Feature: allow duplicates with warnings, UI for discovery (#11815)

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,
-            result="Not consuming task_one.pdf: It is a duplicate of task_one_existing.pdf (#1234).",
+            status=celery.states.SUCCESS,
+            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:
-                consumer.run()
+        with self.get_consumer(self.get_test_file()) as consumer:
+            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:
-                consumer.run()
+        with self.get_consumer(self.get_test_archive_file()) as consumer:
+            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:
-                consumer.run()
+        with self.get_consumer(self.get_test_file()) as consumer:
+            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(
-            ConsumerError,
-            r"sample\.pdf: Not consuming sample\.pdf: It is a duplicate of sample \(#\d+\)",
-        ):
-            with self.get_consumer(dst) as consumer:
-                consumer.run()
+        with self.get_consumer(dst) as consumer:
+            consumer.run()
 
-        self.assertIsFile(dst)
-        self._assert_first_last_send_progress(last_status="FAILED")
+        self.assertIsNotFile(dst)
+        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")