Remove skip_asn stuff

.github/workflows/translate-strings.yml

@@ -12,9 +12,11 @@ jobs:
     steps:
       - name: Checkout code
         uses: actions/checkout@v6
+        env:
+          GH_REF: ${{ github.ref }} # sonar rule:githubactions:S7630 - avoid injection
         with:
           token: ${{ secrets.PNGX_BOT_PAT }}
-          ref: ${{ github.head_ref }}
+          ref: ${{ env.GH_REF }}
       - name: Set up Python
         id: setup-python
         uses: actions/setup-python@v6

src/documents/barcodes.py

@@ -186,11 +186,7 @@ class BarcodePlugin(ConsumeTaskPlugin):
 
         # Update/overwrite an ASN if possible
         # After splitting, as otherwise each split document gets the same ASN
-        if (
-            self.settings.barcode_enable_asn
-            and not self.metadata.skip_asn
-            and (located_asn := self.asn) is not None
-        ):
+        if self.settings.barcode_enable_asn and (located_asn := self.asn) is not None:
             logger.info(f"Found ASN in barcode: {located_asn}")
             self.metadata.asn = located_asn
 

src/documents/bulk_edit.py

@@ -7,7 +7,6 @@ from pathlib import Path
 from typing import TYPE_CHECKING
 from typing import Literal
 
-from celery import chain
 from celery import chord
 from celery import group
 from celery import shared_task
@@ -38,6 +37,42 @@ if TYPE_CHECKING:
 logger: logging.Logger = logging.getLogger("paperless.bulk_edit")
 
 
+@shared_task(bind=True)
+def restore_archive_serial_numbers_task(
+    self,
+    backup: dict[int, int],
+    *args,
+    **kwargs,
+) -> None:
+    restore_archive_serial_numbers(backup)
+
+
+def release_archive_serial_numbers(doc_ids: list[int]) -> dict[int, int]:
+    """
+    Clears ASNs on documents that are about to be replaced so new documents
+    can be assigned ASNs without uniqueness collisions. Returns a backup map
+    of doc_id -> previous ASN for potential restoration.
+    """
+    qs = Document.objects.filter(
+        id__in=doc_ids,
+        archive_serial_number__isnull=False,
+    ).only("pk", "archive_serial_number")
+    backup = dict(qs.values_list("pk", "archive_serial_number"))
+    qs.update(archive_serial_number=None)
+    logger.info(f"Released archive serial numbers for documents {list(backup.keys())}")
+    return backup
+
+
+def restore_archive_serial_numbers(backup: dict[int, int]) -> None:
+    """
+    Restores ASNs using the provided backup map, intended for
+    rollback when replacement consumption fails.
+    """
+    for doc_id, asn in backup.items():
+        Document.objects.filter(pk=doc_id).update(archive_serial_number=asn)
+    logger.info(f"Restored archive serial numbers for documents {list(backup.keys())}")
+
+
 def set_correspondent(
     doc_ids: list[int],
     correspondent: Correspondent,
@@ -386,6 +421,7 @@ def merge(
 
     merged_pdf = pikepdf.new()
     version: str = merged_pdf.pdf_version
+    handoff_asn: int | None = None
     # use doc_ids to preserve order
     for doc_id in doc_ids:
         doc = qs.get(id=doc_id)
@@ -401,6 +437,8 @@ def merge(
                 version = max(version, pdf.pdf_version)
                 merged_pdf.pages.extend(pdf.pages)
             affected_docs.append(doc.id)
+            if handoff_asn is None and doc.archive_serial_number is not None:
+                handoff_asn = doc.archive_serial_number
         except Exception as e:
             logger.exception(
                 f"Error merging document {doc.id}, it will not be included in the merge: {e}",
@@ -426,6 +464,8 @@ def merge(
                 DocumentMetadataOverrides.from_document(metadata_document)
             )
             overrides.title = metadata_document.title + " (merged)"
+            if metadata_document.archive_serial_number is not None:
+                handoff_asn = metadata_document.archive_serial_number
         else:
             overrides = DocumentMetadataOverrides()
     else:
@@ -433,8 +473,9 @@ def merge(
 
     if user is not None:
         overrides.owner_id = user.id
-    # Avoid copying or detecting ASN from merged PDFs to prevent collision
-    overrides.skip_asn = True
+
+    if delete_originals and handoff_asn is not None:
+        overrides.asn = handoff_asn
 
     logger.info("Adding merged document to the task queue.")
 
@@ -447,12 +488,20 @@ def merge(
     )
 
     if delete_originals:
+        backup = release_archive_serial_numbers(affected_docs)
         logger.info(
             "Queueing removal of original documents after consumption of merged document",
         )
-        chain(consume_task, delete.si(affected_docs)).delay()
-    else:
-        consume_task.delay()
+        try:
+            consume_task.apply_async(
+                link=[delete.si(affected_docs)],
+                link_error=[restore_archive_serial_numbers_task.s(backup)],
+            )
+        except Exception:
+            restore_archive_serial_numbers(backup)
+            raise
+        else:
+            consume_task.delay()
 
     return "OK"
 
@@ -508,10 +557,20 @@ def split(
                 )
 
             if delete_originals:
+                backup = release_archive_serial_numbers([doc.id])
                 logger.info(
                     "Queueing removal of original document after consumption of the split documents",
                 )
-                chord(header=consume_tasks, body=delete.si([doc.id])).delay()
+                try:
+                    chord(
+                        header=consume_tasks,
+                        body=delete.si([doc.id]),
+                    ).apply_async(
+                        link_error=[restore_archive_serial_numbers_task.s(backup)],
+                    )
+                except Exception:
+                    restore_archive_serial_numbers(backup)
+                    raise
             else:
                 group(consume_tasks).delay()
 
@@ -614,7 +673,8 @@ def edit_pdf(
             )
             if user is not None:
                 overrides.owner_id = user.id
-
+            if delete_original and len(pdf_docs) == 1:
+                overrides.asn = doc.archive_serial_number
             for idx, pdf in enumerate(pdf_docs, start=1):
                 filepath: Path = (
                     Path(tempfile.mkdtemp(dir=settings.SCRATCH_DIR))
@@ -633,7 +693,17 @@ def edit_pdf(
                 )
 
             if delete_original:
-                chord(header=consume_tasks, body=delete.si([doc.id])).delay()
+                backup = release_archive_serial_numbers([doc.id])
+                try:
+                    chord(
+                        header=consume_tasks,
+                        body=delete.si([doc.id]),
+                    ).apply_async(
+                        link_error=[restore_archive_serial_numbers_task.s(backup)],
+                    )
+                except Exception:
+                    restore_archive_serial_numbers(backup)
+                    raise
             else:
                 group(consume_tasks).delay()
 

src/documents/consumer.py

@@ -696,7 +696,7 @@ class ConsumerPlugin(
                 pk=self.metadata.storage_path_id,
             )
 
-        if self.metadata.asn is not None and not self.metadata.skip_asn:
+        if self.metadata.asn is not None:
             document.archive_serial_number = self.metadata.asn
 
         if self.metadata.owner_id:
@@ -812,8 +812,8 @@ class ConsumerPreflightPlugin(
         """
         Check that if override_asn is given, it is unique and within a valid range
         """
-        if self.metadata.skip_asn or self.metadata.asn is None:
-            # if skip is set or ASN is None
+        if self.metadata.asn is None:
+            # if ASN is None
             return
         # Validate the range is above zero and less than uint32_t max
         # otherwise, Whoosh can't handle it in the index

src/documents/data_models.py

@@ -30,7 +30,6 @@ class DocumentMetadataOverrides:
     change_users: list[int] | None = None
     change_groups: list[int] | None = None
     custom_fields: dict | None = None
-    skip_asn: bool = False
 
     def update(self, other: "DocumentMetadataOverrides") -> "DocumentMetadataOverrides":
         """
@@ -50,8 +49,6 @@ class DocumentMetadataOverrides:
             self.storage_path_id = other.storage_path_id
         if other.owner_id is not None:
             self.owner_id = other.owner_id
-        if other.skip_asn:
-            self.skip_asn = True
 
         # merge
         if self.tag_ids is None:

src/documents/serialisers.py

@@ -18,6 +18,8 @@ from django.core.exceptions import ValidationError
 from django.core.validators import DecimalValidator
 from django.core.validators import EmailValidator
 from django.core.validators import MaxLengthValidator
+from django.core.validators import MaxValueValidator
+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
@@ -875,6 +877,13 @@ class CustomFieldInstanceSerializer(serializers.ModelSerializer):
                 uri_validator(data["value"])
             elif field.data_type == CustomField.FieldDataType.INT:
                 integer_validator(data["value"])
+                try:
+                    value_int = int(data["value"])
+                except (TypeError, ValueError):
+                    raise serializers.ValidationError("Enter a valid integer.")
+                # Keep values within the PostgreSQL integer range
+                MinValueValidator(-2147483648)(value_int)
+                MaxValueValidator(2147483647)(value_int)
             elif (
                 field.data_type == CustomField.FieldDataType.MONETARY
                 and data["value"] != ""

src/documents/tests/test_api_documents.py

@@ -1664,6 +1664,44 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
 
         self.consume_file_mock.assert_not_called()
 
+    def test_patch_document_integer_custom_field_out_of_range(self):
+        """
+        GIVEN:
+            - An integer custom field
+            - A document
+        WHEN:
+            - Patching the document with an integer value exceeding PostgreSQL's range
+        THEN:
+            - HTTP 400 is returned (validation catches the overflow)
+            - No custom field instance is created
+        """
+        cf_int = CustomField.objects.create(
+            name="intfield",
+            data_type=CustomField.FieldDataType.INT,
+        )
+        doc = Document.objects.create(
+            title="Doc",
+            checksum="123",
+            mime_type="application/pdf",
+        )
+
+        response = self.client.patch(
+            f"/api/documents/{doc.pk}/",
+            {
+                "custom_fields": [
+                    {
+                        "field": cf_int.pk,
+                        "value": 2**31,  # overflow for PostgreSQL integer fields
+                    },
+                ],
+            },
+            format="json",
+        )
+
+        self.assertEqual(response.status_code, status.HTTP_400_BAD_REQUEST)
+        self.assertIn("custom_fields", response.data)
+        self.assertEqual(CustomFieldInstance.objects.count(), 0)
+
     def test_upload_with_webui_source(self):
         """
         GIVEN: A document with a source file

src/documents/tests/test_bulk_edit.py

@@ -602,23 +602,21 @@ class TestPDFActions(DirectoriesMixin, TestCase):
             expected_filename,
         )
         self.assertEqual(consume_file_args[1].title, None)
-        self.assertTrue(consume_file_args[1].skip_asn)
+        # No metadata_document_id, delete_originals False, so ASN should be None
+        self.assertIsNone(consume_file_args[1].asn)
 
         # With metadata_document_id overrides
         result = bulk_edit.merge(doc_ids, metadata_document_id=metadata_document_id)
         consume_file_args, _ = mock_consume_file.call_args
         self.assertEqual(consume_file_args[1].title, "B (merged)")
         self.assertEqual(consume_file_args[1].created, self.doc2.created)
-        self.assertTrue(consume_file_args[1].skip_asn)
 
         self.assertEqual(result, "OK")
 
     @mock.patch("documents.bulk_edit.delete.si")
     @mock.patch("documents.tasks.consume_file.s")
-    @mock.patch("documents.bulk_edit.chain")
     def test_merge_and_delete_originals(
         self,
-        mock_chain,
         mock_consume_file,
         mock_delete_documents,
     ):
@@ -632,6 +630,12 @@ class TestPDFActions(DirectoriesMixin, TestCase):
             - Document deletion task should be called
         """
         doc_ids = [self.doc1.id, self.doc2.id, self.doc3.id]
+        self.doc1.archive_serial_number = 101
+        self.doc2.archive_serial_number = 102
+        self.doc3.archive_serial_number = 103
+        self.doc1.save()
+        self.doc2.save()
+        self.doc3.save()
 
         result = bulk_edit.merge(doc_ids, delete_originals=True)
         self.assertEqual(result, "OK")
@@ -642,7 +646,8 @@ class TestPDFActions(DirectoriesMixin, TestCase):
 
         mock_consume_file.assert_called()
         mock_delete_documents.assert_called()
-        mock_chain.assert_called_once()
+        consume_sig = mock_consume_file.return_value
+        consume_sig.apply_async.assert_called_once()
 
         consume_file_args, _ = mock_consume_file.call_args
         self.assertEqual(
@@ -650,7 +655,7 @@ class TestPDFActions(DirectoriesMixin, TestCase):
             expected_filename,
         )
         self.assertEqual(consume_file_args[1].title, None)
-        self.assertTrue(consume_file_args[1].skip_asn)
+        self.assertEqual(consume_file_args[1].asn, 101)
 
         delete_documents_args, _ = mock_delete_documents.call_args
         self.assertEqual(
@@ -658,6 +663,13 @@ class TestPDFActions(DirectoriesMixin, TestCase):
             doc_ids,
         )
 
+        self.doc1.refresh_from_db()
+        self.doc2.refresh_from_db()
+        self.doc3.refresh_from_db()
+        self.assertIsNone(self.doc1.archive_serial_number)
+        self.assertIsNone(self.doc2.archive_serial_number)
+        self.assertIsNone(self.doc3.archive_serial_number)
+
     @mock.patch("documents.tasks.consume_file.s")
     def test_merge_with_archive_fallback(self, mock_consume_file):
         """
@@ -726,6 +738,7 @@ class TestPDFActions(DirectoriesMixin, TestCase):
         self.assertEqual(mock_consume_file.call_count, 2)
         consume_file_args, _ = mock_consume_file.call_args
         self.assertEqual(consume_file_args[1].title, "B (split 2)")
+        self.assertIsNone(consume_file_args[1].asn)
 
         self.assertEqual(result, "OK")
 
@@ -750,6 +763,8 @@ class TestPDFActions(DirectoriesMixin, TestCase):
         """
         doc_ids = [self.doc2.id]
         pages = [[1, 2], [3]]
+        self.doc2.archive_serial_number = 200
+        self.doc2.save()
 
         result = bulk_edit.split(doc_ids, pages, delete_originals=True)
         self.assertEqual(result, "OK")
@@ -767,6 +782,9 @@ class TestPDFActions(DirectoriesMixin, TestCase):
             doc_ids,
         )
 
+        self.doc2.refresh_from_db()
+        self.assertIsNone(self.doc2.archive_serial_number)
+
     @mock.patch("documents.tasks.consume_file.delay")
     @mock.patch("pikepdf.Pdf.save")
     def test_split_with_errors(self, mock_save_pdf, mock_consume_file):
@@ -967,10 +985,16 @@ class TestPDFActions(DirectoriesMixin, TestCase):
         mock_chord.return_value.delay.return_value = None
         doc_ids = [self.doc2.id]
         operations = [{"page": 1}, {"page": 2}]
+        self.doc2.archive_serial_number = 250
+        self.doc2.save()
 
         result = bulk_edit.edit_pdf(doc_ids, operations, delete_original=True)
         self.assertEqual(result, "OK")
         mock_chord.assert_called_once()
+        consume_file_args, _ = mock_consume_file.call_args
+        self.assertEqual(consume_file_args[1].asn, 250)
+        self.doc2.refresh_from_db()
+        self.assertIsNone(self.doc2.archive_serial_number)
 
     @mock.patch("documents.tasks.update_document_content_maybe_archive_file.delay")
     def test_edit_pdf_with_update_document(self, mock_update_document):

src/documents/tests/test_consumer.py

@@ -412,14 +412,6 @@ class TestConsumer(
         self.assertEqual(document.archive_serial_number, 123)
         self._assert_first_last_send_progress()
 
-    def testMetadataOverridesSkipAsnPropagation(self):
-        overrides = DocumentMetadataOverrides()
-        incoming = DocumentMetadataOverrides(skip_asn=True)
-
-        overrides.update(incoming)
-
-        self.assertTrue(overrides.skip_asn)
-
     def testOverrideTitlePlaceholders(self):
         c = Correspondent.objects.create(name="Correspondent Name")
         dt = DocumentType.objects.create(name="DocType Name")

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: 2025-12-24 05:27+0000\n"
+"POT-Creation-Date: 2025-12-29 14:49+0000\n"
 "PO-Revision-Date: 2022-02-17 04:17\n"
 "Last-Translator: \n"
 "Language-Team: English\n"
@@ -1219,35 +1219,35 @@ msgstr ""
 msgid "workflow runs"
 msgstr ""
 
-#: documents/serialisers.py:640
+#: documents/serialisers.py:642
 msgid "Invalid color."
 msgstr ""
 
-#: documents/serialisers.py:1826
+#: documents/serialisers.py:1835
 #, python-format
 msgid "File type %(type)s not supported"
 msgstr ""
 
-#: documents/serialisers.py:1870
+#: documents/serialisers.py:1879
 #, python-format
 msgid "Custom field id must be an integer: %(id)s"
 msgstr ""
 
-#: documents/serialisers.py:1877
+#: documents/serialisers.py:1886
 #, python-format
 msgid "Custom field with id %(id)s does not exist"
 msgstr ""
 
-#: documents/serialisers.py:1894 documents/serialisers.py:1904
+#: documents/serialisers.py:1903 documents/serialisers.py:1913
 msgid ""
 "Custom fields must be a list of integers or an object mapping ids to values."
 msgstr ""
 
-#: documents/serialisers.py:1899
+#: documents/serialisers.py:1908
 msgid "Some custom fields don't exist or were specified twice."
 msgstr ""
 
-#: documents/serialisers.py:2014
+#: documents/serialisers.py:2023
 msgid "Invalid variable detected."
 msgstr ""