Fix: skip vector store document id filter for unrestricted chat users (#13937)

* Fix: skip vector store document id filter for unrestricted chat users

ChatStreamingView built an IN filter from every permitted document id
for the "chat over all documents" case, which exceeds the vector
store's SQLite bound-parameter safety limit on installs with more
than ~32700 documents, silently returning no context. For a user who
can see every document (an active superuser), that filter never
narrows anything, so skip it and let the retriever search the whole
index instead.

* Minor improvements from a Claude review

* When a user is unrestricted chatting, still exclude trashed documents using a 'NOT IN' SQL statement.  Wire that up where we need it

* Update src/paperless_ai/chat.py

Co-authored-by: shamoon <4887959+shamoon@users.noreply.github.com>
This commit is contained in:
Trenton H
2026-09-02 18:08:54 +00:00
committed by GitHub
co-authored by shamoon
parent 5d6ea11828
commit 351892bbab
8 changed files with 327 additions and 56 deletions
+4
View File
@@ -180,6 +180,7 @@ from documents.permissions import has_system_status_permission
from documents.permissions import permitted_document_ids
from documents.permissions import permitted_object_ids
from documents.permissions import set_permissions_for_object
from documents.permissions import user_is_unrestricted
from documents.plugins.date_parsing import get_date_parser
from documents.schema import generate_object_with_permissions_schema
from documents.search import SearchHit
@@ -2329,10 +2330,12 @@ class ChatStreamingView(GenericAPIView[Any]):
return HttpResponseForbidden("Insufficient permissions")
documents = Document.objects.filter(pk=document.pk)
unrestricted = False
else:
documents = Document.objects.filter(
id__in=permitted_document_ids(request.user),
)
unrestricted = user_is_unrestricted(request.user)
output_language = get_llm_output_language(
ai_config=ai_config,
@@ -2343,6 +2346,7 @@ class ChatStreamingView(GenericAPIView[Any]):
stream_chat_with_documents(
query_str=question,
documents=documents,
unrestricted=unrestricted,
output_language=output_language,
),
content_type="text/event-stream",