Skip to content

A community-supported supercharged document management system: scan, index and archive all your documents

Fix: guard build_document_node against stale FK on deleted correspondent/doc type (#13318)

Trenton Hea1f3653a9b9
diff --git a/src/paperless_ai/indexing.py b/src/paperless_ai/indexing.py
index 4230edb8d..16aeda776 100644
--- a/src/paperless_ai/indexing.py
+++ b/src/paperless_ai/indexing.py
@@ -5,6 +5,7 @@ from datetime import timedelta
from typing import TYPE_CHECKING
from django.conf import settings
+from django.core.exceptions import ObjectDoesNotExist
from django.utils import timezone
from filelock import FileLock
from filelock import ReadWriteLock
@@ -167,6 +168,19 @@ def write_store(embed_model_name: str | None = None):
yield store
+def _safe_related_name(document: Document, field: str) -> str | None:
+ """
+ Returns the ``name`` of a related object (correspondent, document_type,
+ storage_path), or None if the FK is unset or points at a row that has
+ since been deleted (e.g. concurrently with this call).
+ """
+ try:
+ related = getattr(document, field)
+ except ObjectDoesNotExist:
+ return None
+ return related.name if related else None
+
+
def build_document_node(
document: Document,
*,
@@ -180,14 +194,10 @@ def build_document_node(
"document_id": str(document.id),
"title": document.title,
"tags": [t.name for t in document.tags.all()],
- "correspondent": document.correspondent.name
- if document.correspondent
- else None,
- "document_type": document.document_type.name
- if document.document_type
- else None,
+ "correspondent": _safe_related_name(document, "correspondent"),
+ "document_type": _safe_related_name(document, "document_type"),
"filename": document.filename,
- "storage_path": document.storage_path.name if document.storage_path else None,
+ "storage_path": _safe_related_name(document, "storage_path"),
"archive_serial_number": document.archive_serial_number,
"created": document.created.isoformat() if document.created else None,
"added": document.added.isoformat() if document.added else None,
diff --git a/src/paperless_ai/tests/test_ai_indexing.py b/src/paperless_ai/tests/test_ai_indexing.py
index c73959895..e69834810 100644
--- a/src/paperless_ai/tests/test_ai_indexing.py
+++ b/src/paperless_ai/tests/test_ai_indexing.py
@@ -8,7 +8,9 @@ from django.test import override_settings
from django.utils import timezone
from llama_index.core.schema import MetadataMode
+from documents.models import Correspondent
from documents.models import Document
+from documents.models import DocumentType
from documents.models import PaperlessTask
from documents.signals import document_consumption_finished
from documents.signals import document_updated
@@ -95,6 +97,38 @@ def test_build_document_node_structured_fields_in_metadata(
assert "modified" in node.metadata
+def test_build_document_node_survives_concurrently_deleted_correspondent(
+ real_document: Document,
+) -> None:
+ """Regression test for #13314.
+
+ If a document's correspondent (or document type) is deleted after the
+ in-memory Document instance was loaded but before build_document_node
+ resolves the relation, accessing the FK must not raise -- it should
+ behave like an unset FK and produce None in the metadata instead of
+ aborting the whole indexing pass.
+ """
+ correspondent = Correspondent.objects.create(name="Stale Correspondent")
+ document_type = DocumentType.objects.create(name="Stale Type")
+ real_document.correspondent = correspondent
+ real_document.document_type = document_type
+ real_document.save()
+
+ # Re-fetch to get an instance whose correspondent/document_type relations
+ # are unresolved (not yet cached), mirroring a task that loaded the
+ # document before the concurrent deletion below.
+ stale_document = Document.objects.get(pk=real_document.pk)
+
+ correspondent.delete()
+ document_type.delete()
+
+ nodes = indexing.build_document_node(stale_document)
+ assert len(nodes) > 0
+ assert nodes[0].metadata["correspondent"] is None
+ assert nodes[0].metadata["document_type"] is None
+
+
@pytest.mark.django_db
def test_build_document_node_excludes_document_id_from_llm_context(
real_document: Document,