Chore(deps): Bump the pre-commit-dependencies group with 3 updates

Bumps the pre-commit-dependencies group with 3 updates: [https://github.com/astral-sh/ruff-pre-commit](https://github.com/astral-sh/ruff-pre-commit), [https://github.com/tox-dev/pyproject-fmt](https://github.com/tox-dev/pyproject-fmt) and [https://github.com/AleksaC/hadolint-py](https://github.com/AleksaC/hadolint-py).


Updates `https://github.com/astral-sh/ruff-pre-commit` from v0.16.1 to 0.16.4
- [Release notes](https://github.com/astral-sh/ruff-pre-commit/releases)
- [Commits](https://github.com/astral-sh/ruff-pre-commit/compare/v0.16.1...v0.16.4)

Updates `https://github.com/tox-dev/pyproject-fmt` from v2.26.0 to 2.28.1
- [Release notes](https://github.com/tox-dev/pyproject-fmt/releases)
- [Commits](https://github.com/tox-dev/pyproject-fmt/compare/v2.26.0...v2.28.1)

Updates `https://github.com/AleksaC/hadolint-py` from v2.14.0 to 2.15.1
- [Commits](https://github.com/AleksaC/hadolint-py/compare/v2.14.0...v2.15.1)

---
updated-dependencies:
- dependency-name: https://github.com/astral-sh/ruff-pre-commit
  dependency-version: 0.16.4
  dependency-type: direct:production
  dependency-group: pre-commit-dependencies
- dependency-name: https://github.com/tox-dev/pyproject-fmt
  dependency-version: 2.28.1
  dependency-type: direct:production
  dependency-group: pre-commit-dependencies
- dependency-name: https://github.com/AleksaC/hadolint-py
  dependency-version: 2.15.1
  dependency-type: direct:production
  dependency-group: pre-commit-dependencies
...

Signed-off-by: dependabot[bot] <support@github.com>

.pre-commit-config.yaml

@@ -50,18 +50,18 @@ repos:
           - 'prettier-plugin-organize-imports@4.3.0'
   # Python hooks
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.16.1
+    rev: v0.16.4
     hooks:
       - id: ruff-check
       - id: ruff-format
   - repo: https://github.com/tox-dev/pyproject-fmt
-    rev: "v2.26.0"
+    rev: "v2.28.1"
     hooks:
       - id: pyproject-fmt
         additional_dependencies: [tomli]
   # Dockerfile hooks
   - repo: https://github.com/AleksaC/hadolint-py
-    rev: v2.14.0
+    rev: v2.15.1
     hooks:
       - id: hadolint
   # Shell script hooks

docs/configuration.md

@@ -2088,6 +2088,12 @@ password. All of these options come from their similarly-named [Django settings]
 
     Defaults to "always".
 
+#### [`PAPERLESS_REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS=<bool>`](#PAPERLESS_REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS) {#PAPERLESS_REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS}
+
+: If set to false, Paperless blocks remote OCR endpoint URLs that resolve to non-public addresses (e.g., localhost, etc).
+
+    Defaults to True.
+
 ## AI {#ai}
 
 #### [`PAPERLESS_AI_ENABLED=<bool>`](#PAPERLESS_AI_ENABLED) {#PAPERLESS_AI_ENABLED}

src/documents/tests/test_api_app_config.py

@@ -1063,3 +1063,79 @@ class TestApiAppConfig(DirectoriesMixin, APITestCase):
         )
         self.assertEqual(response.status_code, status.HTTP_400_BAD_REQUEST)
         self.assertIn("non-public address", str(response.data).lower())
+
+    @override_settings(REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS=False)
+    def test_update_remote_ocr_endpoint_blocks_internal_endpoint_when_disallowed(
+        self,
+    ) -> None:
+        """
+        GIVEN:
+            - Internal remote OCR endpoints are disallowed
+        WHEN:
+            - The config is updated with a remote OCR endpoint resolving internally
+        THEN:
+            - The request is rejected
+        """
+        response = self.client.patch(
+            f"{self.ENDPOINT}1/",
+            json.dumps(
+                {
+                    "remote_ocr_endpoint": "http://127.0.0.1:5000",
+                },
+            ),
+            content_type="application/json",
+        )
+        self.assertEqual(response.status_code, status.HTTP_400_BAD_REQUEST)
+        self.assertIn("non-public address", str(response.data).lower())
+
+    @override_settings(REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS=True)
+    def test_update_remote_ocr_endpoint_allows_internal_endpoint_by_default(
+        self,
+    ) -> None:
+        """
+        GIVEN:
+            - Internal remote OCR endpoints are allowed (the default)
+        WHEN:
+            - The config is updated with a remote OCR endpoint resolving internally
+        THEN:
+            - The request is accepted, preserving existing self-hosted deployments
+        """
+        response = self.client.patch(
+            f"{self.ENDPOINT}1/",
+            json.dumps(
+                {
+                    "remote_ocr_endpoint": "http://127.0.0.1:5000",
+                },
+            ),
+            content_type="application/json",
+        )
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(
+            response.data["remote_ocr_endpoint"],
+            "http://127.0.0.1:5000",
+        )
+
+    @override_settings(REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS=False)
+    def test_update_remote_ocr_endpoint_empty_value_skips_validation(
+        self,
+    ) -> None:
+        """
+        GIVEN:
+            - Internal remote OCR endpoints are disallowed
+        WHEN:
+            - The config is updated with an empty remote OCR endpoint
+        THEN:
+            - The request is accepted; clearing the field never needs
+              outbound URL validation
+        """
+        response = self.client.patch(
+            f"{self.ENDPOINT}1/",
+            json.dumps(
+                {
+                    "remote_ocr_endpoint": "",
+                },
+            ),
+            content_type="application/json",
+        )
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(response.data["remote_ocr_endpoint"], "")

src/paperless/parsers/remote.py

@@ -32,6 +32,8 @@ if TYPE_CHECKING:
     import datetime
     from types import TracebackType
 
+    from azure.core.pipeline import PipelineRequest
+
     from paperless.parsers import MetadataEntry
     from paperless.parsers import ParserContext
 
@@ -436,9 +438,45 @@ class RemoteDocumentParser:
         from azure.ai.documentintelligence.models import DocumentContentFormat
         from azure.core.credentials import AzureKeyCredential
 
+        from paperless.network import validate_outbound_http_url
+
+        allow_internal = settings.REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS
+
+        try:
+            validate_outbound_http_url(config.endpoint, allow_internal=allow_internal)
+        except ValueError as e:
+            raise ParseError(f"Invalid remote OCR endpoint: {e}") from e
+
+        def _revalidate_request_host(request: PipelineRequest) -> None:
+            """Re-validates the destination host of every request sent.
+
+            The check above only covers the moment the client is built. A
+            single analysis involves several requests spread over the
+            polling loop below, and any one of them can be redirected.
+            Wiring this through ``raw_request_hook`` (Azure's built-in
+            CustomHookPolicy) rather than a custom policy means it runs
+            *after* RedirectPolicy in the pipeline, so it sees - and
+            re-checks - every actual outbound URL, including redirect
+            targets, not just the original request.
+            """
+            validate_outbound_http_url(
+                request.http_request.url,
+                allow_internal=allow_internal,
+            )
+
         client = DocumentIntelligenceClient(
             endpoint=config.endpoint,
             credential=AzureKeyCredential(config.api_key),
+            raw_request_hook=_revalidate_request_host,
+            # AzureKeyCredential is sent as Ocp-Apim-Subscription-Key, which
+            # Azure's default SensitiveHeaderCleanupPolicy does not strip on
+            # a cross-domain redirect (only Authorization and
+            # x-ms-authorization-auxiliary are, by default).
+            blocked_redirect_headers=[
+                "Authorization",
+                "x-ms-authorization-auxiliary",
+                "Ocp-Apim-Subscription-Key",
+            ],
         )
 
         try:

src/paperless/serialisers.py

@@ -305,6 +305,22 @@ class ApplicationConfigurationSerializer(
 
     validate_llm_embedding_endpoint = validate_llm_endpoint
 
+    def validate_remote_ocr_endpoint(self, value: str | None) -> str | None:
+        if not value:
+            return value
+
+        try:
+            validate_outbound_http_url(
+                value,
+                allow_internal=settings.REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS,
+            )
+        except ValueError as e:
+            raise serializers.ValidationError(
+                f"Invalid remote OCR endpoint: {e.args[0]}, see logs for details",
+            ) from e
+
+        return value
+
     class Meta:
         model = ApplicationConfiguration
         fields = "__all__"

src/paperless/settings/__init__.py

@@ -1208,6 +1208,10 @@ REMOTE_OCR_MODE = get_choice_from_env(
     {"always", "workflow_only"},
     default="always",
 )
+REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS = get_bool_from_env(
+    "PAPERLESS_REMOTE_OCR_ALLOW_INTERNAL_ENDPOINTS",
+    "true",
+)
 
 ################################################################################
 # AI Settings                                                                  #