Knowledge optimization (#3755)

Co-authored-by: crazywoola <427733928@qq.com>
Co-authored-by: JzoNg <jzongcode@gmail.com>

api/controllers/console/datasets/datasets.py

@@ -48,11 +48,14 @@ class DatasetListApi(Resource):
         limit = request.args.get('limit', default=20, type=int)
         ids = request.args.getlist('ids')
         provider = request.args.get('provider', default="vendor")
+        search = request.args.get('keyword', default=None, type=str)
+        tag_ids = request.args.getlist('tag_ids')
+
         if ids:
             datasets, total = DatasetService.get_datasets_by_ids(ids, current_user.current_tenant_id)
         else:
             datasets, total = DatasetService.get_datasets(page, limit, provider,
-                                                          current_user.current_tenant_id, current_user)
+                                                          current_user.current_tenant_id, current_user, search, tag_ids)
 
         # check embedding setting
         provider_manager = ProviderManager()
@@ -184,6 +187,10 @@ class DatasetApi(Resource):
                             help='Invalid indexing technique.')
         parser.add_argument('permission', type=str, location='json', choices=(
             'only_me', 'all_team_members'), help='Invalid permission.')
+        parser.add_argument('embedding_model', type=str,
+                            location='json', help='Invalid embedding model.')
+        parser.add_argument('embedding_model_provider', type=str,
+                            location='json', help='Invalid embedding model provider.')
         parser.add_argument('retrieval_model', type=dict, location='json', help='Invalid retrieval model.')
         args = parser.parse_args()
 
@@ -506,10 +513,27 @@ class DatasetRetrievalSettingMockApi(Resource):
         else:
             raise ValueError("Unsupported vector db type.")
 
+class DatasetErrorDocs(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, dataset_id):
+        dataset_id_str = str(dataset_id)
+        dataset = DatasetService.get_dataset(dataset_id_str)
+        if dataset is None:
+            raise NotFound("Dataset not found.")
+        results = DocumentService.get_error_documents_by_dataset_id(dataset_id_str)
+
+        return {
+            'data': [marshal(item, document_status_fields) for item in results],
+            'total': len(results)
+        }, 200
+
 
 api.add_resource(DatasetListApi, '/datasets')
 api.add_resource(DatasetApi, '/datasets/<uuid:dataset_id>')
 api.add_resource(DatasetQueryApi, '/datasets/<uuid:dataset_id>/queries')
+api.add_resource(DatasetErrorDocs, '/datasets/<uuid:dataset_id>/error-docs')
 api.add_resource(DatasetIndexingEstimateApi, '/datasets/indexing-estimate')
 api.add_resource(DatasetRelatedAppListApi, '/datasets/<uuid:dataset_id>/related-apps')
 api.add_resource(DatasetIndexingStatusApi, '/datasets/<uuid:dataset_id>/indexing-status')

api/controllers/console/datasets/datasets_document.py

@@ -1,3 +1,4 @@
+import logging
 from datetime import datetime, timezone
 
 from flask import request
@@ -233,7 +234,7 @@ class DatasetDocumentListApi(Resource):
                             location='json')
         parser.add_argument('data_source', type=dict, required=False, location='json')
         parser.add_argument('process_rule', type=dict, required=False, location='json')
-        parser.add_argument('duplicate', type=bool, nullable=False, location='json')
+        parser.add_argument('duplicate', type=bool, default=True, nullable=False, location='json')
         parser.add_argument('original_document_id', type=str, required=False, location='json')
         parser.add_argument('doc_form', type=str, default='text_model', required=False, nullable=False, location='json')
         parser.add_argument('doc_language', type=str, default='English', required=False, nullable=False,
@@ -883,6 +884,49 @@ class DocumentRecoverApi(DocumentResource):
         return {'result': 'success'}, 204
 
 
+class DocumentRetryApi(DocumentResource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def post(self, dataset_id):
+        """retry document."""
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('document_ids', type=list, required=True, nullable=False,
+                            location='json')
+        args = parser.parse_args()
+        dataset_id = str(dataset_id)
+        dataset = DatasetService.get_dataset(dataset_id)
+        retry_documents = []
+        if not dataset:
+            raise NotFound('Dataset not found.')
+        for document_id in args['document_ids']:
+            try:
+                document_id = str(document_id)
+
+                document = DocumentService.get_document(dataset.id, document_id)
+
+                # 404 if document not found
+                if document is None:
+                    raise NotFound("Document Not Exists.")
+
+                # 403 if document is archived
+                if DocumentService.check_archived(document):
+                    raise ArchivedDocumentImmutableError()
+
+                # 400 if document is completed
+                if document.indexing_status == 'completed':
+                    raise DocumentAlreadyFinishedError()
+                retry_documents.append(document)
+            except Exception as e:
+                logging.error(f"Document {document_id} retry failed: {str(e)}")
+                continue
+        # retry document
+        DocumentService.retry_document(dataset_id, retry_documents)
+
+        return {'result': 'success'}, 204
+
+
 api.add_resource(GetProcessRuleApi, '/datasets/process-rule')
 api.add_resource(DatasetDocumentListApi,
                  '/datasets/<uuid:dataset_id>/documents')
@@ -908,3 +952,4 @@ api.add_resource(DocumentStatusApi,
                  '/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/status/<string:action>')
 api.add_resource(DocumentPauseApi, '/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/processing/pause')
 api.add_resource(DocumentRecoverApi, '/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/processing/resume')
+api.add_resource(DocumentRetryApi, '/datasets/<uuid:dataset_id>/retry')