Bump version to 3.3.0

.github/workflows/ci-backend.yml

@@ -80,7 +80,7 @@ jobs:
     needs: changes
     if: needs.changes.outputs.backend_changed == 'true'
     name: "Python ${{ matrix.python-version }}"
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
     strategy:
@@ -114,7 +114,7 @@ jobs:
           packages: unpaper tesseract-ocr imagemagick ghostscript poppler-utils
       - name: Configure ImageMagick
         run: |
-          sudo cp docker/rootfs/etc/ImageMagick-6/paperless-policy.xml /etc/ImageMagick-7/policy.xml
+          sudo cp docker/rootfs/etc/ImageMagick-6/paperless-policy.xml /etc/ImageMagick-6/policy.xml
       - name: Install Python dependencies
         env:
           PYTHON_VERSION: ${{ steps.setup-python.outputs.python-version }}
@@ -158,7 +158,7 @@ jobs:
     needs: changes
     if: needs.changes.outputs.backend_changed == 'true'
     name: Check project typing
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
     env:

.github/workflows/ci-docker.yml

@@ -24,10 +24,10 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - runner: ubuntu-26.04
+          - runner: ubuntu-24.04
             arch: amd64
             platform: linux/amd64
-          - runner: ubuntu-26.04-arm
+          - runner: ubuntu-24.04-arm
             arch: arm64
             platform: linux/arm64
     runs-on: ${{ matrix.runner }}
@@ -163,7 +163,7 @@ jobs:
           archive: false
   merge-and-push:
     name: Merge and Push Manifest
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     needs: build-arch
     if: needs.build-arch.outputs.should-push == 'true'
     environment: image-publishing

.github/workflows/ci-docs.yml

@@ -65,7 +65,7 @@ jobs:
     needs: changes
     if: needs.changes.outputs.docs_changed == 'true'
     name: Build Documentation
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     steps:
       - uses: actions/configure-pages@45bfe0192ca1faeb007ade9deae92b16b8254a0d # v6.0.0
       - name: Checkout
@@ -102,7 +102,7 @@ jobs:
     name: Deploy Documentation
     needs: [changes, build]
     if: github.event_name == 'push' && github.ref == 'refs/heads/main' && needs.changes.outputs.docs_changed == 'true'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       pages: write
       id-token: write

.github/workflows/ci-frontend.yml

@@ -72,7 +72,7 @@ jobs:
     needs: changes
     if: needs.changes.outputs.frontend_changed == 'true'
     name: Install Dependencies
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
     steps:
@@ -104,7 +104,7 @@ jobs:
     name: Lint
     needs: [changes, install-dependencies]
     if: needs.changes.outputs.frontend_changed == 'true'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
     steps:
@@ -137,7 +137,7 @@ jobs:
     name: "Unit Tests (${{ matrix.shard-index }}/${{ matrix.shard-count }})"
     needs: [changes, install-dependencies]
     if: needs.changes.outputs.frontend_changed == 'true'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
     strategy:
@@ -188,10 +188,10 @@ jobs:
     name: E2E Tests
     needs: [changes, install-dependencies]
     if: needs.changes.outputs.frontend_changed == 'true'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
-    container: mcr.microsoft.com/playwright:v1.62.1-resolute
+    container: mcr.microsoft.com/playwright:v1.62.1-noble
     env:
       PLAYWRIGHT_BROWSERS_PATH: /ms-playwright
       PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD: 1
@@ -246,7 +246,7 @@ jobs:
     name: Frontend Build
     needs: [changes, unit-tests, e2e-tests]
     if: needs.changes.outputs.frontend_changed == 'true'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
     steps:

.github/workflows/ci-release.yml

@@ -14,7 +14,7 @@ permissions: {}
 jobs:
   wait-for-docker:
     name: Wait for Docker Build
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       checks: read
       statuses: read
@@ -30,7 +30,7 @@ jobs:
   build-release:
     name: Build Release
     needs: wait-for-docker
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
     steps:
@@ -73,7 +73,7 @@ jobs:
         timeout-minutes: 12
         uses: $/.github/actions/apt-install
         with:
-          packages: gettext libleptonica6
+          packages: gettext liblept5
       # ---- Build Documentation ----
       - name: Build documentation
         env:
@@ -145,7 +145,7 @@ jobs:
   publish-release:
     name: Publish Release
     needs: build-release
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: write
       pull-requests: write
@@ -197,7 +197,7 @@ jobs:
     name: Append Changelog
     needs: publish-release
     if: needs.publish-release.outputs.prerelease == 'false'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: write
       pull-requests: write

.github/workflows/ci-static-analysis.yml

@@ -15,7 +15,7 @@ permissions:
 jobs:
   zizmor:
     name: Run zizmor
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       contents: read
       actions: read
@@ -29,7 +29,7 @@ jobs:
         uses: zizmorcore/zizmor-action@cc914d7f3750a2d13d75c7f184a1060aa0e9d482 # v0.6.4
   semgrep:
     name: Semgrep CE
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     container:
       image: semgrep/semgrep:1.155.0@sha256:cc869c685dcc0fe497c86258da9f205397d8108e56d21a86082ea4886e52784d
     if: github.actor != 'dependabot[bot]'

.github/workflows/cleanup-tags.yml

@@ -17,7 +17,7 @@ jobs:
   cleanup-images:
     name: Cleanup Image Tags for ${{ matrix.primary-name }}
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     environment: registry-maintenance
     strategy:
       fail-fast: false
@@ -42,7 +42,7 @@ jobs:
   cleanup-untagged-images:
     name: Cleanup Untagged Images Tags for ${{ matrix.primary-name }}
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     needs:
       - cleanup-images
     environment: registry-maintenance

.github/workflows/codeql-analysis.yml

@@ -21,7 +21,7 @@ on:
 jobs:
   analyze:
     name: Analyze
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       actions: read
       contents: read

.github/workflows/crowdin.yml

@@ -13,7 +13,7 @@ jobs:
   synchronize-with-crowdin:
     name: Crowdin Sync
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     environment: translation-sync
     steps:
       - name: Checkout

.github/workflows/issue-bot.yml

@@ -7,7 +7,7 @@ jobs:
     # Note: peakoss/anti-slop does not support the `issues` event yet (all of its
     # issue inputs are still commented out upstream), so the checks that the PR Bot
     # workflow gets from the action are implemented manually here.
-    runs-on: ubuntu-slim
+    runs-on: ubuntu-latest
     permissions:
       issues: write
     steps:

.github/workflows/pr-bot.yml

@@ -4,7 +4,7 @@ on:
     types: [opened]
 jobs:
   Anti-slop:
-    runs-on: ubuntu-slim
+    runs-on: ubuntu-latest
     permissions:
       contents: read
       issues: read
@@ -24,7 +24,7 @@ jobs:
             ASLOP-PR-VERIFY
   pr-bot:
     name: Automated PR Bot
-    runs-on: ubuntu-slim
+    runs-on: ubuntu-latest
     # Runs after Anti-slop so the welcome comment can see whether the PR was closed
     # instead of racing it. Still runs if that job fails, so labeling is not lost.
     needs: Anti-slop

.github/workflows/project-actions.yml

@@ -12,7 +12,7 @@ permissions:
 jobs:
   pr_opened_or_reopened:
     name: pr_opened_or_reopened
-    runs-on: ubuntu-slim
+    runs-on: ubuntu-24.04
     permissions:
       # write permission is required for autolabeler
       pull-requests: write

.github/workflows/repo-maintenance.yml

@@ -9,7 +9,7 @@ jobs:
   stale:
     name: 'Stale'
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       issues: write
       pull-requests: write
@@ -34,7 +34,7 @@ jobs:
   lock-threads:
     name: 'Lock Old Threads'
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-24.04
     permissions:
       issues: write
       pull-requests: write
@@ -58,7 +58,7 @@ jobs:
   close-answered-discussions:
     name: 'Close Answered Discussions'
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-slim
+    runs-on: ubuntu-24.04
     permissions:
       discussions: write
     steps:
@@ -117,7 +117,7 @@ jobs:
   close-outdated-discussions:
     name: 'Close Outdated Discussions'
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-slim
+    runs-on: ubuntu-24.04
     permissions:
       discussions: write
     steps:
@@ -211,7 +211,7 @@ jobs:
   close-unsupported-feature-requests:
     name: 'Close Unsupported Feature Requests'
     if: github.repository_owner == 'paperless-ngx'
-    runs-on: ubuntu-slim
+    runs-on: ubuntu-24.04
     permissions:
       discussions: write
     steps:

.github/workflows/translate-strings.yml

@@ -8,7 +8,7 @@ env:
 jobs:
   generate-translate-strings:
     name: Generate Translation Strings
-    runs-on: ubuntu-26.04
+    runs-on: ubuntu-latest
     environment: translation-sync
     permissions:
       contents: write

docs/faq.md

@@ -76,7 +76,9 @@ is not supported by any of the available parsers.
 
 **A:** Not by default. As of v3, a file whose contents match an existing document is still
 consumed, and the duplicate is flagged in the UI — open the document and check the
-**Duplicates** tab to review documents that share the same content. If you prefer the old
+**Duplicates** tab to review documents that share the same content, or filter the document
+list by **Duplicates** to find all of them (see
+[Duplicate documents](usage.md#duplicate-documents)). If you prefer the old
 behavior of rejecting duplicates during consumption, set
 [`PAPERLESS_CONSUMER_DELETE_DUPLICATES`](configuration.md#PAPERLESS_CONSUMER_DELETE_DUPLICATES)
 to `true`.

docs/usage.md

@@ -299,19 +299,18 @@ for details.
 ### Duplicate documents
 
 By default, Paperless-ngx **does not reject duplicates**. If you consume a file whose
-contents exactly match an existing document (same checksum), the new copy is still
+contents match an existing document (same original or archive checksum), the new copy is
-consumed and a warning is logged. The task entry for the upload also flags that a
+still consumed and a warning is logged.
-duplicate was detected and links to the existing document(s).
 
-To review duplicates, open a document and switch to the **Duplicates** tab on the
+When a document has duplicates, a **Duplicates** tab appears on its detail page, listing
-document detail page. It lists other documents that share the same content, including any
+the other documents you can view that share the same content (including any in the trash).
-that are in the trash (shown with a badge), and links to each so you can decide which to
+To find all documents with duplicates, choose **Duplicates** in the document list's text
-keep.
+filter dropdown, or use `has_duplicates=true` in the REST API.
 
 If you would rather reject duplicates at consumption time (the pre-v3 behavior), set
 [`PAPERLESS_CONSUMER_DELETE_DUPLICATES`](configuration.md#PAPERLESS_CONSUMER_DELETE_DUPLICATES)
 to `true`. The duplicate file is then deleted instead of consumed, and the task fails with
-a "document already exists" message.
+a "Document already exists" message linking to the existing document.
 
 ## Document Suggestions
 

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "paperless-ngx"
-version = "3.2.1"
+version = "3.3.0"
 description = """\
   A community-supported supercharged document management system: scan, index and archive all your physical documents\
   """

src-ui/package.json

@@ -1,6 +1,6 @@
 {
   "name": "paperless-ngx-ui",
-  "version": "3.2.1",
+  "version": "3.3.0",
   "scripts": {
     "preinstall": "npx only-allow pnpm",
     "ng": "ng",

src-ui/src/environments/environment.prod.ts

@@ -8,7 +8,7 @@ export const environment = {
   apiVersion: '10', // match src/paperless/settings.py
   appTitle: DEFAULT_APP_TITLE,
   tag: 'prod',
-  version: '3.2.1',
+  version: '3.3.0',
   webSocketHost: window.location.host,
   webSocketProtocol: window.location.protocol == 'https:' ? 'wss:' : 'ws:',
   webSocketBaseUrl: base_url.pathname + 'ws/',

src/documents/bulk_edit.py

@@ -13,14 +13,8 @@ from celery import group
 from celery import shared_task
 from django.conf import settings
 from django.db import transaction
-from django.db.models import Case
-from django.db.models import F
 from django.db.models import Max
-from django.db.models import OuterRef
 from django.db.models import Q
-from django.db.models import Subquery
-from django.db.models import When
-from django.db.models.functions import Coalesce
 from django.utils import timezone
 
 from documents.data_models import ConsumableDocument
@@ -42,7 +36,6 @@ from documents.tasks import remove_document_from_index
 from documents.tasks import update_document_content_maybe_archive_file
 from documents.versioning import get_latest_version_for_root
 from documents.versioning import get_root_document
-from documents.versioning import versions_newest_first
 
 if TYPE_CHECKING:
     from collections.abc import Mapping
@@ -415,28 +408,10 @@ def reprocess(doc_ids: list[int], *, remote_ocr: bool = False) -> Literal["OK"]:
 
     Consumption workflows do not run here, so ``remote_ocr`` is how the user
     asks for the remote engine when it is not configured to handle everything.
-
-    A root document with versions reprocesses its latest version, which is the
-    file whose content, archive and thumbnail are shown for it.
     """
-    latest_version = versions_newest_first(
+    for document_id in doc_ids:
-        Document.objects.filter(root_document=OuterRef("pk")),
-    ).values("id")[:1]
-    source_ids = (
-        Document.objects.filter(id__in=doc_ids)
-        .annotate(
-            source_id=Case(
-                When(root_document__isnull=False, then=F("id")),
-                default=Coalesce(Subquery(latest_version), F("id")),
-            ),
-        )
-        .order_by()
-        .values_list("source_id", flat=True)
-        .distinct()
-    )
-    for source_id in source_ids:
         update_document_content_maybe_archive_file.apply_async(
-            kwargs={"document_id": source_id, "remote_ocr": remote_ocr},
+            kwargs={"document_id": document_id, "remote_ocr": remote_ocr},
             headers={"trigger_source": PaperlessTask.TriggerSource.MANUAL},
         )
 

src/documents/search/_backend.py

@@ -31,7 +31,6 @@ from documents.search._query import parse_user_query
 from documents.search._schema import _write_sentinels
 from documents.search._schema import build_schema
 from documents.search._schema import open_or_rebuild_index
-from documents.search._schema import rebuild_in_progress
 from documents.search._schema import wipe_index
 from documents.search._tokenizer import ascii_fold
 from documents.search._tokenizer import autocomplete_tokens
@@ -1111,44 +1110,39 @@ class TantivyBackend:
                 flushing a segment, deferring merge work; they do not avoid it.
         """
         wipe_index(self._path)
-        # The marker covers the window where the empty index is already stamped
+        new_index = tantivy.Index(build_schema(), path=str(self._path))
-        # as current but not yet populated, so an interrupted rebuild is retried.
+        _write_sentinels(self._path)
-        with rebuild_in_progress(self._path):
+        register_tokenizers(new_index, settings.SEARCH_LANGUAGE)
-            new_index = tantivy.Index(build_schema(), path=str(self._path))
-            _write_sentinels(self._path)
-            register_tokenizers(new_index, settings.SEARCH_LANGUAGE)
 
-            # Point instance at the new index so _build_tantivy_doc uses it
+        # Point instance at the new index so _build_tantivy_doc uses it
-            old_index, old_schema = self._raw_index, self._raw_schema
+        old_index, old_schema = self._raw_index, self._raw_schema
-            self._raw_index = new_index
+        self._raw_index = new_index
-            self._raw_schema = new_index.schema
+        self._raw_schema = new_index.schema
-            # Stream documents one-by-one (so the progress bar advances per
+        # Stream documents one-by-one (so the progress bar advances per
-            # document) while fetching viewer permissions one SQL query per
+        # document) while fetching viewer permissions one SQL query per chunk.
-            # chunk. The stream is Sized, so iter_wrapper can still discover
+        # The stream is Sized, so iter_wrapper can still discover the total.
-            # the total.
+        documents_stream = _DocumentViewerStream(documents, chunk_size=1000)
-            documents_stream = _DocumentViewerStream(documents, chunk_size=1000)
+        try:
-            try:
+            writer = new_index.writer(heap_size=writer_heap_bytes)
-                writer = new_index.writer(heap_size=writer_heap_bytes)
+            for document, (viewer_ids, viewer_group_ids) in iter_wrapper(
-                for document, (viewer_ids, viewer_group_ids) in iter_wrapper(
+                documents_stream,
-                    documents_stream,
+            ):
-                ):
+                doc = self._build_tantivy_doc(
-                    doc = self._build_tantivy_doc(
+                    document,
-                        document,
+                    viewer_ids=viewer_ids,
-                        viewer_ids=viewer_ids,
+                    viewer_group_ids=viewer_group_ids,
-                        viewer_group_ids=viewer_group_ids,
+                )
-                    )
+                writer.add_document(doc)
-                    writer.add_document(doc)
+            writer.commit()
-                writer.commit()
+            # Wait for background merge threads to finish so all segments are
-                # Wait for background merge threads to finish so all segments
+            # fully merged and persisted before the index is considered rebuilt.
-                # are fully merged and persisted before the index is considered
+            writer.wait_merging_threads()
-                # rebuilt.
+            new_index.reload()
-                writer.wait_merging_threads()
+        except BaseException:  # pragma: no cover
-                new_index.reload()
+            # Restore old index on failure so the backend remains usable
-            except BaseException:  # pragma: no cover
+            self._raw_index = old_index
-                # Restore old index on failure so the backend remains usable
+            self._raw_schema = old_schema
-                self._raw_index = old_index
+            raise
-                self._raw_schema = old_schema
-                raise
 
 
 def chunked(iterable, size):

src/documents/search/_schema.py

@@ -4,7 +4,6 @@ import hashlib
 import json
 import logging
 import shutil
-from contextlib import contextmanager
 from typing import TYPE_CHECKING
 from typing import Final
 from typing import NamedTuple
@@ -17,7 +16,6 @@ from whoosh_compat import FieldKind
 from documents.search._fields import PUBLIC_FIELDS
 
 if TYPE_CHECKING:
-    from collections.abc import Iterator
     from pathlib import Path
 
 logger = logging.getLogger("paperless.search")
@@ -30,11 +28,6 @@ logger = logging.getLogger("paperless.search")
 # v3 - barcodes JSON field for stored barcode contents
 SCHEMA_VERSION: Final[int] = 3
 
-# Present in the index directory from the moment a full rebuild starts until it
-# finishes. If a rebuild is interrupted it is left behind, so the half-built
-# index is not mistaken for a complete one.
-REBUILD_MARKER: Final[str] = ".rebuilding"
-
 
 class FieldDescriptor(NamedTuple):
     """One tantivy field, in declaration order.
@@ -262,9 +255,9 @@ def needs_rebuild(index_dir: Path) -> bool:
     """
     Check if the search index needs rebuilding.
 
-    True if a previous full rebuild never finished (the rebuild marker is still
+    Reads .index_settings.json to compare the stored schema version, search
-    present), or if the index's stamped settings no longer match the current
+    language and schema fingerprint against the current configuration. Returns
-    configuration. See _settings_mismatch().
+    True if the file is missing, unparsable, or any value mismatches.
 
     Args:
         index_dir: Path to the search index directory
@@ -272,40 +265,6 @@ def needs_rebuild(index_dir: Path) -> bool:
     Returns:
         True if the index needs rebuilding, False if it's up to date
     """
-    if (index_dir / REBUILD_MARKER).exists():
-        logger.warning("Previous search index rebuild did not finish - rebuilding.")
-        return True
-    return _settings_mismatch(index_dir)
-
-
-@contextmanager
-def rebuild_in_progress(index_dir: Path) -> Iterator[None]:
-    """
-    Flag the index as incomplete for the duration of a full rebuild.
-
-    The marker is cleared only if the block exits cleanly. There is deliberately
-    no try/finally: an exception must leave the marker behind so the next
-    needs_rebuild() check retries the rebuild.
-    """
-    marker = index_dir / REBUILD_MARKER
-    marker.touch()
-    yield
-    marker.unlink(missing_ok=True)
-
-
-def _settings_mismatch(index_dir: Path) -> bool:
-    """
-    Check the stamped settings against the current configuration.
-
-    Reads .index_settings.json to compare the stored schema version, search
-    language and schema fingerprint. Returns True if the file is missing,
-    unparsable, or any value mismatches.
-
-    This deliberately ignores the rebuild marker: open_or_rebuild_index() uses it
-    so that a process opening the index while another process is mid-rebuild
-    (or after one died) does not wipe the partial index out from under it.
-    Repopulating is the job of ``document_index reindex``.
-    """
     settings_file = index_dir / ".index_settings.json"
     if not settings_file.exists():
         return True
@@ -374,7 +333,7 @@ def open_or_rebuild_index(index_dir: Path | None = None) -> tantivy.Index:
         index_dir = cast("Path", settings.INDEX_DIR)
     if not index_dir.exists():
         return tantivy.Index(build_schema())
-    if _settings_mismatch(index_dir):
+    if needs_rebuild(index_dir):
         wipe_index(index_dir)
         idx = tantivy.Index(build_schema(), path=str(index_dir))
         _write_sentinels(index_dir)

src/documents/tasks.py

@@ -490,23 +490,18 @@ def update_document_content_maybe_archive_file(
                     shutil.move(thumbnail, document.thumbnail_path)
 
             document.refresh_from_db()
-            root_document = (
-                document.root_document if document.root_document_id else document
-            )
             logger.info(
-                f"Updating index for document {root_document.pk} ({document.archive_checksum})",
+                f"Updating index for document {document_id} ({document.archive_checksum})",
             )
             from documents.search import get_backend
 
-            get_backend().add_or_update(root_document)
+            get_backend().add_or_update(document)
 
             ai_config = AIConfig()
             if ai_config.llm_index_enabled:
-                llm_index_add_or_update_document(root_document)
+                llm_index_add_or_update_document(document)
 
             clear_document_caches(document.pk)
-            if root_document.pk != document.pk:
-                clear_document_caches(root_document.pk)
 
         except Exception:
             logger.exception(

src/documents/tests/search/test_backend.py

@@ -16,10 +16,7 @@ from documents.search._backend import TantivyBackend
 from documents.search._backend import WriteBatch
 from documents.search._backend import get_backend
 from documents.search._backend import reset_backend
-from documents.search._schema import REBUILD_MARKER
-from documents.search._schema import needs_rebuild
 from documents.signals.handlers import add_to_index
-from paperless_testing.dirs import PaperlessDirs
 from paperless_testing.factories import CorrespondentFactory
 from paperless_testing.factories import DocumentFactory
 from paperless_testing.factories import DocumentTypeFactory
@@ -826,53 +823,6 @@ class TestRebuild:
         backend.rebuild(Document.objects.all(), iter_wrapper=wrapper)
         assert 30 in seen
 
-    def test_successful_rebuild_leaves_index_up_to_date(
-        self,
-        backend: TantivyBackend,
-        paperless_dirs: PaperlessDirs,
-    ) -> None:
-        """
-        GIVEN:
-            - A backend and one document
-        WHEN:
-            - rebuild() completes
-        THEN:
-            - needs_rebuild() is False and no rebuild marker remains
-        """
-        DocumentFactory.create()
-
-        backend.rebuild(Document.objects.all())
-
-        assert needs_rebuild(paperless_dirs.index_dir) is False
-        assert not (paperless_dirs.index_dir / REBUILD_MARKER).exists()
-
-    def test_interrupted_rebuild_is_retried(
-        self,
-        backend: TantivyBackend,
-        paperless_dirs: PaperlessDirs,
-    ) -> None:
-        """
-        GIVEN:
-            - A rebuild that dies while indexing documents (e.g. the database
-              connection is lost)
-        WHEN:
-            - needs_rebuild() is checked afterwards
-        THEN:
-            - It is True, even though the empty index was already stamped with
-              current settings, so the next start rebuilds instead of reporting
-              the index as up to date
-        """
-        DocumentFactory.create()
-
-        def die(pairs):
-            raise RuntimeError("terminating connection due to administrator command")
-            yield  # pragma: no cover
-
-        with pytest.raises(RuntimeError):
-            backend.rebuild(Document.objects.all(), iter_wrapper=die)
-
-        assert needs_rebuild(paperless_dirs.index_dir) is True
-
     def test_includes_group_granted_viewers(self, backend: TantivyBackend) -> None:
         """Rebuild must index viewer ids for group-only grants, not just direct ones.
 

src/documents/tests/test_bulk_edit.py

@@ -4,7 +4,6 @@ from pathlib import Path
 from unittest import mock
 
 import pikepdf
-import pytest
 from django.contrib.auth.models import Group
 from django.contrib.auth.models import Permission
 from django.contrib.auth.models import User
@@ -13,7 +12,6 @@ from django.test import TestCase
 from django.test.utils import CaptureQueriesContext
 from guardian.shortcuts import get_groups_with_perms
 from guardian.shortcuts import get_users_with_perms
-from pytest_mock import MockerFixture
 
 from documents import bulk_edit
 from documents.models import Correspondent
@@ -25,7 +23,6 @@ from documents.models import StoragePath
 from documents.models import Tag
 from documents.permissions import set_permissions_for_objects
 from paperless_testing.dirs import DirectoriesMixin
-from paperless_testing.factories import DocumentFactory
 from paperless_testing.permissions import grant_object
 
 
@@ -1973,22 +1970,18 @@ class TestPDFActions(DirectoriesMixin, TestCase):
         self.assertIn("Error removing password from document", cm.output[0])
 
 
-@pytest.mark.django_db
+class TestBulkEditReprocess(DirectoriesMixin, TestCase):
-class TestBulkEditReprocess:
+    def setUp(self) -> None:
-    @pytest.fixture
+        super().setUp()
-    def mock_task(self, mocker: MockerFixture) -> mock.MagicMock:
+
-        return mocker.patch(
+        self.doc = Document.objects.create(
-            "documents.bulk_edit.update_document_content_maybe_archive_file",
+            title="test",
+            checksum="A",
+            mime_type="application/pdf",
         )
 
-    @staticmethod
+    @mock.patch("documents.bulk_edit.update_document_content_maybe_archive_file")
-    def _queued_ids(mock_task: mock.MagicMock) -> list[int]:
+    def test_reprocess_defaults_to_local(self, mock_task: mock.Mock) -> None:
-        return [
-            call.kwargs["kwargs"]["document_id"]
-            for call in mock_task.apply_async.call_args_list
-        ]
-
-    def test_reprocess_defaults_to_local(self, mock_task: mock.MagicMock) -> None:
         """
         GIVEN:
             - A reprocess request that says nothing about remote OCR
@@ -1997,17 +1990,18 @@ class TestBulkEditReprocess:
         THEN:
             - The task is queued without asking for the remote engine
         """
-        doc = DocumentFactory()
+        result = bulk_edit.reprocess([self.doc.id])
-
-        assert bulk_edit.reprocess([doc.id]) == "OK"
 
+        self.assertEqual(result, "OK")
         mock_task.apply_async.assert_called_once()
-        assert mock_task.apply_async.call_args.kwargs["kwargs"] == {
+        _, kwargs = mock_task.apply_async.call_args
-            "document_id": doc.id,
+        self.assertEqual(
-            "remote_ocr": False,
+            kwargs["kwargs"],
-        }
+            {"document_id": self.doc.id, "remote_ocr": False},
+        )
 
-    def test_reprocess_passes_remote_ocr(self, mock_task: mock.MagicMock) -> None:
+    @mock.patch("documents.bulk_edit.update_document_content_maybe_archive_file")
+    def test_reprocess_passes_remote_ocr(self, mock_task: mock.Mock) -> None:
         """
         GIVEN:
             - A reprocess request that explicitly asks for remote OCR
@@ -2016,72 +2010,14 @@ class TestBulkEditReprocess:
         THEN:
             - The request is forwarded to the task for every document
         """
-        docs = DocumentFactory.create_batch(2)
+        other = Document.objects.create(
-
+            title="test2",
-        bulk_edit.reprocess([doc.id for doc in docs], remote_ocr=True)
+            checksum="B",
-
+            mime_type="application/pdf",
-        assert mock_task.apply_async.call_count == 2
-        for call in mock_task.apply_async.call_args_list:
-            assert call.kwargs["kwargs"]["remote_ocr"]
-
-    def test_reprocess_root_uses_latest_version(
-        self,
-        mock_task: mock.MagicMock,
-    ) -> None:
-        """
-        GIVEN:
-            - A root document with two versions
-        WHEN:
-            - reprocess is called with the root document
-        THEN:
-            - The latest version is reprocessed, not the root's original file
-        """
-        root = DocumentFactory()
-        DocumentFactory(root_document=root, version_index=1)
-        latest = DocumentFactory(root_document=root, version_index=2)
-
-        bulk_edit.reprocess([root.id])
-
-        assert self._queued_ids(mock_task) == [latest.id]
-
-    def test_reprocess_explicit_version(self, mock_task: mock.MagicMock) -> None:
-        """
-        GIVEN:
-            - A root document with two versions
-        WHEN:
-            - reprocess is called with the older version
-        THEN:
-            - That version is reprocessed
-        """
-        root = DocumentFactory()
-        older = DocumentFactory(root_document=root, version_index=1)
-        DocumentFactory(root_document=root, version_index=2)
-
-        bulk_edit.reprocess([older.id])
-
-        assert self._queued_ids(mock_task) == [older.id]
-
-    def test_reprocess_root_and_latest_version_dispatches_once(
-        self,
-        mock_task: mock.MagicMock,
-    ) -> None:
-        """
-        GIVEN:
-            - A root document with two versions, the latest created on a
-              different date than the root
-        WHEN:
-            - reprocess is called with both the root and its latest version
-        THEN:
-            - The latest version is reprocessed only once
-        """
-        root = DocumentFactory(created=date(2024, 1, 1))
-        DocumentFactory(root_document=root, version_index=1)
-        latest = DocumentFactory(
-            root_document=root,
-            version_index=2,
-            created=date(2025, 1, 1),
         )
 
-        bulk_edit.reprocess([root.id, latest.id])
+        bulk_edit.reprocess([self.doc.id, other.id], remote_ocr=True)
 
-        assert self._queued_ids(mock_task) == [latest.id]
+        self.assertEqual(mock_task.apply_async.call_count, 2)
+        for call in mock_task.apply_async.call_args_list:
+            self.assertTrue(call.kwargs["kwargs"]["remote_ocr"])

src/documents/tests/test_tasks.py

@@ -20,7 +20,6 @@ from documents.sanity_checker import SanityCheckMessages
 from documents.tests.helpers import dummy_preprocess
 from paperless_testing.assertions import FileSystemAssertsMixin
 from paperless_testing.dirs import DirectoriesMixin
-from paperless_testing.factories import DocumentFactory
 
 
 @pytest.mark.django_db
@@ -288,82 +287,6 @@ class TestUpdateContent(DirectoriesMixin, TestCase):
         tasks.update_document_content_maybe_archive_file(doc.pk)
         self.assertNotEqual(Document.objects.get(pk=doc.pk).content, "test")
 
-    def _create_root_with_version(self) -> tuple[Document, Document]:
-        sample1 = self.dirs.scratch_dir / "sample.pdf"
-        shutil.copy(
-            Path(__file__).parent
-            / "samples"
-            / "documents"
-            / "originals"
-            / "0000001.pdf",
-            sample1,
-        )
-        root = DocumentFactory(content="root content", mime_type="application/pdf")
-        version = DocumentFactory(
-            content="my document",
-            filename=sample1,
-            mime_type="application/pdf",
-            root_document=root,
-            version_index=1,
-        )
-        return root, version
-
-    @mock.patch("documents.tasks.clear_document_caches")
-    @mock.patch("documents.search.get_backend")
-    def test_update_content_version_indexes_root(
-        self,
-        mock_get_backend: mock.Mock,
-        mock_clear_caches: mock.Mock,
-    ) -> None:
-        """
-        GIVEN:
-            - A root document with a version
-        WHEN:
-            - Update content task is called for the version
-        THEN:
-            - The version's content is updated
-            - The root document is indexed rather than the version
-            - Caches are cleared for both
-        """
-        root, version = self._create_root_with_version()
-
-        tasks.update_document_content_maybe_archive_file(version.pk)
-
-        self.assertNotEqual(
-            Document.objects.get(pk=version.pk).content,
-            "my document",
-        )
-        self.assertEqual(Document.objects.get(pk=root.pk).content, "root content")
-        indexed = mock_get_backend.return_value.add_or_update.call_args.args[0]
-        self.assertEqual(indexed.pk, root.pk)
-        mock_clear_caches.assert_has_calls(
-            [mock.call(version.pk), mock.call(root.pk)],
-        )
-
-    @override_settings(AI_ENABLED=True, LLM_EMBEDDING_BACKEND="huggingface")
-    @mock.patch("documents.tasks.llm_index_add_or_update_document")
-    @mock.patch("documents.search.get_backend")
-    def test_update_content_version_updates_llm_index_for_root(
-        self,
-        mock_get_backend: mock.Mock,
-        mock_llm_index: mock.Mock,
-    ) -> None:
-        """
-        GIVEN:
-            - A root document with a version
-            - The LLM index is enabled
-        WHEN:
-            - Update content task is called for the version
-        THEN:
-            - The LLM index is updated for the root document, not the version
-        """
-        root, version = self._create_root_with_version()
-
-        tasks.update_document_content_maybe_archive_file(version.pk)
-
-        mock_llm_index.assert_called_once()
-        self.assertEqual(mock_llm_index.call_args.args[0].pk, root.pk)
-
 
 class TestUpdateContentRemoteOCR(DirectoriesMixin, TestCase):
     """

src/locale/en_US/LC_MESSAGES/django.po

@@ -2,7 +2,7 @@ msgid ""
 msgstr ""
 "Project-Id-Version: paperless-ngx\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2026-10-06 15:12+0000\n"
+"POT-Creation-Date: 2026-10-05 16:26+0000\n"
 "PO-Revision-Date: 2022-02-17 04:17\n"
 "Last-Translator: \n"
 "Language-Team: English\n"
@@ -1941,8 +1941,25 @@ msgstr ""
 msgid "As a final step, please complete the following form:"
 msgstr ""
 
+#: documents/validators.py:24
+#, python-brace-format
+msgid "Unable to parse URI {value}, missing scheme"
+msgstr ""
+
+#: documents/validators.py:29
+#, python-brace-format
+msgid "Unable to parse URI {value}, missing net location or path"
+msgstr ""
+
 #: documents/validators.py:36
-msgid ", "
+msgid ""
+"URI scheme '{parts.scheme}' is not allowed. Allowed schemes: {', '."
+"join(allowed_schemes)}"
+msgstr ""
+
+#: documents/validators.py:45
+#, python-brace-format
+msgid "Unable to parse URI {value}"
 msgstr ""
 
 #: documents/views.py:336 documents/views.py:2729

src/paperless/tests/parsers/test_mail_parser_live.py

@@ -137,20 +137,9 @@ class TestNginxService:
     reason="No Gotenberg/Tika servers to test with",
 )
 class TestParserLive:
-    # Rasterizer versions shift a few pixels, so compare perceptual hashes by
+    @staticmethod
-    # Hamming distance (out of 18 * 18 = 324 bits) rather than for equality
+    def imagehash(file: Path, hash_size: int = 18) -> str:
-    MAX_HASH_DISTANCE = 8
+        return f"{average_hash(Image.open(file), hash_size)}"
-
-    @classmethod
-    def assert_thumbnails_similar(cls, generated: Path, expected: Path) -> None:
-        distance = average_hash(Image.open(generated), 18) - average_hash(
-            Image.open(expected),
-            18,
-        )
-        assert distance <= cls.MAX_HASH_DISTANCE, (
-            f"Thumbnail {generated} differs from {expected} by {distance} bits "
-            f"(max {cls.MAX_HASH_DISTANCE})"
-        )
 
     def test_get_thumbnail(
         self,
@@ -179,7 +168,12 @@ class TestParserLive:
         assert thumb.exists()
         assert thumb.is_file()
 
-        self.assert_thumbnails_similar(thumb, simple_txt_email_thumbnail_file)
+        assert self.imagehash(thumb) == self.imagehash(
+            simple_txt_email_thumbnail_file,
+        ), (
+            f"Created thumbnail {thumb} differs from expected file "
+            f"{simple_txt_email_thumbnail_file}"
+        )
 
     def test_tika_parse_successful(self, mail_parser: MailDocumentParser) -> None:
         """
@@ -261,7 +255,7 @@ class TestParserLive:
         THEN:
             - Gotenberg shall be called to generate the PDF
             - The archive PDF shall contain the expected content
-            - The generated thumbnail shall be perceptually close to the expected image
+            - The generated thumbnail shall match the expected image hash
         """
         util_call_with_backoff(mail_parser.parse, [html_email_file, "message/rfc822"])
 
@@ -278,4 +272,14 @@ class TestParserLive:
             html_email_file,
             "message/rfc822",
         )
-        self.assert_thumbnails_similar(generated_thumbnail, html_email_thumbnail_file)
+        generated_thumbnail_hash = self.imagehash(generated_thumbnail)
+
+        # The created PDF is not reproducible, but the converted image
+        # should always look the same
+        expected_hash = self.imagehash(html_email_thumbnail_file)
+
+        assert generated_thumbnail_hash == expected_hash, (
+            f"PDF thumbnail differs from expected. "
+            f"Generated: {generated_thumbnail}, "
+            f"Hash: {generated_thumbnail_hash} vs {expected_hash}"
+        )

src/paperless/version.py

@@ -1,6 +1,6 @@
 from typing import Final
 
-__version__: Final[tuple[int, int, int]] = (3, 2, 1)
+__version__: Final[tuple[int, int, int]] = (3, 3, 0)
 # Version string like X.Y.Z
 __full_version_str__: Final[str] = ".".join(map(str, __version__))
 # Version string like X.Y

uv.lock

@@ -2971,7 +2971,7 @@ wheels = [
 
 [[package]]
 name = "paperless-ngx"
-version = "3.2.1"
+version = "3.3.0"
 source = { virtual = "." }
 dependencies = [
     { name = "azure-ai-documentintelligence" },