refactor: port ChildChunk (#30920)

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

api/services/dataset_service.py

@@ -3748,6 +3748,7 @@ class SegmentService:
                     ChildChunk.segment_id == segment.id,
                 )
             )
+            assert current_user.current_tenant_id
             child_chunk = ChildChunk(
                 tenant_id=current_user.current_tenant_id,
                 dataset_id=dataset.id,
@@ -3758,7 +3759,7 @@ class SegmentService:
                 index_node_hash=index_node_hash,
                 content=content,
                 word_count=len(content),
-                type="customized",
+                type=SegmentType.CUSTOMIZED,
                 created_by=current_user.id,
             )
             db.session.add(child_chunk)
@@ -3818,6 +3819,7 @@ class SegmentService:
             if new_child_chunks_args:
                 child_chunk_count = len(child_chunks)
                 for position, args in enumerate(new_child_chunks_args, start=child_chunk_count + 1):
+                    assert current_user.current_tenant_id
                     index_node_id = str(uuid.uuid4())
                     index_node_hash = helper.generate_text_hash(args.content)
                     child_chunk = ChildChunk(
@@ -3830,7 +3832,7 @@ class SegmentService:
                         index_node_hash=index_node_hash,
                         content=args.content,
                         word_count=len(args.content),
-                        type="customized",
+                        type=SegmentType.CUSTOMIZED,
                         created_by=current_user.id,
                     )
 

api/services/vector_service.py

@@ -16,6 +16,7 @@ from graphon.model_runtime.entities.model_entities import ModelType
 from models import UploadFile
 from models.dataset import ChildChunk, Dataset, DatasetProcessRule, DocumentSegment, SegmentAttachmentBinding
 from models.dataset import Document as DatasetDocument
+from models.enums import SegmentType
 
 logger = logging.getLogger(__name__)
 
@@ -178,7 +179,7 @@ class VectorService:
                     index_node_hash=child_chunk.metadata["doc_hash"],
                     content=child_chunk.page_content,
                     word_count=len(child_chunk.page_content),
-                    type="automatic",
+                    type=SegmentType.AUTOMATIC,
                     created_by=dataset_document.created_by,
                 )
                 db.session.add(child_segment)
@@ -222,6 +223,7 @@ class VectorService:
             )
             documents.append(new_child_document)
         for update_child_chunk in update_child_chunks:
+            assert update_child_chunk.index_node_id
             child_document = Document(
                 page_content=update_child_chunk.content,
                 metadata={
@@ -234,6 +236,7 @@ class VectorService:
             documents.append(child_document)
             delete_node_ids.append(update_child_chunk.index_node_id)
         for delete_child_chunk in delete_child_chunks:
+            assert delete_child_chunk.index_node_id
             delete_node_ids.append(delete_child_chunk.index_node_id)
         if dataset.indexing_technique == IndexTechniqueType.HIGH_QUALITY:
             # update vector index
@@ -246,6 +249,7 @@ class VectorService:
     @classmethod
     def delete_child_chunk_vector(cls, child_chunk: ChildChunk, dataset: Dataset):
         vector = Vector(dataset=dataset)
+        assert child_chunk.index_node_id
         vector.delete_by_ids([child_chunk.index_node_id])
 
     @classmethod