✅(views) be a bit more permissive for indexable document content

In indexation view, raise a validation error only when both the
title & content of a document are empty.

Signed-off-by: Fabre Florian <ffabre@hybird.org>

src/backend/core/schemas.py

@@ -23,11 +23,11 @@ class DocumentSchema(BaseModel):
     """Schema for validating the documents submitted to our API for indexing"""
 
     id: UUID4
-    title: Annotated[str, Field(max_length=300)]
+    title: Annotated[str, Field(max_length=300, min_length=0)]
     depth: Annotated[int, Field(ge=0)]
     path: Annotated[str, Field(max_length=300)]
     numchild: Annotated[int, Field(ge=0)]
-    content: str
+    content: Annotated[str, Field(min_length=0)]
     created_at: AwareDatetime
     updated_at: AwareDatetime
     size: Annotated[int, Field(ge=0, le=100 * 1024**3)]  # File size limited to 100GB
@@ -56,6 +56,12 @@ class DocumentSchema(BaseModel):
             raise ValueError(f"{info.field_name} must be earlier than now")
         return value
 
+    @model_validator(mode="after")
+    def check_empty_content(self):
+        if not self.title and not self.content:
+            raise ValueError('Either title or content should have at least 1 character')
+        return self
+
     @model_validator(mode="after")
     def check_update_at_after_created_at(self):
         """Date and time of last modification should be later than date and time of creation"""
@@ -83,7 +89,7 @@ class SearchQueryParametersSchema(BaseModel):
 
     q: str
     services: Union[str, List[str], None] = Field(default_factory=list)
-    visited: List[str] = Field(default_factory=list)
+    visited: Union[List[str], None] = Field(default_factory=list)
     reach: Optional[enums.ReachEnum] = None
     order_by: Optional[Literal[enums.ORDER_BY_OPTIONS]] = Field(default=enums.RELEVANCE)
     order_direction: Optional[Literal["asc", "desc"]] = Field(default="desc")

src/backend/core/tests/test_api_documents_index_bulk.py

@@ -334,3 +334,32 @@ def test_api_documents_index_bulk_datetime_future(field):
 
     for i in [1, 2]:
         assert responses[i]["status"] == "valid"
+
+
+def test_api_documents_index_empty_content_check():
+    """Test bulkk document indexing with both empty title & content."""
+    service = factories.ServiceFactory(name="test-service")
+    documents = factories.DocumentSchemaFactory.build_batch(3)
+
+    documents[0]['content'] = ''
+    documents[0]['title'] = ''
+
+    response = APIClient().post(
+        "/api/v1.0/documents/index/",
+        documents,
+        HTTP_AUTHORIZATION=f"Bearer {service.token:s}",
+        format="json",
+    )
+
+    assert response.status_code == 400
+    responses = response.json()
+    assert responses[0]["status"] == "error"
+    assert len(responses[0]["errors"]) == 1
+    assert (
+        responses[0]["errors"][0]["msg"]
+        == "Value error, Either title or content should have at least 1 character"
+    )
+    assert responses[0]["errors"][0]["type"] == "value_error"
+
+    for i in [1, 2]:
+        assert responses[i]["status"] == "valid"

src/backend/core/tests/test_api_documents_index_single.py

@@ -298,3 +298,23 @@ def test_api_documents_index_single_datetime_future(field):
     assert response.status_code == 400
     assert response.data[0]["msg"] == f"Value error, {field:s} must be earlier than now"
     assert response.data[0]["type"] == "value_error"
+
+
+def test_api_documents_index_empty_content_check():
+    """Test document indexing with both empty title & content."""
+    service = factories.ServiceFactory(name="test-service")
+    document = factories.DocumentSchemaFactory.build()
+
+    document['content'] = ''
+    document['title'] = ''
+
+    response = APIClient().post(
+        "/api/v1.0/documents/index/",
+        document,
+        HTTP_AUTHORIZATION=f"Bearer {service.token:s}",
+        format="json",
+    )
+
+    assert response.status_code == 400
+    assert response.data[0]["msg"] == "Value error, Either title or content should have at least 1 character"
+    assert response.data[0]["type"] == "value_error"