Fix/vdb index issue (#1776)

Co-authored-by: jyong <jyong@dify.ai>

api/controllers/console/app/annotation.py

     @setup_required
     @login_required
     @account_initialization_required
-    @cloud_edition_billing_resource_check('annotation')
     def delete(self, app_id, annotation_id):
         # The role of the current user in the ta table must be admin or owner
         if current_user.current_tenant.current_role not in ['admin', 'owner']:

api/controllers/console/wraps.py

                     abort(403, error_msg)
                 elif resource == 'workspace_custom' and not billing_info['can_replace_logo']:
                     abort(403, error_msg)
-                elif resource == 'annotation' and 0 < annotation_quota_limit['limit'] <= annotation_quota_limit['size']:
+                elif resource == 'annotation' and 0 < annotation_quota_limit['limit'] < annotation_quota_limit['size']:
                     abort(403, error_msg)
                 else:
                     return view(*args, **kwargs)

api/core/completion.py

                 vector_index = VectorIndex(
                     dataset=dataset,
                     config=current_app.config,
-                    embeddings=embeddings
+                    embeddings=embeddings,
+                    attributes=['doc_id', 'annotation_id', 'app_id']
                 )
 
                 documents = vector_index.search(

api/core/index/vector_index/milvus_vector_index.py

         """Only for created index."""
         if self._vector_store:
             return self._vector_store
-        attributes = ['doc_id', 'dataset_id', 'document_id']
 
         return MilvusVectorStore(
             collection_name=self.get_index_name(self.dataset),

api/core/index/vector_index/vector_index.py

 
 
 class VectorIndex:
-    def __init__(self, dataset: Dataset, config: dict, embeddings: Embeddings):
+    def __init__(self, dataset: Dataset, config: dict, embeddings: Embeddings,
+                 attributes: list = None):
+        if attributes is None:
+            attributes = ['doc_id', 'dataset_id', 'document_id', 'doc_hash']
         self._dataset = dataset
         self._embeddings = embeddings
-        self._vector_index = self._init_vector_index(dataset, config, embeddings)
+        self._vector_index = self._init_vector_index(dataset, config, embeddings, attributes)
+        self._attributes = attributes
 
-    def _init_vector_index(self, dataset: Dataset, config: dict, embeddings: Embeddings) -> BaseVectorIndex:
+    def _init_vector_index(self, dataset: Dataset, config: dict, embeddings: Embeddings,
+                           attributes: list) -> BaseVectorIndex:
         vector_type = config.get('VECTOR_STORE')
 
         if self._dataset.index_struct_dict:
                     api_key=config.get('WEAVIATE_API_KEY'),
                     batch_size=int(config.get('WEAVIATE_BATCH_SIZE'))
                 ),
-                embeddings=embeddings
+                embeddings=embeddings,
+                attributes=attributes
             )
         elif vector_type == "qdrant":
             from core.index.vector_index.qdrant_vector_index import QdrantVectorIndex, QdrantConfig

api/core/index/vector_index/weaviate_vector_index.py

 
 class WeaviateVectorIndex(BaseVectorIndex):
 
-    def __init__(self, dataset: Dataset, config: WeaviateConfig, embeddings: Embeddings):
+    def __init__(self, dataset: Dataset, config: WeaviateConfig, embeddings: Embeddings, attributes: list):
         super().__init__(dataset, embeddings)
         self._client = self._init_client(config)
+        self._attributes = attributes
 
     def _init_client(self, config: WeaviateConfig) -> weaviate.Client:
         auth_config = weaviate.auth.AuthApiKey(api_key=config.api_key)
         if self._vector_store:
             return self._vector_store
 
-        attributes = ['doc_id', 'dataset_id', 'document_id', 'doc_hash']
+        attributes = self._attributes
         if self._is_origin():
             attributes = ['doc_id']
 

api/tasks/annotation/add_annotation_to_index_task.py

             id=app_id,
             tenant_id=tenant_id,
             indexing_technique='high_quality',
+            embedding_model_provider=dataset_collection_binding.provider_name,
+            embedding_model=dataset_collection_binding.model_name,
             collection_binding_id=dataset_collection_binding.id
         )