Make content follow the version

- store content per version
- root doc retrieval returns latest content
- updating content affects the latest version
- load metadata per version

src-ui/src/app/components/document-detail/document-detail.component.ts

@@ -330,13 +330,19 @@ export class DocumentDetailComponent
   }
 
   get archiveContentRenderType(): ContentRenderType {
-    return this.document?.archived_file_name
+    const hasArchiveVersion =
+      this.metadata?.has_archive_version ?? !!this.document?.archived_file_name
+    return hasArchiveVersion
       ? this.getRenderType('application/pdf')
-      : this.getRenderType(this.document?.mime_type)
+      : this.getRenderType(
+          this.metadata?.original_mime_type || this.document?.mime_type
+        )
   }
 
   get originalContentRenderType(): ContentRenderType {
-    return this.getRenderType(this.document?.mime_type)
+    return this.getRenderType(
+      this.metadata?.original_mime_type || this.document?.mime_type
+    )
   }
 
   get showThumbnailOverlay(): boolean {
@@ -372,6 +378,39 @@ export class DocumentDetailComponent
     }
   }
 
+  private loadMetadataForSelectedVersion() {
+    this.documentsService
+      .getMetadata(this.documentId, this.selectedVersionId)
+      .pipe(
+        first(),
+        takeUntil(this.unsubscribeNotifier),
+        takeUntil(this.docChangeNotifier)
+      )
+      .subscribe({
+        next: (result) => {
+          this.metadata = result
+          this.tiffURL = null
+          this.tiffError = null
+          if (this.archiveContentRenderType === ContentRenderType.TIFF) {
+            this.tryRenderTiff()
+          }
+          if (
+            this.archiveContentRenderType !== ContentRenderType.PDF ||
+            this.useNativePdfViewer
+          ) {
+            this.previewLoaded = true
+          }
+        },
+        error: (error) => {
+          this.metadata = {} // allow display to fallback to <object> tag
+          this.toastService.showError(
+            $localize`Error retrieving metadata`,
+            error
+          )
+        },
+      })
+  }
+
   get isRTL() {
     if (!this.metadata || !this.metadata.lang) return false
     else {
@@ -724,36 +763,10 @@ export class DocumentDetailComponent
     this.selectedVersionId = versions.length
       ? Math.max(...versions.map((version) => version.id))
       : doc.id
+    this.previewLoaded = false
     this.requiresPassword = false
     this.updateFormForCustomFields()
-    if (this.archiveContentRenderType === ContentRenderType.TIFF) {
-      this.tryRenderTiff()
-    }
-    this.documentsService
-      .getMetadata(doc.id)
-      .pipe(
-        first(),
-        takeUntil(this.unsubscribeNotifier),
-        takeUntil(this.docChangeNotifier)
-      )
-      .subscribe({
-        next: (result) => {
-          this.metadata = result
-          if (
-            this.archiveContentRenderType !== ContentRenderType.PDF ||
-            this.useNativePdfViewer
-          ) {
-            this.previewLoaded = true
-          }
-        },
-        error: (error) => {
-          this.metadata = {} // allow display to fallback to <object> tag
-          this.toastService.showError(
-            $localize`Error retrieving metadata`,
-            error
-          )
-        },
-      })
+    this.loadMetadataForSelectedVersion()
     if (
       this.permissionsService.currentUserHasObjectPermissions(
         PermissionAction.Change,
@@ -785,6 +798,7 @@ export class DocumentDetailComponent
   // Update file preview and download target to a specific version (by document id)
   selectVersion(versionId: number) {
     this.selectedVersionId = versionId
+    this.previewLoaded = false
     this.previewUrl = this.documentsService.getPreviewUrl(
       this.documentId,
       false,
@@ -792,6 +806,7 @@ export class DocumentDetailComponent
     )
     this.updatePdfSource()
     this.thumbUrl = this.documentsService.getThumbUrl(this.selectedVersionId)
+    this.loadMetadataForSelectedVersion()
     // For text previews, refresh content
     this.http
       .get(this.previewUrl, { responseType: 'text' })
@@ -1840,7 +1855,7 @@ export class DocumentDetailComponent
     const modal = this.modalService.open(ShareLinksDialogComponent)
     modal.componentInstance.documentId = this.document.id
     modal.componentInstance.hasArchiveVersion =
-      !!this.document?.archived_file_name
+      this.metadata?.has_archive_version ?? !!this.document?.archived_file_name
   }
 
   get emailEnabled(): boolean {
@@ -1853,7 +1868,7 @@ export class DocumentDetailComponent
     })
     modal.componentInstance.documentIds = [this.document.id]
     modal.componentInstance.hasArchiveVersion =
-      !!this.document?.archived_file_name
+      this.metadata?.has_archive_version ?? !!this.document?.archived_file_name
   }
 
   private tryRenderTiff() {

src-ui/src/app/services/rest/document.service.spec.ts

@@ -165,6 +165,14 @@ describe(`DocumentService`, () => {
     expect(req.request.method).toEqual('GET')
   })
 
+  it('should call appropriate api endpoint for versioned metadata', () => {
+    subscription = service.getMetadata(documents[0].id, 123).subscribe()
+    const req = httpTestingController.expectOne(
+      `${environment.apiBaseUrl}${endpoint}/${documents[0].id}/metadata/?version=123`
+    )
+    expect(req.request.method).toEqual('GET')
+  })
+
   it('should call appropriate api endpoint for getting selection data', () => {
     const ids = [documents[0].id]
     subscription = service.getSelectionData(ids).subscribe()

src-ui/src/app/services/rest/document.service.ts

@@ -242,8 +242,15 @@ export class DocumentService extends AbstractPaperlessService<Document> {
     )
   }
 
-  getMetadata(id: number): Observable<DocumentMetadata> {
-    return this.http.get<DocumentMetadata>(this.getResourceUrl(id, 'metadata'))
+  getMetadata(
+    id: number,
+    versionID: number = null
+  ): Observable<DocumentMetadata> {
+    let url = new URL(this.getResourceUrl(id, 'metadata'))
+    if (versionID) {
+      url.searchParams.append('version', versionID.toString())
+    }
+    return this.http.get<DocumentMetadata>(url.toString())
   }
 
   bulkEdit(ids: number[], method: string, args: any) {

src/documents/consumer.py

@@ -520,7 +520,7 @@ class ConsumerPlugin(
                     original_document.checksum = hashlib.md5(
                         file_for_checksum.read_bytes(),
                     ).hexdigest()
-                    original_document.content = ""
+                    original_document.content = text
                     original_document.page_count = page_count
                     original_document.mime_type = mime_type
                     original_document.original_filename = self.filename

src/documents/filters.py

@@ -8,6 +8,7 @@ from contextlib import contextmanager
 from typing import TYPE_CHECKING
 
 from django.contrib.contenttypes.models import ContentType
+from django.core.exceptions import FieldError
 from django.db.models import Case
 from django.db.models import CharField
 from django.db.models import Count
@@ -163,11 +164,34 @@ class TitleContentFilter(Filter):
     def filter(self, qs, value):
         value = value.strip() if isinstance(value, str) else value
         if value:
-            return qs.filter(Q(title__icontains=value) | Q(content__icontains=value))
+            try:
+                return qs.filter(
+                    Q(title__icontains=value) | Q(effective_content__icontains=value),
+                )
+            except FieldError:
+                return qs.filter(
+                    Q(title__icontains=value) | Q(content__icontains=value),
+                )
         else:
             return qs
 
 
+@extend_schema_field(serializers.CharField)
+class EffectiveContentFilter(Filter):
+    def filter(self, qs, value):
+        value = value.strip() if isinstance(value, str) else value
+        if not value:
+            return qs
+        try:
+            return qs.filter(
+                **{f"effective_content__{self.lookup_expr}": value},
+            )
+        except FieldError:
+            return qs.filter(
+                **{f"content__{self.lookup_expr}": value},
+            )
+
+
 @extend_schema_field(serializers.BooleanField)
 class SharedByUser(Filter):
     def filter(self, qs, value):
@@ -724,6 +748,11 @@ class DocumentFilterSet(FilterSet):
 
     title_content = TitleContentFilter()
 
+    content__istartswith = EffectiveContentFilter(lookup_expr="istartswith")
+    content__iendswith = EffectiveContentFilter(lookup_expr="iendswith")
+    content__icontains = EffectiveContentFilter(lookup_expr="icontains")
+    content__iexact = EffectiveContentFilter(lookup_expr="iexact")
+
     owner__id__none = ObjectFilter(field_name="owner", exclude=True)
 
     custom_fields__icontains = CustomFieldsFilter()
@@ -764,7 +793,6 @@ class DocumentFilterSet(FilterSet):
         fields = {
             "id": ID_KWARGS,
             "title": CHAR_KWARGS,
-            "content": CHAR_KWARGS,
             "archive_serial_number": INT_KWARGS,
             "created": DATE_KWARGS,
             "added": DATETIME_KWARGS,

src/documents/index.py

@@ -185,10 +185,20 @@ def update_document(writer: AsyncWriter, doc: Document) -> None:
         only_with_perms_in=["view_document"],
     )
     viewer_ids: str = ",".join([str(u.id) for u in users_with_perms])
+    effective_content = doc.content
+    if doc.root_document_id is None:
+        latest_version = (
+            Document.objects.filter(root_document=doc)
+            .only("content")
+            .order_by("-id")
+            .first()
+        )
+        if latest_version is not None:
+            effective_content = latest_version.content
     writer.update_document(
         id=doc.pk,
         title=doc.title,
-        content=doc.content,
+        content=effective_content,
         correspondent=doc.correspondent.name if doc.correspondent else None,
         correspondent_id=doc.correspondent.id if doc.correspondent else None,
         has_correspondent=doc.correspondent is not None,

src/documents/serialisers.py

@@ -1177,6 +1177,8 @@ class DocumentSerializer(
 
     def to_representation(self, instance):
         doc = super().to_representation(instance)
+        if "content" in self.fields and hasattr(instance, "effective_content"):
+            doc["content"] = getattr(instance, "effective_content") or ""
         if self.truncate_content and "content" in self.fields:
             doc["content"] = doc.get("content")[0:550]
 

src/documents/tests/test_api_document_versions.py

@@ -116,18 +116,21 @@ class TestDocumentVersioningApi(DirectoriesMixin, APITestCase):
             title="root",
             checksum="root",
             mime_type="application/pdf",
+            content="root-content",
         )
         v1 = Document.objects.create(
             title="v1",
             checksum="v1",
             mime_type="application/pdf",
             root_document=root,
+            content="v1-content",
         )
         v2 = Document.objects.create(
             title="v2",
             checksum="v2",
             mime_type="application/pdf",
             root_document=root,
+            content="v2-content",
         )
 
         with mock.patch("documents.index.remove_document_from_index"):
@@ -136,6 +139,8 @@ class TestDocumentVersioningApi(DirectoriesMixin, APITestCase):
         self.assertEqual(resp.status_code, status.HTTP_200_OK)
         self.assertFalse(Document.objects.filter(id=v2.id).exists())
         self.assertEqual(resp.data["current_version_id"], v1.id)
+        root.refresh_from_db()
+        self.assertEqual(root.content, "root-content")
 
         with mock.patch("documents.index.remove_document_from_index"):
             resp = self.client.delete(f"/api/documents/{root.id}/versions/{v1.id}/")
@@ -143,6 +148,8 @@ class TestDocumentVersioningApi(DirectoriesMixin, APITestCase):
         self.assertEqual(resp.status_code, status.HTTP_200_OK)
         self.assertFalse(Document.objects.filter(id=v1.id).exists())
         self.assertEqual(resp.data["current_version_id"], root.id)
+        root.refresh_from_db()
+        self.assertEqual(root.content, "root-content")
 
     def test_delete_version_writes_audit_log_entry(self) -> None:
         root = Document.objects.create(
@@ -454,3 +461,60 @@ class TestDocumentVersioningApi(DirectoriesMixin, APITestCase):
         )
 
         self.assertEqual(resp.status_code, status.HTTP_400_BAD_REQUEST)
+
+    def test_patch_content_updates_latest_version_content(self) -> None:
+        root = Document.objects.create(
+            title="root",
+            checksum="root",
+            mime_type="application/pdf",
+            content="root-content",
+        )
+        v1 = Document.objects.create(
+            title="v1",
+            checksum="v1",
+            mime_type="application/pdf",
+            root_document=root,
+            content="v1-content",
+        )
+        v2 = Document.objects.create(
+            title="v2",
+            checksum="v2",
+            mime_type="application/pdf",
+            root_document=root,
+            content="v2-content",
+        )
+
+        resp = self.client.patch(
+            f"/api/documents/{root.id}/",
+            {"content": "edited-content"},
+            format="json",
+        )
+
+        self.assertEqual(resp.status_code, status.HTTP_200_OK)
+        self.assertEqual(resp.data["content"], "edited-content")
+        root.refresh_from_db()
+        v1.refresh_from_db()
+        v2.refresh_from_db()
+        self.assertEqual(v2.content, "edited-content")
+        self.assertEqual(root.content, "root-content")
+        self.assertEqual(v1.content, "v1-content")
+
+    def test_retrieve_returns_latest_version_content(self) -> None:
+        root = Document.objects.create(
+            title="root",
+            checksum="root",
+            mime_type="application/pdf",
+            content="root-content",
+        )
+        Document.objects.create(
+            title="v1",
+            checksum="v1",
+            mime_type="application/pdf",
+            root_document=root,
+            content="v1-content",
+        )
+
+        resp = self.client.get(f"/api/documents/{root.id}/")
+
+        self.assertEqual(resp.status_code, status.HTTP_200_OK)
+        self.assertEqual(resp.data["content"], "v1-content")

src/documents/tests/test_consumer.py

@@ -745,6 +745,7 @@ class TestConsumer(
         assert version.original_filename is not None
         self.assertEqual(version.version_label, "v2")
         self.assertTrue(version.original_filename.endswith("_v0.pdf"))
+        self.assertTrue(bool(version.content))
 
     @mock.patch("documents.consumer.load_classifier")
     def testClassifyDocument(self, m) -> None:

src/documents/views.py

@@ -30,12 +30,16 @@ from django.db.migrations.loader import MigrationLoader
 from django.db.migrations.recorder import MigrationRecorder
 from django.db.models import Case
 from django.db.models import Count
+from django.db.models import F
 from django.db.models import IntegerField
 from django.db.models import Max
 from django.db.models import Model
+from django.db.models import OuterRef
 from django.db.models import Q
+from django.db.models import Subquery
 from django.db.models import Sum
 from django.db.models import When
+from django.db.models.functions import Coalesce
 from django.db.models.functions import Lower
 from django.db.models.manager import Manager
 from django.db.models.query import QuerySet
@@ -763,7 +767,7 @@ class DocumentViewSet(
         ObjectOwnedOrGrantedPermissionsFilter,
     )
     filterset_class = DocumentFilterSet
-    search_fields = ("title", "correspondent__name", "content")
+    search_fields = ("title", "correspondent__name", "effective_content")
     ordering_fields = (
         "id",
         "title",
@@ -781,10 +785,16 @@ class DocumentViewSet(
     )
 
     def get_queryset(self):
+        latest_version_content = Subquery(
+            Document.objects.filter(root_document=OuterRef("pk"))
+            .order_by("-id")
+            .values("content")[:1],
+        )
         return (
             Document.objects.filter(root_document__isnull=True)
             .distinct()
             .order_by("-created")
+            .annotate(effective_content=Coalesce(latest_version_content, F("content")))
             .annotate(num_notes=Count("notes"))
             .select_related("correspondent", "storage_path", "document_type", "owner")
             .prefetch_related("tags", "custom_fields", "notes")
@@ -847,14 +857,45 @@ class DocumentViewSet(
         return Response({"root_id": root_doc.id})
 
     def update(self, request, *args, **kwargs):
-        response = super().update(request, *args, **kwargs)
+        partial = kwargs.pop("partial", False)
+        root_doc = self.get_object()
+        content_updated = "content" in request.data
+        updated_content = request.data.get("content") if content_updated else None
+        latest_doc = self._get_latest_doc_for_root(root_doc)
+
+        data = request.data.copy()
+        serializer_partial = partial
+        if content_updated and latest_doc.id != root_doc.id:
+            if updated_content is None:
+                raise ValidationError({"content": ["This field may not be null."]})
+            data.pop("content", None)
+            serializer_partial = True
+
+        serializer = self.get_serializer(
+            root_doc,
+            data=data,
+            partial=serializer_partial,
+        )
+        serializer.is_valid(raise_exception=True)
+        self.perform_update(serializer)
+
+        if content_updated and latest_doc.id != root_doc.id:
+            latest_doc.content = updated_content
+            latest_doc.save(update_fields=["content", "modified"])
+
+        if getattr(root_doc, "_prefetched_objects_cache", None):
+            root_doc._prefetched_objects_cache = {}
+
+        refreshed_doc = self.get_queryset().get(pk=root_doc.pk)
+        response = Response(self.get_serializer(refreshed_doc).data)
+
         from documents import index
 
-        index.add_or_update_document(self.get_object())
+        index.add_or_update_document(refreshed_doc)
 
         document_updated.send(
             sender=self.__class__,
-            document=self.get_object(),
+            document=refreshed_doc,
         )
 
         return response
@@ -904,6 +945,11 @@ class DocumentViewSet(
         latest = Document.objects.filter(root_document=root_doc).order_by("id").last()
         return latest or root_doc
 
+    @staticmethod
+    def _get_latest_doc_for_root(root_doc: Document) -> Document:
+        latest = Document.objects.filter(root_document=root_doc).order_by("-id").first()
+        return latest or root_doc
+
     def file_response(self, pk, request, disposition):
         request_doc = Document.global_objects.select_related(
             "owner",