MindsDB 25.3.3.0__py3-none-any.whl → 25.3.4.0__py3-none-any.whl

mindsdb/interfaces/knowledge_base/controller.py

@@ -26,9 +26,7 @@ from mindsdb.integrations.libs.vectordatabase_handler import (
 )
 from mindsdb.integrations.utilities.rag.rag_pipeline_builder import RAG
 from mindsdb.integrations.utilities.rag.config_loader import load_rag_config
-from mindsdb.integrations.utilities.sql_utils import (
-    extract_comparison_conditions, filter_dataframe, FilterCondition, FilterOperator
-)
+
 from mindsdb.interfaces.agents.constants import DEFAULT_EMBEDDINGS_MODEL_CLASS
 from mindsdb.interfaces.agents.langchain_agent import create_chat_model, get_llm_provider
 from mindsdb.interfaces.database.projects import ProjectController
@@ -105,19 +103,9 @@ class KnowledgeBaseTable:
         db_handler = self.get_vector_db()
         logger.debug(f"Using vector db handler: {type(db_handler)}")
 
-        vector_filters, outer_filters = [], []
-        # update vector handlers, mark conditions as applied inside
-        for op, arg1, arg2 in extract_comparison_conditions(query.where):
-            condition = FilterCondition(arg1, FilterOperator(op.upper()), arg2)
-            if arg1 in (TableField.ID.value, TableField.CONTENT.value, TableField.EMBEDDINGS.value):
-                vector_filters.append(condition)
-            else:
-                outer_filters.append([op, arg1, arg2])
-
-        df = db_handler.dispatch_select(query, conditions=vector_filters)
+        df = db_handler.dispatch_select(query)
 
         if df is not None:
-            df = filter_dataframe(df, outer_filters)
 
             logger.debug(f"Query returned {len(df)} rows")
             logger.debug(f"Columns in response: {df.columns.tolist()}")
@@ -229,7 +217,7 @@ class KnowledgeBaseTable:
 
         # send to vectordb
         db_handler = self.get_vector_db()
-        db_handler.query(query)
+        db_handler.dispatch_delete(query)
 
     def hybrid_search(
         self,
@@ -293,7 +281,6 @@ class KnowledgeBaseTable:
                         **base_metadata,
                         'original_row_id': str(row_id),
                         'content_column': col,
-                        'content_type': col.split('_')[-1] if '_' in col else 'text'
                     }
 
                     raw_documents.append(Document(
@@ -364,7 +351,7 @@ class KnowledgeBaseTable:
             logger.debug(f"Added IDs: {df_out[TableField.ID.value].tolist()}")
 
         # -- prepare content and metadata --
-        content_columns = params.get('content_columns')
+        content_columns = params.get('content_columns', [TableField.CONTENT.value])
         metadata_columns = params.get('metadata_columns')
 
         logger.debug(f"Processing with: content_columns={content_columns}, metadata_columns={metadata_columns}")
@@ -399,17 +386,6 @@ class KnowledgeBaseTable:
                 # all the rest columns
                 metadata_columns = list(set(columns).difference(content_columns))
 
-        elif metadata_columns is not None:
-            metadata_columns = list(set(metadata_columns).intersection(columns))
-            # use all unused columns is content
-            content_columns = list(set(columns).difference(metadata_columns))
-        elif TableField.METADATA.value in columns:
-            metadata_columns = [TableField.METADATA.value]
-            content_columns = list(set(columns).difference(metadata_columns))
-        else:
-            # all columns go to content
-            content_columns = columns
-
         # Add content columns directly (don't combine them)
         for col in content_columns:
             df_out[col] = df[col]
@@ -429,6 +405,9 @@ class KnowledgeBaseTable:
                         value = float(value)
                     elif pd.api.types.is_bool_dtype(value):
                         value = bool(value)
+                    elif isinstance(value, dict):
+                        metadata.update(value)
+                        continue
                     else:
                         value = str(value)
                     metadata[col] = value
@@ -852,6 +831,8 @@ class KnowledgeBaseController:
         # drop objects if they were created automatically
         if 'default_vector_storage' in kb.params:
             try:
+                handler = self.session.datahub.get(kb.params['default_vector_storage']).integration_handler
+                handler.drop_table(kb.vector_database_table)
                 self.session.integration_controller.delete(kb.params['default_vector_storage'])
             except EntityNotExistsError:
                 pass

mindsdb/interfaces/knowledge_base/preprocessing/document_preprocessor.py

@@ -98,7 +98,7 @@ class DocumentPreprocessor:
         provided_id: str = None,
     ) -> str:
         """Generate deterministic ID for a chunk"""
-        base_id = self._generate_deterministic_id(content, content_column, provided_id)
+        base_id = provided_id
         chunk_id = (
             f"{base_id}_chunk_{chunk_index}" if chunk_index is not None else base_id
         )

mindsdb/interfaces/skills/skills_controller.py

@@ -42,12 +42,12 @@ class SkillsController:
             db.Skills.deleted_at == null()
         ).first()
 
-    def get_skills(self, project_name: str) -> List[dict]:
+    def get_skills(self, project_name: Optional[str]) -> List[dict]:
         '''
         Gets all skills in a project.
 
         Parameters:
-            project_name (str): The name of the containing project
+            project_name (Optional[str]): The name of the containing project
 
         Returns:
             all_skills (List[db.Skills]): List of database skill object
@@ -56,11 +56,12 @@ class SkillsController:
             ValueError: If `project_name` does not exist
         '''
 
-        project_controller = ProjectController()
-        projects = project_controller.get_list()
-        if project_name is not None:
-            projects = list([p for p in projects if p.name == project_name])
-        project_ids = list([p.id for p in projects])
+        if project_name is None:
+            projects = self.project_controller.get_list()
+            project_ids = list([p.id for p in projects])
+        else:
+            project = self.project_controller.get(name=project_name)
+            project_ids = [project.id]
 
         query = (
             db.session.query(db.Skills)