Chore: Backend dependencies update (#5676)

.github/workflows/ci.yml

@@ -16,7 +16,7 @@ on:
 env:
   # This is the version of pipenv all the steps will use
   # If changing this, change Dockerfile
-  DEFAULT_PIP_ENV_VERSION: "2023.11.15"
+  DEFAULT_PIP_ENV_VERSION: "2023.12.1"
   # This is the default version of Python to use in most steps which aren't specific
   DEFAULT_PYTHON_VERSION: "3.10"
 

.pre-commit-config.yaml

@@ -47,11 +47,11 @@ repos:
         exclude: "(^Pipfile\\.lock$)"
   # Python hooks
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: 'v0.1.11'
+    rev: 'v0.2.1'
     hooks:
       - id: ruff
   - repo: https://github.com/psf/black-pre-commit-mirror
-    rev: 23.12.1
+    rev: 24.1.1
     hooks:
       - id: black
   # Dockerfile hooks

.ruff.toml

@@ -1,5 +1,14 @@
+fix = true
+line-length = 88
+respect-gitignore = true
+src = ["src"]
+target-version = "py39"
+output-format = "grouped"
+show-fixes = true
+
 # https://docs.astral.sh/ruff/settings/
 # https://docs.astral.sh/ruff/rules/
+[lint]
 extend-select = [
   "W",     # https://docs.astral.sh/ruff/rules/#pycodestyle-e-w
   "I",     # https://docs.astral.sh/ruff/rules/#isort-i
@@ -25,15 +34,8 @@ extend-select = [
 ]
 # TODO PTH https://docs.astral.sh/ruff/rules/#flake8-use-pathlib-pth
 ignore = ["DJ001", "SIM105", "RUF012"]
-fix = true
-line-length = 88
-respect-gitignore = true
-src = ["src"]
-target-version = "py39"
-output-format = "grouped"
-show-fixes = true
 
-[per-file-ignores]
+[lint.per-file-ignores]
 ".github/scripts/*.py" = ["E501", "INP001", "SIM117"]
 "docker/wait-for-redis.py" = ["INP001", "T201"]
 "*/tests/*.py" = ["E501", "SIM117"]
@@ -41,5 +43,5 @@ show-fixes = true
 "src/paperless_tesseract/tests/test_parser.py" = ["RUF001"]
 "src/documents/models.py" = ["SIM115"]
 
-[isort]
+[lint.isort]
 force-single-line = true

Dockerfile

@@ -29,7 +29,7 @@ COPY Pipfile* ./
 
 RUN set -eux \
   && echo "Installing pipenv" \
-    && python3 -m pip install --no-cache-dir --upgrade pipenv==2023.11.15 \
+    && python3 -m pip install --no-cache-dir --upgrade pipenv==2023.12.1 \
   && echo "Generating requirement.txt" \
     && pipenv requirements > requirements.txt
 

Pipfile

@@ -7,7 +7,7 @@ name = "pypi"
 dateparser = "~=1.2"
 # WARNING: django does not use semver.
 #          Only patch versions are guaranteed to not introduce breaking changes.
-django = "~=4.2.9"
+django = "~=4.2.10"
 django-allauth = "*"
 django-auditlog = "*"
 django-celery-results = "*"
@@ -46,11 +46,11 @@ python-magic = "*"
 pyzbar = "*"
 rapidfuzz = "*"
 redis = {extras = ["hiredis"], version = "*"}
-scikit-learn = "~=1.3"
+scikit-learn = "~=1.4"
 setproctitle = "*"
 tika-client = "*"
 tqdm = "*"
-uvicorn = {extras = ["standard"], version = "*"}
+uvicorn = {extras = ["standard"], version = "==0.25.0"}
 watchdog = "~=3.0"
 whitenoise = "~=6.6"
 whoosh="~=2.7"

src/documents/serialisers.py

@@ -82,13 +82,14 @@ class MatchingModelSerializer(serializers.ModelSerializer):
 
     def validate(self, data):
         # TODO: remove pending https://github.com/encode/django-rest-framework/issues/7173
-        name = data["name"] if "name" in data else self.instance.name
+        name = data.get(
+            "name",
+            self.instance.name if hasattr(self.instance, "name") else None,
+        )
         owner = (
             data["owner"]
             if "owner" in data
-            else self.user
+            else self.user if hasattr(self, "user") else None
-            if hasattr(self, "user")
-            else None
         )
         pk = self.instance.pk if hasattr(self.instance, "pk") else None
         if ("name" in data or "owner" in data) and self.Meta.model.objects.filter(
@@ -261,7 +262,7 @@ class OwnedObjectSerializer(serializers.ModelSerializer, SetPermissionsMixin):
         if "set_permissions" in validated_data:
             self._set_permissions(validated_data["set_permissions"], instance)
         if "owner" in validated_data and "name" in self.Meta.fields:
-            name = validated_data["name"] if "name" in validated_data else instance.name
+            name = validated_data.get("name", instance.name)
             not_unique = (
                 self.Meta.model.objects.exclude(pk=instance.pk)
                 .filter(owner=validated_data["owner"], name=name)
@@ -443,7 +444,10 @@ class CustomFieldSerializer(serializers.ModelSerializer):
 
     def validate(self, attrs):
         # TODO: remove pending https://github.com/encode/django-rest-framework/issues/7173
-        name = attrs["name"] if "name" in attrs else self.instance.name
+        name = attrs.get(
+            "name",
+            self.instance.name if hasattr(self.instance, "name") else None,
+        )
         if ("name" in attrs) and self.Meta.model.objects.filter(
             name=name,
         ).exists():
@@ -697,10 +701,7 @@ class DocumentSerializer(
                             custom_field_instance.field,
                             doc_id,
                         )
-        if (
+        if validated_data.get("remove_inbox_tags"):
-            "remove_inbox_tags" in validated_data
-            and validated_data["remove_inbox_tags"]
-        ):
             tag_ids_being_added = (
                 [
                     tag.id
@@ -1352,7 +1353,7 @@ class BulkEditObjectPermissionsSerializer(serializers.Serializer, SetPermissions
     def validate(self, attrs):
         object_type = attrs["object_type"]
         objects = attrs["objects"]
-        permissions = attrs["permissions"] if "permissions" in attrs else None
+        permissions = attrs.get("permissions")
 
         self._validate_objects(objects, object_type)
         if permissions is not None:
@@ -1514,7 +1515,7 @@ class WorkflowSerializer(serializers.ModelSerializer):
             for trigger in triggers:
                 filter_has_tags = trigger.pop("filter_has_tags", None)
                 trigger_instance, _ = WorkflowTrigger.objects.update_or_create(
-                    id=trigger["id"] if "id" in trigger else None,
+                    id=trigger.get("id"),
                     defaults=trigger,
                 )
                 if filter_has_tags is not None:
@@ -1530,7 +1531,7 @@ class WorkflowSerializer(serializers.ModelSerializer):
                 assign_change_groups = action.pop("assign_change_groups", None)
                 assign_custom_fields = action.pop("assign_custom_fields", None)
                 action_instance, _ = WorkflowAction.objects.update_or_create(
-                    id=action["id"] if "id" in action else None,
+                    id=action.get("id"),
                     defaults=action,
                 )
                 if assign_tags is not None:

src/documents/signals/handlers.py

@@ -559,15 +559,21 @@ def run_workflow(
                     try:
                         document.title = parse_doc_title_w_placeholders(
                             action.assign_title,
+                            (
                                 document.correspondent.name
                                 if document.correspondent is not None
-                            else "",
+                                else ""
+                            ),
+                            (
                                 document.document_type.name
                                 if document.document_type is not None
-                            else "",
+                                else ""
+                            ),
+                            (
                                 document.owner.username
                                 if document.owner is not None
-                            else "",
+                                else ""
+                            ),
                             timezone.localtime(document.added),
                             document.original_filename,
                             timezone.localtime(document.created),

src/documents/views.py

@@ -175,16 +175,16 @@ class IndexView(TemplateView):
         context["full_name"] = self.request.user.get_full_name()
         context["styles_css"] = f"frontend/{self.get_frontend_language()}/styles.css"
         context["runtime_js"] = f"frontend/{self.get_frontend_language()}/runtime.js"
-        context[
+        context["polyfills_js"] = (
-            "polyfills_js"
+            f"frontend/{self.get_frontend_language()}/polyfills.js"
-        ] = f"frontend/{self.get_frontend_language()}/polyfills.js"
+        )
         context["main_js"] = f"frontend/{self.get_frontend_language()}/main.js"
-        context[
+        context["webmanifest"] = (
-            "webmanifest"
+            f"frontend/{self.get_frontend_language()}/manifest.webmanifest"
-        ] = f"frontend/{self.get_frontend_language()}/manifest.webmanifest"
+        )
-        context[
+        context["apple_touch_icon"] = (
-            "apple_touch_icon"
+            f"frontend/{self.get_frontend_language()}/apple-touch-icon.png"
-        ] = f"frontend/{self.get_frontend_language()}/apple-touch-icon.png"
+        )
         return context
 
 
@@ -722,9 +722,9 @@ class SearchResultSerializer(DocumentSerializer, PassUserMixin):
         r["__search_hit__"] = {
             "score": instance.score,
             "highlights": instance.highlights("content", text=doc.content),
-            "note_highlights": instance.highlights("notes", text=notes)
+            "note_highlights": (
-            if doc
+                instance.highlights("notes", text=notes) if doc else None
-            else None,
+            ),
             "rank": instance.rank,
         }
 

src/paperless/settings.py

@@ -1100,7 +1100,7 @@ def _get_nltk_language_setting(ocr_lang: str) -> Optional[str]:
         "tur": "turkish",
     }
 
-    return iso_code_to_nltk.get(ocr_lang, None)
+    return iso_code_to_nltk.get(ocr_lang)
 
 
 NLTK_ENABLED: Final[bool] = __get_boolean("PAPERLESS_ENABLE_NLTK", "yes")

src/paperless/wsgi.py

@@ -6,6 +6,7 @@ It exposes the WSGI callable as a module-level variable named ``application``.
 For more information on this file, see
 https://docs.djangoproject.com/en/1.10/howto/deployment/wsgi/
 """
+
 import os
 
 from django.core.wsgi import get_wsgi_application

src/paperless_mail/admin.py

@@ -9,11 +9,9 @@ from paperless_mail.models import ProcessedMail
 
 
 class MailAccountAdminForm(forms.ModelForm):
-
     """Metadata classes used by Django admin to display the form."""
 
     class Meta:
-
         """Metadata class used by Django admin to display the form."""
 
         model = MailAccount

src/paperless_mail/mail.py

@@ -739,9 +739,11 @@ class MailAccountHandler(LoggingMixin):
                     correspondent_id=correspondent.id if correspondent else None,
                     document_type_id=doc_type.id if doc_type else None,
                     tag_ids=tag_ids,
-                    owner_id=rule.owner.id
+                    owner_id=(
+                        rule.owner.id
                         if (rule.assign_owner_from_rule and rule.owner)
-                    else None,
+                        else None
+                    ),
                 )
 
                 consume_task = consume_file.s(

src/paperless_tesseract/parsers.py

@@ -205,9 +205,9 @@ class RasterisedDocumentParser(DocumentParser):
         }
 
         if "pdfa" in ocrmypdf_args["output_type"]:
-            ocrmypdf_args[
+            ocrmypdf_args["color_conversion_strategy"] = (
-                "color_conversion_strategy"
+                self.settings.color_conversion_strategy
-            ] = self.settings.color_conversion_strategy
+            )
 
         if self.settings.mode == ModeChoices.FORCE or safe_fallback:
             ocrmypdf_args["force_ocr"] = True