paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2026-01-30 23:08:59 -06:00
Code Issues Packages Projects Releases Wiki Activity

Refactor document and custom field count queries

Browse Source
This commit is contained in:
shamoon
2026-01-27 11:49:25 -08:00
parent 4d16d1c812
commit 13eac46f1c
1 changed files with 22 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files

68
src/documents/management/commands/document_perf_benchmark.py
View File

@@ -16,6 +16,7 @@ from documents.models import CustomFieldInstance
from documents.models import Document from documents.models import Document
from documents.models import Tag from documents.models import Tag
from documents.permissions import get_objects_for_user_owner_aware from documents.permissions import get_objects_for_user_owner_aware
from documents.permissions import permitted_document_ids
class Command(BaseCommand): class Command(BaseCommand):
@@ -530,23 +531,15 @@ class Command(BaseCommand):
return return
self._time_query( self._time_query(
label="tag document_count (values_list)", label="tag document_count (grouped)",
iterations=iterations, iterations=iterations,
fn=lambda: list( fn=lambda: list(
self._tag_queryset( Tag.documents.through.objects.filter(
prefix=prefix, document_id__in=Subquery(permitted_document_ids(user)),
filter_q=self._document_filter(user, use_subquery=False), )
).values_list("id", "document_count"), .values("tag_id")
), .annotate(c=Count("document_id"))
) .values_list("tag_id", "c"),
self._time_query(
label="tag document_count (Subquery)",
iterations=iterations,
fn=lambda: list(
self._tag_queryset(
prefix=prefix,
filter_q=self._document_filter(user, use_subquery=True),
).values_list("id", "document_count"),
), ),
) )
@@ -561,48 +554,31 @@ class Command(BaseCommand):
if not CustomField.objects.filter(name__startswith=prefix).exists(): if not CustomField.objects.filter(name__startswith=prefix).exists():
return return
def _filter(for_user, *, use_subquery: bool): permitted = Subquery(permitted_document_ids(user))
if for_user.is_superuser: super_permitted = CustomFieldInstance.objects.filter(
return Q(fields__document__deleted_at__isnull=True) document__deleted_at__isnull=True,
).values_list("document_id")
qs = get_objects_for_user_owner_aware( def _run(ids_subquery):
for_user,
"documents.view_document",
Document,
)
ids = (
Subquery(qs.values_list("id"))
if use_subquery
else qs.values_list("id", flat=True)
)
return Q(
fields__document__deleted_at__isnull=True,
fields__document__id__in=ids,
)
def _run(filter_q):
return list( return list(
CustomField.objects.filter(name__startswith=prefix) CustomFieldInstance.objects.filter(
.annotate( document_id__in=ids_subquery,
document_count=Count("fields", filter=filter_q), field__name__startswith=prefix,
) )
.values_list("id", "document_count"), .values("field_id")
.annotate(c=Count("document_id"))
.values_list("field_id", "c"),
) )
self._time_query( self._time_query(
label="custom fields document_count (values_list)", label="custom fields document_count (grouped permitted)",
iterations=iterations, iterations=iterations,
fn=lambda: _run(_filter(user, use_subquery=False)), fn=lambda: _run(permitted),
)
self._time_query(
label="custom fields document_count (Subquery)",
iterations=iterations,
fn=lambda: _run(_filter(user, use_subquery=True)),
) )
self._time_query( self._time_query(
label="custom fields document_count superuser baseline", label="custom fields document_count superuser baseline",
iterations=iterations, iterations=iterations,
fn=lambda: _run(_filter(superuser, use_subquery=False)), fn=lambda: _run(super_permitted),
) )
def _time_query(self, *, label: str, iterations: int, fn): def _time_query(self, *, label: str, iterations: int, fn):