feat: Support multiple knowledge base binding in pipelines (#1766)

* Initial plan

* Add multi-knowledge base support to pipelines

- Created database migration dbm010 to convert knowledge-base field from string to array
- Updated default pipeline config to use knowledge-bases array
- Updated pipeline metadata to use knowledge-base-multi-selector type
- Modified localagent.py to retrieve from multiple knowledge bases and concatenate results
- Added KNOWLEDGE_BASE_MULTI_SELECTOR type to frontend form entities
- Implemented multi-selector UI component with dialog for selecting multiple knowledge bases

Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com>

* Add i18n translations for multi-knowledge base selector

Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com>

* Fix prettier formatting errors in DynamicFormItemComponent

Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com>

* Add accessibility attributes to knowledge base selector checkbox

Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com>

* fix: minor fix

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com>
Co-authored-by: Junyan Qin <rockchinq@gmail.com>

pkg/persistence/migrations/dbm010_pipeline_multi_knowledge_base.py

@@ -0,0 +1,88 @@
+from .. import migration
+
+import sqlalchemy
+
+from ...entity.persistence import pipeline as persistence_pipeline
+
+
+@migration.migration_class(10)
+class DBMigratePipelineMultiKnowledgeBase(migration.DBMigration):
+    """Pipeline support multiple knowledge base binding"""
+
+    async def upgrade(self):
+        """Upgrade"""
+        # read all pipelines
+        pipelines = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_pipeline.LegacyPipeline))
+
+        for pipeline in pipelines:
+            serialized_pipeline = self.ap.persistence_mgr.serialize_model(persistence_pipeline.LegacyPipeline, pipeline)
+
+            config = serialized_pipeline['config']
+
+            # Convert knowledge-base from string to array
+            if 'local-agent' in config['ai']:
+                current_kb = config['ai']['local-agent'].get('knowledge-base', '')
+                
+                # If it's already a list, skip
+                if isinstance(current_kb, list):
+                    continue
+                
+                # Convert string to list
+                if current_kb and current_kb != '__none__':
+                    config['ai']['local-agent']['knowledge-bases'] = [current_kb]
+                else:
+                    config['ai']['local-agent']['knowledge-bases'] = []
+                
+                # Remove old field
+                if 'knowledge-base' in config['ai']['local-agent']:
+                    del config['ai']['local-agent']['knowledge-base']
+
+            await self.ap.persistence_mgr.execute_async(
+                sqlalchemy.update(persistence_pipeline.LegacyPipeline)
+                .where(persistence_pipeline.LegacyPipeline.uuid == serialized_pipeline['uuid'])
+                .values(
+                    {
+                        'config': config,
+                        'for_version': self.ap.ver_mgr.get_current_version(),
+                    }
+                )
+            )
+
+    async def downgrade(self):
+        """Downgrade"""
+        # read all pipelines
+        pipelines = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_pipeline.LegacyPipeline))
+
+        for pipeline in pipelines:
+            serialized_pipeline = self.ap.persistence_mgr.serialize_model(persistence_pipeline.LegacyPipeline, pipeline)
+
+            config = serialized_pipeline['config']
+
+            # Convert knowledge-bases from array back to string
+            if 'local-agent' in config['ai']:
+                current_kbs = config['ai']['local-agent'].get('knowledge-bases', [])
+                
+                # If it's already a string, skip
+                if isinstance(current_kbs, str):
+                    continue
+                
+                # Convert list to string (take first one or empty)
+                if current_kbs and len(current_kbs) > 0:
+                    config['ai']['local-agent']['knowledge-base'] = current_kbs[0]
+                else:
+                    config['ai']['local-agent']['knowledge-base'] = ''
+                
+                # Remove new field
+                if 'knowledge-bases' in config['ai']['local-agent']:
+                    del config['ai']['local-agent']['knowledge-bases']
+
+            await self.ap.persistence_mgr.execute_async(
+                sqlalchemy.update(persistence_pipeline.LegacyPipeline)
+                .where(persistence_pipeline.LegacyPipeline.uuid == serialized_pipeline['uuid'])
+                .values(
+                    {
+                        'config': config,
+                        'for_version': self.ap.ver_mgr.get_current_version(),
+                    }
+                )
+            )

pkg/provider/runners/localagent.py

@@ -40,10 +40,14 @@ class LocalAgentRunner(runner.RequestRunner):
         """运行请求"""
         pending_tool_calls = []
 
-        kb_uuid = query.pipeline_config['ai']['local-agent']['knowledge-base']
-
-        if kb_uuid == '__none__':
-            kb_uuid = None
+        # Get knowledge bases list (new field)
+        kb_uuids = query.pipeline_config['ai']['local-agent'].get('knowledge-bases', [])
+        
+        # Fallback to old field for backward compatibility
+        if not kb_uuids:
+            old_kb_uuid = query.pipeline_config['ai']['local-agent'].get('knowledge-base', '')
+            if old_kb_uuid and old_kb_uuid != '__none__':
+                kb_uuids = [old_kb_uuid]
 
         user_message = copy.deepcopy(query.user_message)
 
@@ -57,21 +61,28 @@ class LocalAgentRunner(runner.RequestRunner):
                     user_message_text += ce.text
                     break
 
-        if kb_uuid and user_message_text:
+        if kb_uuids and user_message_text:
             # only support text for now
-            kb = await self.ap.rag_mgr.get_knowledge_base_by_uuid(kb_uuid)
+            all_results = []
+            
+            # Retrieve from each knowledge base
+            for kb_uuid in kb_uuids:
+                kb = await self.ap.rag_mgr.get_knowledge_base_by_uuid(kb_uuid)
 
-            if not kb:
-                self.ap.logger.warning(f'Knowledge base {kb_uuid} not found')
-                raise ValueError(f'Knowledge base {kb_uuid} not found')
+                if not kb:
+                    self.ap.logger.warning(f'Knowledge base {kb_uuid} not found, skipping')
+                    continue
 
-            result = await kb.retrieve(user_message_text, kb.knowledge_base_entity.top_k)
+                result = await kb.retrieve(user_message_text, kb.knowledge_base_entity.top_k)
+                
+                if result:
+                    all_results.extend(result)
 
             final_user_message_text = ''
 
-            if result:
+            if all_results:
                 rag_context = '\n\n'.join(
-                    f'[{i + 1}] {entry.metadata.get("text", "")}' for i, entry in enumerate(result)
+                    f'[{i + 1}] {entry.metadata.get("text", "")}' for i, entry in enumerate(all_results)
                 )
                 final_user_message_text = rag_combined_prompt_template.format(
                     rag_context=rag_context, user_message=user_message_text

pkg/utils/constants.py

@@ -1,6 +1,6 @@
 semantic_version = 'v4.4.1'
 
-required_database_version = 9
+required_database_version = 10
 """Tag the version of the database schema, used to check if the database needs to be migrated"""
 
 debug_mode = False