update_share from update_search

api/share/utils.py

@@ -4,13 +4,13 @@
 """
 from django.apps import apps
 import random
-import requests
 from framework.celery_tasks import app as celery_app
+from framework.celery_tasks.handlers import enqueue_task
 from framework.sentry import log_exception
 
 from website import settings
 from celery.exceptions import Retry
-from osf.metadata.tools import pls_update_trove_indexcard, pls_delete_trove_indexcard
+from osf.metadata.tools import pls_send_trove_indexcard, pls_delete_trove_indexcard
 
 def is_qa_resource(resource):
     """
@@ -28,36 +28,31 @@ def is_qa_resource(resource):
 
 
 def update_share(resource):
-    if _should_delete_indexcard(resource):
-        resp = pls_delete_trove_indexcard(resource)
-    else:
-        resp = pls_update_trove_indexcard(resource)
-    status_code = resp.status_code
-    try:
-        resp.raise_for_status()
-    except requests.HTTPError:
-        if status_code >= 500:
-            async_update_resource_share.delay(resource._id)
-        else:
-            log_exception()
+    if not settings.SHARE_ENABLED:
+        return
+    enqueue_task(task__update_share.s(resource._id))
 
 
 @celery_app.task(bind=True, max_retries=4, acks_late=True)
-def async_update_resource_share(self, guid, **kwargs):
+def task__update_share(self, guid: str, **kwargs):
     """
     This function updates share  takes Preprints, Projects and Registrations.
     :param self:
     :param guid:
     :return:
     """
     resource = apps.get_model('osf.Guid').load(guid).referent
-    resp = pls_update_trove_indexcard(resource)
+    resp = (
+        pls_delete_trove_indexcard(resource)
+        if _should_delete_indexcard(resource)
+        else pls_send_trove_indexcard(resource)
+    )
     try:
         resp.raise_for_status()
     except Exception as e:
         if self.request.retries == self.max_retries:
             log_exception()
-        elif resp.status_code >= 500:
+        elif resp.status_code >= 500 and settings.USE_CELERY:
             try:
                 self.retry(
                     exc=e,

osf/management/commands/recatalog_metadata.py

@@ -80,6 +80,11 @@ def add_arguments(self, parser):
             action='store_true',
             help='recatalog metadata for non-registration projects (and components)',
         )
+        # type_group.add_argument(
+        #     '--files',
+        #     action='store_true',
+        #     help='recatalog metadata for files',
+        # )
 
         parser.add_argument(
             '--start-id',

osf/management/commands/reindex_quickfiles.py

@@ -6,7 +6,6 @@
 
 PAGE_SIZE = 100
 from tqdm import tqdm
-from api.share.utils import update_share
 
 def paginated_progressbar(queryset, page_size, function):
     paginator = Paginator(queryset, page_size)
@@ -32,7 +31,6 @@ def reindex_quickfiles():
     paginated_progressbar(files_to_reindex, PAGE_SIZE, update_file)
 
     for node in nodes:
-        update_share(node)
         node.update_search()
 
 

osf/metadata/tools.py

@@ -46,7 +46,7 @@ def pls_gather_metadata_file(osf_item, format_key, serializer_config=None) -> Se
     )
 
 
-def pls_update_trove_indexcard(osf_item):
+def pls_send_trove_indexcard(osf_item):
     _iri = osf_iri(osf_item)
     if not _iri:
         raise ValueError(f'could not get iri for {osf_item}')

osf/models/files.py

@@ -23,6 +23,7 @@
 from osf.utils.datetime_aware_jsonfield import DateTimeAwareJSONField
 from osf.utils.fields import NonNaiveDateTimeField
 from api.base.utils import waterbutler_api_url_for
+from api.share.utils import update_share
 from website.files import utils
 from website.files.exceptions import VersionNotFoundError
 from website.util import api_v2_url, web_url_for, api_url_for
@@ -443,6 +444,7 @@ def delete(self, user=None, save=True, deleted_on=None):
         return self
 
     def update_search(self):
+        update_share(self)
         from website import search
         try:
             search.search.update_file(self)

osf/models/institution.py

@@ -188,7 +188,7 @@ def banner_path(self):
             return '/static/img/institutions/banners/placeholder-banner.png'
 
     def update_search(self):
-        from website.search.search import update_institution, update_node
+        from website.search.search import update_institution
         from website.search.exceptions import SearchUnavailableError
 
         try:
@@ -197,10 +197,7 @@ def update_search(self):
             logger.exception(e)
 
         for node in self.nodes.filter(is_deleted=False):
-            try:
-                update_node(node, async_update=False)
-            except SearchUnavailableError as e:
-                logger.exception(e)
+            node.update_search()
 
     def save(self, *args, **kwargs):
         saved_fields = self.get_dirty_fields()

osf/models/node.py

@@ -710,6 +710,8 @@ def should_request_identifiers(self):
 
     @classmethod
     def bulk_update_search(cls, nodes, index=None):
+        for _node in nodes:
+            update_share(_node)
         from website import search
         try:
             serialize = functools.partial(search.search.update_node, index=index, bulk=True, async_update=False)
@@ -719,8 +721,8 @@ def bulk_update_search(cls, nodes, index=None):
             log_exception()
 
     def update_search(self):
+        update_share(self)
         from website import search
-
         try:
             search.search.update_node(self, bulk=False, async_update=True)
             if self.collection_submissions.exists() and self.is_public:
@@ -1038,8 +1040,6 @@ def add_tag_log(self, tag, auth):
     # Override Taggable
     def on_tag_added(self, tag):
         self.update_search()
-        if settings.SHARE_ENABLED:
-            update_share(self)
 
     def remove_tag(self, tag, auth, save=True):
         if not tag:
@@ -1062,9 +1062,6 @@ def remove_tag(self, tag, auth, save=True):
             if save:
                 self.save()
             self.update_search()
-            if settings.SHARE_ENABLED:
-                update_share(self)
-
             return True
 
     def remove_tags(self, tags, auth, save=True):
@@ -1074,9 +1071,6 @@ def remove_tags(self, tags, auth, save=True):
         """
         super(AbstractNode, self).remove_tags(tags, auth, save)
         self.update_search()
-        if settings.SHARE_ENABLED:
-            update_share(self)
-
         return True
 
     def set_visible(self, user, visible, log=True, auth=None, save=False):

osf/models/preprint.py

@@ -52,6 +52,7 @@
     TagNotFoundError
 )
 from django.contrib.postgres.fields import ArrayField
+from api.share.utils import update_share
 
 logger = logging.getLogger(__name__)
 
@@ -901,6 +902,8 @@ def set_contributor_order(self, contributor_ids):
 
     @classmethod
     def bulk_update_search(cls, preprints, index=None):
+        for _preprint in preprints:
+            update_share(_preprint)
         from website import search
         try:
             serialize = functools.partial(search.search.update_preprint, index=index, bulk=True, async_update=False)
@@ -910,6 +913,7 @@ def bulk_update_search(cls, preprints, index=None):
             log_exception()
 
     def update_search(self):
+        update_share(self)
         from website import search
         try:
             search.search.update_preprint(self, bulk=False, async_update=True)

osf/models/registrations.py

@@ -14,7 +14,6 @@
 )
 from dirtyfields import DirtyFieldsMixin
 
-from api.share.utils import update_share
 from framework.auth import Auth
 from framework.exceptions import PermissionsError
 from osf.utils.fields import NonNaiveDateTimeField
@@ -825,17 +824,9 @@ def withdraw(self):
         # Pass auth=None because the registration initiator may not be
         # an admin on components (component admins had the opportunity
         # to disapprove the retraction by this point)
-        for node in self.get_descendants_recursive(primary_only=True):
+        for node in self.node_and_primary_descendants():
             node.set_privacy('public', auth=None, log=False)
-            node.update_search()
-
-        # force a save before sending data to share or retraction will not be updated
-        self.set_privacy('public', auth=None, log=False)
-        self.update_search()
-        self.save()
-
-        if settings.SHARE_ENABLED:
-            update_share(self)
+        AbstractNode.bulk_update_search(self.node_and_primary_descendants())
 
     def copy_registration_responses_into_schema_response(self, draft_registration=None, save=True):
         """Copies registration metadata into schema responses"""
@@ -862,8 +853,7 @@ def on_schema_response_completed(self):
             archive_to_ia(children)
 
     def related_resource_updated(self, log_action=None, api_request=None, **log_params):
-        if settings.SHARE_ENABLED:
-            update_share(self)
+        self.update_search()
         if not log_action:
             return
 

osf/models/user.py

@@ -35,6 +35,7 @@
                                        MergeConflictError)
 from framework.exceptions import PermissionsError
 from framework.sessions.utils import remove_sessions_for_user
+from api.share.utils import update_share
 from osf.utils.requests import get_current_request
 from osf.exceptions import reraise_django_validation_errors, UserStateError
 from osf.models.base import BaseModel, GuidMixin, GuidMixinQuerySet
@@ -1444,6 +1445,7 @@ def is_assumed_ham(self):
         return user_has_trusted_email
 
     def update_search(self):
+        update_share(self)
         from website.search.search import update_user
         update_user(self)
 

scripts/migration/migrate_share_preprint_data.py

@@ -31,7 +31,7 @@ def migrate(dry=True):
         logger.info('{}/{} - {}'.format(count, target_count, preprint_id))
         try:
             if not dry:
-                on_preprint_updated(preprint_id, update_share=True)
+                on_preprint_updated(preprint_id)
                 # Sleep in order to be nice to EZID
                 time.sleep(1)
         except Exception as e:

website/archiver/tasks.py

@@ -313,5 +313,4 @@ def archive_success(dst_pk, job_pk):
         job.save()
         dst.sanction.ask(dst.get_active_contributors_recursive(unique_users=True))
 
-    if settings.SHARE_ENABLED:
-        update_share(dst)
+    update_share(dst)

website/preprints/tasks.py

@@ -5,8 +5,6 @@
 from framework.celery_tasks import app as celery_app
 from framework.postcommit_tasks.handlers import enqueue_postcommit_task, get_task_from_postcommit_queue
 
-from website import settings
-from api.share.utils import update_share
 
 logger = logging.getLogger(__name__)
 
@@ -27,9 +25,6 @@ def on_preprint_updated(preprint_id, old_subjects=None, saved_fields=None):
     if should_update_preprint_identifiers(preprint, old_subjects, saved_fields):
         update_or_create_preprint_identifiers(preprint)
 
-    if settings.SHARE_ENABLED:
-        update_share(preprint)
-
 
 def should_update_preprint_identifiers(preprint, old_subjects, saved_fields):
     # Only update identifier metadata iff...