fix: Add the validation of doc_form in the Document-related service APIs.

api/controllers/console/datasets/datasets.py

@@ -187,6 +187,14 @@ class IndexingEstimatePayload(BaseModel):
             raise ValueError("indexing_technique is required.")
         return result
 
+    @field_validator("doc_form")
+    @classmethod
+    def validate_doc_form(cls, value: str) -> str:
+        result = _validate_doc_form(value)
+        if result is None:
+            return "text_model"
+        return result
+
 
 class ConsoleDatasetListQuery(BaseModel):
     page: int = Field(default=1, description="Page number")

api/models/dataset.py

@@ -19,6 +19,7 @@ from sqlalchemy.orm import Mapped, Session, mapped_column
 
 from configs import dify_config
 from core.rag.index_processor.constant.built_in_field import BuiltInField, MetadataDataSource
+from core.rag.index_processor.constant.index_type import IndexStructureType
 from core.rag.index_processor.constant.query_type import QueryType
 from core.rag.retrieval.retrieval_methods import RetrievalMethod
 from core.tools.signature import sign_upload_file
@@ -51,7 +52,7 @@ class Dataset(Base):
 
     INDEXING_TECHNIQUE_LIST = ["high_quality", "economy", None]
     PROVIDER_LIST = ["vendor", "external", None]
-    DOC_FORM_LIST = ["text_model", "qa_model", "hierarchical_model"]
+    DOC_FORM_LIST = [member.value for member in IndexStructureType]
 
     id: Mapped[str] = mapped_column(StringUUID, default=lambda: str(uuid4()))
     tenant_id: Mapped[str] = mapped_column(StringUUID)