoned_preprint = True
preprint.node.save()
return self.update(preprint, validated_data)
def on_preprint_updated(preprint_id,
update_share=True,
share_type=None,
old_subjects=None):
# WARNING: Only perform Read-Only operations in an asynchronous task, until Repeatable Read/Serializable
# transactions are implemented in View and Task application layers.
from osf.models import PreprintService
preprint = PreprintService.load(preprint_id)
if old_subjects is None:
old_subjects = []
if preprint.node:
update_or_create_preprint_identifiers(preprint)
if update_share:
update_preprint_share(preprint, old_subjects, share_type)
def on_preprint_updated(preprint_id, update_share=True, share_type=None, old_subjects=None):
# WARNING: Only perform Read-Only operations in an asynchronous task, until Repeatable Read/Serializable
# transactions are implemented in View and Task application layers.
from osf.models import PreprintService
preprint = PreprintService.load(preprint_id)
if old_subjects is None:
old_subjects = []
if preprint.node:
status = 'public' if preprint.verified_publishable else 'unavailable'
try:
update_ezid_metadata_on_change(preprint._id, status=status)
except HTTPError as err:
sentry.log_exception()
sentry.log_message(err.args[0])
if update_share:
update_preprint_share(preprint, old_subjects, share_type)
def test_create_preprint_adds_log_if_published(self, mock_get_identifiers):
public_project_payload = build_preprint_create_payload(
self.public_project._id, self.provider._id,
self.file_one_public_project._id, {
'is_published': True,
'subjects': [[SubjectFactory()._id]],
})
res = self.app.post_json_api(self.url,
public_project_payload,
auth=self.user.auth)
assert_equal(res.status_code, 201)
preprint_id = res.json['data']['id']
preprint = PreprintService.load(preprint_id)
log = preprint.node.logs.latest()
assert_equal(log.action, 'preprint_initiated')
assert_equal(log.params.get('preprint'), preprint_id)
def on_preprint_updated(preprint_id, update_share=True, share_type=None, old_subjects=None):
# WARNING: Only perform Read-Only operations in an asynchronous task, until Repeatable Read/Serializable
# transactions are implemented in View and Task application layers.
from osf.models import PreprintService
preprint = PreprintService.load(preprint_id)
if old_subjects is None:
old_subjects = []
if preprint.node:
status = 'public' if preprint.verified_publishable else 'unavailable'
try:
update_ezid_metadata_on_change(preprint._id, status=status)
except HTTPError as err:
sentry.log_exception()
sentry.log_message(err.args[0])
if update_share:
update_preprint_share(preprint, old_subjects, share_type)
def test_create_preprint_adds_log_if_published(self):
public_project_payload = build_preprint_create_payload(
self.public_project._id,
self.provider._id,
self.file_one_public_project._id,
{
'is_published': True,
'subjects': [[SubjectFactory()._id]],
}
)
res = self.app.post_json_api(self.url, public_project_payload, auth=self.user.auth)
assert_equal(res.status_code, 201)
preprint_id = res.json['data']['id']
preprint = PreprintService.load(preprint_id)
log = preprint.node.logs.latest()
assert_equal(log.action, 'preprint_initiated')
assert_equal(log.params.get('preprint'), preprint_id)
def _async_update_preprint_share(self, preprint_id, old_subjects, share_type):
# Any modifications to this function may need to change _update_preprint_share
# Takes preprint_id to ensure async retries push fresh data
PreprintService = apps.get_model('osf.PreprintService')
preprint = PreprintService.load(preprint_id)
data = serialize_share_preprint_data(preprint, share_type, old_subjects)
resp = send_share_preprint_data(preprint, data)
try:
resp.raise_for_status()
except Exception as e:
if resp.status_code >= 500:
if self.request.retries == self.max_retries:
send_desk_share_preprint_error(preprint, resp, self.request.retries)
raise self.retry(
exc=e,
countdown=(random.random() + 1) * min(60 + settings.CELERY_RETRY_BACKOFF_BASE ** self.request.retries, 60 * 10)
)
else:
send_desk_share_preprint_error(preprint, resp, self.request.retries)
def _async_update_preprint_share(self, preprint_id, old_subjects, share_type):
# Any modifications to this function may need to change _update_preprint_share
# Takes preprint_id to ensure async retries push fresh data
PreprintService = apps.get_model('osf.PreprintService')
preprint = PreprintService.load(preprint_id)
data = serialize_share_preprint_data(preprint, share_type, old_subjects)
resp = send_share_preprint_data(preprint, data)
try:
resp.raise_for_status()
except Exception as e:
if resp.status_code >= 500:
if self.request.retries == self.max_retries:
send_desk_share_preprint_error(preprint, resp,
self.request.retries)
raise self.retry(exc=e,
countdown=(random.random() + 1) * min(
60 + settings.CELERY_RETRY_BACKOFF_BASE**
self.request.retries, 60 * 10))
else:
send_desk_share_preprint_error(preprint, resp,
self.request.retries)
def perform_destroy(self, instance):
if instance.is_published:
raise Conflict('Published preprints cannot be deleted.')
PreprintService.delete(instance)
def perform_destroy(self, instance):
if instance.is_published:
raise Conflict('Published preprints cannot be deleted.')
PreprintService.delete(instance)
