Fix/refactor

docs/configuration.md

@@ -1146,8 +1146,9 @@ via the consumption directory, you can disable the consumer to save resources.
 
 #### [`PAPERLESS_CONSUMER_DELETE_DUPLICATES=<bool>`](#PAPERLESS_CONSUMER_DELETE_DUPLICATES) {#PAPERLESS_CONSUMER_DELETE_DUPLICATES}
 
-: When the consumer detects a duplicate document, it will not touch
+: As of version 3.0 Paperless-ngx allows duplicate documents to be consumed by default, _except_ when
-the original document. This default behavior can be changed here.
+this setting is enabled. When enabled, Paperless will check if a document with the same hash already
+exists in the system and delete the duplicate file from the consumption directory without consuming it.
 
     Defaults to false.
 

src-ui/src/app/components/admin/tasks/tasks.component.html

@@ -97,6 +97,12 @@
                   <br/><em>(<ng-container i18n>click for full output</ng-container>)</em>
                 }
               </ng-template>
+              @if (task.duplicate_documents?.length > 0) {
+                <div class="small text-warning-emphasis d-flex align-items-center gap-1">
+                  <i-bs class="lh-1" width="1em" height="1em" name="exclamation-triangle"></i-bs>
+                  <span i18n>Duplicate(s) detected</span>
+                </div>
+              }
             </td>
           }
           <td class="d-lg-none">

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

@@ -370,6 +370,37 @@
             </ng-template>
           </li>
         }
+
+        @if (document?.duplicate_documents?.length) {
+          <li [ngbNavItem]="DocumentDetailNavIDs.Duplicates">
+            <a class="text-nowrap" ngbNavLink i18n>
+              Duplicates
+              <span class="badge text-bg-secondary ms-1">{{ document.duplicate_documents.length }}</span>
+            </a>
+            <ng-template ngbNavContent>
+              <div class="d-flex flex-column gap-2">
+                <div class="fst-italic" i18n>Duplicate documents detected:</div>
+                <div class="list-group">
+                  @for (duplicate of document.duplicate_documents; track duplicate.id) {
+                    <a
+                      class="list-group-item list-group-item-action d-flex justify-content-between align-items-center"
+                      [routerLink]="['/documents', duplicate.id, 'details']"
+                      [class.disabled]="duplicate.deleted_at"
+                    >
+                      <span class="d-flex align-items-center gap-2">
+                        <span>{{ duplicate.title || ('#' + duplicate.id) }}</span>
+                        @if (duplicate.deleted_at) {
+                          <span class="badge text-bg-secondary" i18n>In trash</span>
+                        }
+                      </span>
+                      <span class="text-secondary">#{{ duplicate.id }}</span>
+                    </a>
+                  }
+                </div>
+              </div>
+            </ng-template>
+          </li>
+        }
       </ul>
 
       <div [ngbNavOutlet]="nav" class="mt-3"></div>

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

@@ -301,16 +301,16 @@ describe('DocumentDetailComponent', () => {
       .spyOn(openDocumentsService, 'openDocument')
       .mockReturnValueOnce(of(true))
     fixture.detectChanges()
-    expect(component.activeNavID).toEqual(5) // DocumentDetailNavIDs.Notes
+    expect(component.activeNavID).toEqual(component.DocumentDetailNavIDs.Notes)
   })
 
   it('should change url on tab switch', () => {
     initNormally()
     const navigateSpy = jest.spyOn(router, 'navigate')
-    component.nav.select(5)
+    component.nav.select(component.DocumentDetailNavIDs.Notes)
     component.nav.navChange.next({
       activeId: 1,
-      nextId: 5,
+      nextId: component.DocumentDetailNavIDs.Notes,
       preventDefault: () => {},
     })
     fixture.detectChanges()
@@ -352,6 +352,18 @@ describe('DocumentDetailComponent', () => {
     expect(component.document).toEqual(doc)
   })
 
+  it('should fall back to details tab when duplicates tab is active but no duplicates', () => {
+    initNormally()
+    component.activeNavID = component.DocumentDetailNavIDs.Duplicates
+    const noDupDoc = { ...doc, duplicate_documents: [] }
+
+    component.updateComponent(noDupDoc)
+
+    expect(component.activeNavID).toEqual(
+      component.DocumentDetailNavIDs.Details
+    )
+  })
+
   it('should load already-opened document via param', () => {
     initNormally()
     jest.spyOn(documentService, 'get').mockReturnValueOnce(of(doc))
@@ -367,6 +379,38 @@ describe('DocumentDetailComponent', () => {
     expect(component.document).toEqual(doc)
   })
 
+  it('should update cached open document duplicates when reloading an open doc', () => {
+    const openDoc = { ...doc, duplicate_documents: [{ id: 1, title: 'Old' }] }
+    const updatedDuplicates = [
+      { id: 2, title: 'Newer duplicate', deleted_at: null },
+    ]
+    jest
+      .spyOn(activatedRoute, 'paramMap', 'get')
+      .mockReturnValue(of(convertToParamMap({ id: 3, section: 'details' })))
+    jest.spyOn(documentService, 'get').mockReturnValue(
+      of({
+        ...doc,
+        modified: new Date('2024-01-02T00:00:00Z'),
+        duplicate_documents: updatedDuplicates,
+      })
+    )
+    jest.spyOn(openDocumentsService, 'getOpenDocument').mockReturnValue(openDoc)
+    const saveSpy = jest.spyOn(openDocumentsService, 'save')
+    jest.spyOn(openDocumentsService, 'openDocument').mockReturnValue(of(true))
+    jest.spyOn(customFieldsService, 'listAll').mockReturnValue(
+      of({
+        count: customFields.length,
+        all: customFields.map((f) => f.id),
+        results: customFields,
+      })
+    )
+
+    fixture.detectChanges()
+
+    expect(openDoc.duplicate_documents).toEqual(updatedDuplicates)
+    expect(saveSpy).toHaveBeenCalled()
+  })
+
   it('should disable form if user cannot edit', () => {
     currentUserHasObjectPermissions = false
     initNormally()

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

@@ -8,7 +8,7 @@ import {
   FormsModule,
   ReactiveFormsModule,
 } from '@angular/forms'
-import { ActivatedRoute, Router } from '@angular/router'
+import { ActivatedRoute, Router, RouterModule } from '@angular/router'
 import {
   NgbDateStruct,
   NgbDropdownModule,
@@ -124,6 +124,7 @@ enum DocumentDetailNavIDs {
   Notes = 5,
   Permissions = 6,
   History = 7,
+  Duplicates = 8,
 }
 
 enum ContentRenderType {
@@ -181,6 +182,7 @@ export enum ZoomSetting {
     NgxBootstrapIconsModule,
     PdfViewerModule,
     TextAreaComponent,
+    RouterModule,
   ],
 })
 export class DocumentDetailComponent
@@ -454,6 +456,11 @@ export class DocumentDetailComponent
           const openDocument = this.openDocumentService.getOpenDocument(
             this.documentId
           )
+          // update duplicate documents if present
+          if (openDocument && doc?.duplicate_documents) {
+            openDocument.duplicate_documents = doc.duplicate_documents
+            this.openDocumentService.save()
+          }
           const useDoc = openDocument || doc
           if (openDocument) {
             if (
@@ -704,6 +711,13 @@ export class DocumentDetailComponent
     }
     this.title = this.documentTitlePipe.transform(doc.title)
     this.prepareForm(doc)
+
+    if (
+      this.activeNavID === DocumentDetailNavIDs.Duplicates &&
+      !doc?.duplicate_documents?.length
+    ) {
+      this.activeNavID = DocumentDetailNavIDs.Details
+    }
   }
 
   get customFieldFormFields(): FormArray {

src-ui/src/app/data/document.ts

@@ -159,6 +159,8 @@ export interface Document extends ObjectWithPermissions {
 
   page_count?: number
 
+  duplicate_documents?: Document[]
+
   // Frontend only
   __changedFields?: string[]
 }

src-ui/src/app/data/paperless-task.ts

@@ -1,3 +1,4 @@
+import { Document } from './document'
 import { ObjectWithId } from './object-with-id'
 
 export enum PaperlessTaskType {
@@ -42,5 +43,7 @@ export interface PaperlessTask extends ObjectWithId {
 
   related_document?: number
 
+  duplicate_documents?: Document[]
+
   owner?: number
 }

src/documents/consumer.py

@@ -785,19 +785,45 @@ class ConsumerPreflightPlugin(
             Q(checksum=checksum) | Q(archive_checksum=checksum),
         )
         if existing_doc.exists():
-            msg = ConsumerStatusShortMessage.DOCUMENT_ALREADY_EXISTS
+            existing_doc = existing_doc.order_by("-created")
-            log_msg = f"Not consuming {self.filename}: It is a duplicate of {existing_doc.get().title} (#{existing_doc.get().pk})."
+            duplicates_in_trash = existing_doc.filter(deleted_at__isnull=False)
+            log_msg = (
+                f"Consuming duplicate {self.filename}: "
+                f"{existing_doc.count()} existing document(s) share the same content."
+            )
 
-            if existing_doc.first().deleted_at is not None:
+            if duplicates_in_trash.exists():
-                msg = ConsumerStatusShortMessage.DOCUMENT_ALREADY_EXISTS_IN_TRASH
+                log_msg += " Note: at least one existing document is in the trash."
-                log_msg += " Note: existing document is in the trash."
+
+            self.log.warning(log_msg)
 
             if settings.CONSUMER_DELETE_DUPLICATES:
+                duplicate = existing_doc.first()
+                duplicate_label = (
+                    duplicate.title
+                    or duplicate.original_filename
+                    or (Path(duplicate.filename).name if duplicate.filename else None)
+                    or str(duplicate.pk)
+                )
+
                 Path(self.input_doc.original_file).unlink()
-            self._fail(
+
-                msg,
+                failure_msg = (
-                log_msg,
+                    f"Not consuming {self.filename}: "
-            )
+                    f"It is a duplicate of {duplicate_label} (#{duplicate.pk})"
+                )
+                status_msg = ConsumerStatusShortMessage.DOCUMENT_ALREADY_EXISTS
+
+                if duplicates_in_trash.exists():
+                    status_msg = (
+                        ConsumerStatusShortMessage.DOCUMENT_ALREADY_EXISTS_IN_TRASH
+                    )
+                    failure_msg += " Note: existing document is in the trash."
+
+                self._fail(
+                    status_msg,
+                    failure_msg,
+                )
 
     def pre_check_directories(self):
         """

src/documents/migrations/1077_alter_document_checksum_unique.py

@@ -0,0 +1,23 @@
+# Generated by Django 5.2.7 on 2026-01-14 17:45
+
+from django.db import migrations
+from django.db import models
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ("documents", "1076_alter_paperlesstask_task_name"),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name="document",
+            name="checksum",
+            field=models.CharField(
+                editable=False,
+                max_length=32,
+                verbose_name="checksum",
+                help_text="The checksum of the original document.",
+            ),
+        ),
+    ]

src/documents/models.py

@@ -212,7 +212,6 @@ class Document(SoftDeleteModel, ModelWithOwner):
         _("checksum"),
         max_length=32,
         editable=False,
-        unique=True,
         help_text=_("The checksum of the original document."),
     )
 

src/documents/permissions.py

@@ -148,13 +148,29 @@ def get_document_count_filter_for_user(user):
     )
 
 
-def get_objects_for_user_owner_aware(user, perms, Model) -> QuerySet:
+def get_objects_for_user_owner_aware(
-    objects_owned = Model.objects.filter(owner=user)
+    user,
-    objects_unowned = Model.objects.filter(owner__isnull=True)
+    perms,
+    Model,
+    *,
+    include_deleted=False,
+) -> QuerySet:
+    """
+    Returns objects the user owns, are unowned, or has explicit perms.
+    When include_deleted is True, soft-deleted items are also included.
+    """
+    manager = (
+        Model.global_objects
+        if include_deleted and hasattr(Model, "global_objects")
+        else Model.objects
+    )
+
+    objects_owned = manager.filter(owner=user)
+    objects_unowned = manager.filter(owner__isnull=True)
     objects_with_perms = get_objects_for_user(
         user=user,
         perms=perms,
-        klass=Model,
+        klass=manager.all(),
         accept_global_perms=False,
     )
     return objects_owned | objects_unowned | objects_with_perms

src/documents/serialisers.py

@@ -23,6 +23,7 @@ from django.core.validators import MinValueValidator
 from django.core.validators import RegexValidator
 from django.core.validators import integer_validator
 from django.db.models import Count
+from django.db.models import Q
 from django.db.models.functions import Lower
 from django.utils.crypto import get_random_string
 from django.utils.dateparse import parse_datetime
@@ -72,6 +73,7 @@ from documents.models import WorkflowTrigger
 from documents.parsers import is_mime_type_supported
 from documents.permissions import get_document_count_filter_for_user
 from documents.permissions import get_groups_with_only_permission
+from documents.permissions import get_objects_for_user_owner_aware
 from documents.permissions import set_permissions_for_object
 from documents.regex import validate_regex_pattern
 from documents.templating.filepath import validate_filepath_template_and_render
@@ -1014,6 +1016,29 @@ class NotesSerializer(serializers.ModelSerializer):
         return ret
 
 
+def _get_viewable_duplicates(document: Document, user: User | None):
+    checksums = {document.checksum}
+    if document.archive_checksum:
+        checksums.add(document.archive_checksum)
+    duplicates = Document.global_objects.filter(
+        Q(checksum__in=checksums) | Q(archive_checksum__in=checksums),
+    ).exclude(pk=document.pk)
+    duplicates = duplicates.order_by("-created")
+    allowed = get_objects_for_user_owner_aware(
+        user,
+        "documents.view_document",
+        Document,
+        include_deleted=True,
+    )
+    return duplicates.filter(id__in=allowed.values_list("id", flat=True))
+
+
+class DuplicateDocumentSummarySerializer(serializers.Serializer):
+    id = serializers.IntegerField()
+    title = serializers.CharField()
+    deleted_at = serializers.DateTimeField(allow_null=True)
+
+
 @extend_schema_serializer(
     deprecate_fields=["created_date"],
 )
@@ -1031,6 +1056,7 @@ class DocumentSerializer(
     archived_file_name = SerializerMethodField()
     created_date = serializers.DateField(required=False)
     page_count = SerializerMethodField()
+    duplicate_documents = SerializerMethodField()
 
     notes = NotesSerializer(many=True, required=False, read_only=True)
 
@@ -1056,6 +1082,16 @@ class DocumentSerializer(
     def get_page_count(self, obj) -> int | None:
         return obj.page_count
 
+    @extend_schema_field(DuplicateDocumentSummarySerializer(many=True))
+    def get_duplicate_documents(self, obj):
+        view = self.context.get("view")
+        if view and getattr(view, "action", None) != "retrieve":
+            return []
+        request = self.context.get("request")
+        user = request.user if request else None
+        duplicates = _get_viewable_duplicates(obj, user)
+        return list(duplicates.values("id", "title", "deleted_at"))
+
     def get_original_file_name(self, obj) -> str | None:
         return obj.original_filename
 
@@ -1233,6 +1269,7 @@ class DocumentSerializer(
             "archive_serial_number",
             "original_file_name",
             "archived_file_name",
+            "duplicate_documents",
             "owner",
             "permissions",
             "user_can_change",
@@ -2094,10 +2131,12 @@ class TasksViewSerializer(OwnedObjectSerializer):
             "result",
             "acknowledged",
             "related_document",
+            "duplicate_documents",
             "owner",
         )
 
     related_document = serializers.SerializerMethodField()
+    duplicate_documents = serializers.SerializerMethodField()
     created_doc_re = re.compile(r"New document id (\d+) created")
     duplicate_doc_re = re.compile(r"It is a duplicate of .* \(#(\d+)\)")
 
@@ -2122,6 +2161,17 @@ class TasksViewSerializer(OwnedObjectSerializer):
 
         return result
 
+    @extend_schema_field(DuplicateDocumentSummarySerializer(many=True))
+    def get_duplicate_documents(self, obj):
+        related_document = self.get_related_document(obj)
+        request = self.context.get("request")
+        user = request.user if request else None
+        document = Document.global_objects.filter(pk=related_document).first()
+        if not related_document or not user or not document:
+            return []
+        duplicates = _get_viewable_duplicates(document, user)
+        return list(duplicates.values("id", "title", "deleted_at"))
+
 
 class RunTaskViewSerializer(serializers.Serializer):
     task_name = serializers.ChoiceField(

src/documents/tests/test_api_tasks.py

@@ -7,6 +7,7 @@ from django.contrib.auth.models import User
 from rest_framework import status
 from rest_framework.test import APITestCase
 
+from documents.models import Document
 from documents.models import PaperlessTask
 from documents.tests.utils import DirectoriesMixin
 from documents.views import TasksViewSet
@@ -258,7 +259,7 @@ class TestTasks(DirectoriesMixin, APITestCase):
             task_id=str(uuid.uuid4()),
             task_file_name="task_one.pdf",
             status=celery.states.FAILURE,
-            result="test.pdf: Not consuming test.pdf: It is a duplicate.",
+            result="test.pdf: Unexpected error during ingestion.",
         )
 
         response = self.client.get(self.ENDPOINT)
@@ -270,7 +271,7 @@ class TestTasks(DirectoriesMixin, APITestCase):
 
         self.assertEqual(
             returned_data["result"],
-            "test.pdf: Not consuming test.pdf: It is a duplicate.",
+            "test.pdf: Unexpected error during ingestion.",
         )
 
     def test_task_name_webui(self):
@@ -325,20 +326,34 @@ class TestTasks(DirectoriesMixin, APITestCase):
 
         self.assertEqual(returned_data["task_file_name"], "anothertest.pdf")
 
-    def test_task_result_failed_duplicate_includes_related_doc(self):
+    def test_task_result_duplicate_warning_includes_count(self):
         """
         GIVEN:
-            - A celery task failed with a duplicate error
+            - A celery task succeeds, but a duplicate exists
         WHEN:
             - API call is made to get tasks
         THEN:
-            - The returned data includes a related document link
+            - The returned data includes duplicate warning metadata
         """
+        checksum = "duplicate-checksum"
+        Document.objects.create(
+            title="Existing",
+            content="",
+            mime_type="application/pdf",
+            checksum=checksum,
+        )
+        created_doc = Document.objects.create(
+            title="Created",
+            content="",
+            mime_type="application/pdf",
+            checksum=checksum,
+            archive_checksum="another-checksum",
+        )
         PaperlessTask.objects.create(
             task_id=str(uuid.uuid4()),
             task_file_name="task_one.pdf",
-            status=celery.states.FAILURE,
+            status=celery.states.SUCCESS,
-            result="Not consuming task_one.pdf: It is a duplicate of task_one_existing.pdf (#1234).",
+            result=f"Success. New document id {created_doc.pk} created",
         )
 
         response = self.client.get(self.ENDPOINT)
@@ -348,7 +363,7 @@ class TestTasks(DirectoriesMixin, APITestCase):
 
         returned_data = response.data[0]
 
-        self.assertEqual(returned_data["related_document"], "1234")
+        self.assertEqual(returned_data["related_document"], str(created_doc.pk))
 
     def test_run_train_classifier_task(self):
         """

src/documents/tests/test_consumer.py

@@ -485,21 +485,21 @@ class TestConsumer(
         with self.get_consumer(self.get_test_file()) as consumer:
             consumer.run()
 
-        with self.assertRaisesMessage(ConsumerError, "It is a duplicate"):
+        with self.get_consumer(self.get_test_file()) as consumer:
-            with self.get_consumer(self.get_test_file()) as consumer:
+            consumer.run()
-                consumer.run()
 
-        self._assert_first_last_send_progress(last_status="FAILED")
+        self.assertEqual(Document.objects.count(), 2)
+        self._assert_first_last_send_progress()
 
     def testDuplicates2(self):
         with self.get_consumer(self.get_test_file()) as consumer:
             consumer.run()
 
-        with self.assertRaisesMessage(ConsumerError, "It is a duplicate"):
+        with self.get_consumer(self.get_test_archive_file()) as consumer:
-            with self.get_consumer(self.get_test_archive_file()) as consumer:
+            consumer.run()
-                consumer.run()
 
-        self._assert_first_last_send_progress(last_status="FAILED")
+        self.assertEqual(Document.objects.count(), 2)
+        self._assert_first_last_send_progress()
 
     def testDuplicates3(self):
         with self.get_consumer(self.get_test_archive_file()) as consumer:
@@ -513,9 +513,10 @@ class TestConsumer(
 
         Document.objects.all().delete()
 
-        with self.assertRaisesMessage(ConsumerError, "document is in the trash"):
+        with self.get_consumer(self.get_test_file()) as consumer:
-            with self.get_consumer(self.get_test_file()) as consumer:
+            consumer.run()
-                consumer.run()
+
+        self.assertEqual(Document.objects.count(), 1)
 
     def testAsnExists(self):
         with self.get_consumer(
@@ -718,12 +719,45 @@ class TestConsumer(
         dst = self.get_test_file()
         self.assertIsFile(dst)
 
-        with self.assertRaises(ConsumerError):
+        expected_message = (
+            f"{dst.name}: Not consuming {dst.name}: "
+            f"It is a duplicate of {document.title} (#{document.pk})"
+        )
+
+        with self.assertRaisesMessage(ConsumerError, expected_message):
             with self.get_consumer(dst) as consumer:
                 consumer.run()
 
         self.assertIsNotFile(dst)
-        self._assert_first_last_send_progress(last_status="FAILED")
+        self.assertEqual(Document.objects.count(), 1)
+        self._assert_first_last_send_progress(last_status=ProgressStatusOptions.FAILED)
+
+    @override_settings(CONSUMER_DELETE_DUPLICATES=True)
+    def test_delete_duplicate_in_trash(self):
+        dst = self.get_test_file()
+        with self.get_consumer(dst) as consumer:
+            consumer.run()
+
+        # Move the existing document to trash
+        document = Document.objects.first()
+        document.delete()
+
+        dst = self.get_test_file()
+        self.assertIsFile(dst)
+
+        expected_message = (
+            f"{dst.name}: Not consuming {dst.name}: "
+            f"It is a duplicate of {document.title} (#{document.pk})"
+            f" Note: existing document is in the trash."
+        )
+
+        with self.assertRaisesMessage(ConsumerError, expected_message):
+            with self.get_consumer(dst) as consumer:
+                consumer.run()
+
+        self.assertIsNotFile(dst)
+        self.assertEqual(Document.global_objects.count(), 1)
+        self.assertEqual(Document.objects.count(), 0)
 
     @override_settings(CONSUMER_DELETE_DUPLICATES=False)
     def test_no_delete_duplicate(self):
@@ -743,15 +777,12 @@ class TestConsumer(
         dst = self.get_test_file()
         self.assertIsFile(dst)
 
-        with self.assertRaisesRegex(
+        with self.get_consumer(dst) as consumer:
-            ConsumerError,
+            consumer.run()
-            r"sample\.pdf: Not consuming sample\.pdf: It is a duplicate of sample \(#\d+\)",
-        ):
-            with self.get_consumer(dst) as consumer:
-                consumer.run()
 
-        self.assertIsFile(dst)
+        self.assertIsNotFile(dst)
-        self._assert_first_last_send_progress(last_status="FAILED")
+        self.assertEqual(Document.objects.count(), 2)
+        self._assert_first_last_send_progress()
 
     @override_settings(FILENAME_FORMAT="{title}")
     @mock.patch("documents.parsers.document_consumer_declaration.send")