diff --git a/api/README.md b/api/README.md index 5ecf92a4f0..e75ea3d354 100644 --- a/api/README.md +++ b/api/README.md @@ -80,10 +80,10 @@ 1. If you need to handle and debug the async tasks (e.g. dataset importing and documents indexing), please start the worker service. ```bash -uv run celery -A app.celery worker -P gevent -c 1 --loglevel INFO -Q dataset,generation,mail,ops_trace,app_deletion,plugin,workflow_storage,conversation +uv run celery -A app.celery worker -P gevent -c 2 --loglevel INFO -Q dataset,generation,mail,ops_trace,app_deletion,plugin,workflow_storage,conversation ``` -Addition, if you want to debug the celery scheduled tasks, you can use the following command in another terminal: +Additionally, if you want to debug the celery scheduled tasks, you can run the following command in another terminal to start the beat service: ```bash uv run celery -A app.celery beat diff --git a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_datasets.py b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_datasets.py index 34faa4ec85..c741bfbf82 100644 --- a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_datasets.py +++ b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_datasets.py @@ -4,7 +4,7 @@ from sqlalchemy.orm import Session from werkzeug.exceptions import Forbidden import services -from controllers.console import api +from controllers.console import console_ns from controllers.console.datasets.error import DatasetNameDuplicateError from controllers.console.wraps import ( account_initialization_required, @@ -32,6 +32,7 @@ def _validate_description_length(description): return description +@console_ns.route("/rag/pipeline/dataset") class CreateRagPipelineDatasetApi(Resource): @setup_required @login_required @@ -84,6 +85,7 @@ class CreateRagPipelineDatasetApi(Resource): return import_info, 201 +@console_ns.route("/rag/pipeline/empty-dataset") class CreateEmptyRagPipelineDatasetApi(Resource): @setup_required @login_required @@ -108,7 +110,3 @@ class CreateEmptyRagPipelineDatasetApi(Resource): ), ) return marshal(dataset, dataset_detail_fields), 201 - - -api.add_resource(CreateRagPipelineDatasetApi, "/rag/pipeline/dataset") -api.add_resource(CreateEmptyRagPipelineDatasetApi, "/rag/pipeline/empty-dataset") diff --git a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_draft_variable.py b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_draft_variable.py index db07e7729a..38f75402a8 100644 --- a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_draft_variable.py +++ b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_draft_variable.py @@ -6,7 +6,7 @@ from flask_restx import Resource, fields, inputs, marshal, marshal_with, reqpars from sqlalchemy.orm import Session from werkzeug.exceptions import Forbidden -from controllers.console import api +from controllers.console import console_ns from controllers.console.app.error import ( DraftWorkflowNotExist, ) @@ -111,6 +111,7 @@ def _api_prerequisite(f): return wrapper +@console_ns.route("/rag/pipelines//workflows/draft/variables") class RagPipelineVariableCollectionApi(Resource): @_api_prerequisite @marshal_with(_WORKFLOW_DRAFT_VARIABLE_LIST_WITHOUT_VALUE_FIELDS) @@ -168,6 +169,7 @@ def validate_node_id(node_id: str) -> NoReturn | None: return None +@console_ns.route("/rag/pipelines//workflows/draft/nodes//variables") class RagPipelineNodeVariableCollectionApi(Resource): @_api_prerequisite @marshal_with(_WORKFLOW_DRAFT_VARIABLE_LIST_FIELDS) @@ -190,6 +192,7 @@ class RagPipelineNodeVariableCollectionApi(Resource): return Response("", 204) +@console_ns.route("/rag/pipelines//workflows/draft/variables/") class RagPipelineVariableApi(Resource): _PATCH_NAME_FIELD = "name" _PATCH_VALUE_FIELD = "value" @@ -284,6 +287,7 @@ class RagPipelineVariableApi(Resource): return Response("", 204) +@console_ns.route("/rag/pipelines//workflows/draft/variables//reset") class RagPipelineVariableResetApi(Resource): @_api_prerequisite def put(self, pipeline: Pipeline, variable_id: str): @@ -325,6 +329,7 @@ def _get_variable_list(pipeline: Pipeline, node_id) -> WorkflowDraftVariableList return draft_vars +@console_ns.route("/rag/pipelines//workflows/draft/system-variables") class RagPipelineSystemVariableCollectionApi(Resource): @_api_prerequisite @marshal_with(_WORKFLOW_DRAFT_VARIABLE_LIST_FIELDS) @@ -332,6 +337,7 @@ class RagPipelineSystemVariableCollectionApi(Resource): return _get_variable_list(pipeline, SYSTEM_VARIABLE_NODE_ID) +@console_ns.route("/rag/pipelines//workflows/draft/environment-variables") class RagPipelineEnvironmentVariableCollectionApi(Resource): @_api_prerequisite def get(self, pipeline: Pipeline): @@ -364,26 +370,3 @@ class RagPipelineEnvironmentVariableCollectionApi(Resource): ) return {"items": env_vars_list} - - -api.add_resource( - RagPipelineVariableCollectionApi, - "/rag/pipelines//workflows/draft/variables", -) -api.add_resource( - RagPipelineNodeVariableCollectionApi, - "/rag/pipelines//workflows/draft/nodes//variables", -) -api.add_resource( - RagPipelineVariableApi, "/rag/pipelines//workflows/draft/variables/" -) -api.add_resource( - RagPipelineVariableResetApi, "/rag/pipelines//workflows/draft/variables//reset" -) -api.add_resource( - RagPipelineSystemVariableCollectionApi, "/rag/pipelines//workflows/draft/system-variables" -) -api.add_resource( - RagPipelineEnvironmentVariableCollectionApi, - "/rag/pipelines//workflows/draft/environment-variables", -) diff --git a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_import.py b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_import.py index a447f2848a..e0b918456b 100644 --- a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_import.py +++ b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_import.py @@ -5,7 +5,7 @@ from flask_restx import Resource, marshal_with, reqparse # type: ignore from sqlalchemy.orm import Session from werkzeug.exceptions import Forbidden -from controllers.console import api +from controllers.console import console_ns from controllers.console.datasets.wraps import get_rag_pipeline from controllers.console.wraps import ( account_initialization_required, @@ -20,6 +20,7 @@ from services.app_dsl_service import ImportStatus from services.rag_pipeline.rag_pipeline_dsl_service import RagPipelineDslService +@console_ns.route("/rag/pipelines/imports") class RagPipelineImportApi(Resource): @setup_required @login_required @@ -66,6 +67,7 @@ class RagPipelineImportApi(Resource): return result.model_dump(mode="json"), 200 +@console_ns.route("/rag/pipelines/imports//confirm") class RagPipelineImportConfirmApi(Resource): @setup_required @login_required @@ -90,6 +92,7 @@ class RagPipelineImportConfirmApi(Resource): return result.model_dump(mode="json"), 200 +@console_ns.route("/rag/pipelines/imports//check-dependencies") class RagPipelineImportCheckDependenciesApi(Resource): @setup_required @login_required @@ -107,6 +110,7 @@ class RagPipelineImportCheckDependenciesApi(Resource): return result.model_dump(mode="json"), 200 +@console_ns.route("/rag/pipelines//exports") class RagPipelineExportApi(Resource): @setup_required @login_required @@ -128,22 +132,3 @@ class RagPipelineExportApi(Resource): ) return {"data": result}, 200 - - -# Import Rag Pipeline -api.add_resource( - RagPipelineImportApi, - "/rag/pipelines/imports", -) -api.add_resource( - RagPipelineImportConfirmApi, - "/rag/pipelines/imports//confirm", -) -api.add_resource( - RagPipelineImportCheckDependenciesApi, - "/rag/pipelines/imports//check-dependencies", -) -api.add_resource( - RagPipelineExportApi, - "/rag/pipelines//exports", -) diff --git a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_workflow.py b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_workflow.py index 01ddb8a871..a75c121fbe 100644 --- a/api/controllers/console/datasets/rag_pipeline/rag_pipeline_workflow.py +++ b/api/controllers/console/datasets/rag_pipeline/rag_pipeline_workflow.py @@ -9,7 +9,7 @@ from sqlalchemy.orm import Session from werkzeug.exceptions import Forbidden, InternalServerError, NotFound import services -from controllers.console import api +from controllers.console import console_ns from controllers.console.app.error import ( ConversationCompletedError, DraftWorkflowNotExist, @@ -50,6 +50,7 @@ from services.rag_pipeline.rag_pipeline_transform_service import RagPipelineTran logger = logging.getLogger(__name__) +@console_ns.route("/rag/pipelines//workflows/draft") class DraftRagPipelineApi(Resource): @setup_required @login_required @@ -147,6 +148,7 @@ class DraftRagPipelineApi(Resource): } +@console_ns.route("/rag/pipelines//workflows/draft/iteration/nodes//run") class RagPipelineDraftRunIterationNodeApi(Resource): @setup_required @login_required @@ -181,6 +183,7 @@ class RagPipelineDraftRunIterationNodeApi(Resource): raise InternalServerError() +@console_ns.route("/rag/pipelines//workflows/draft/loop/nodes//run") class RagPipelineDraftRunLoopNodeApi(Resource): @setup_required @login_required @@ -215,6 +218,7 @@ class RagPipelineDraftRunLoopNodeApi(Resource): raise InternalServerError() +@console_ns.route("/rag/pipelines//workflows/draft/run") class DraftRagPipelineRunApi(Resource): @setup_required @login_required @@ -249,6 +253,7 @@ class DraftRagPipelineRunApi(Resource): raise InvokeRateLimitHttpError(ex.description) +@console_ns.route("/rag/pipelines//workflows/published/run") class PublishedRagPipelineRunApi(Resource): @setup_required @login_required @@ -369,6 +374,7 @@ class PublishedRagPipelineRunApi(Resource): # # return result # +@console_ns.route("/rag/pipelines//workflows/published/datasource/nodes//run") class RagPipelinePublishedDatasourceNodeRunApi(Resource): @setup_required @login_required @@ -411,6 +417,7 @@ class RagPipelinePublishedDatasourceNodeRunApi(Resource): ) +@console_ns.route("/rag/pipelines//workflows/draft/datasource/nodes//run") class RagPipelineDraftDatasourceNodeRunApi(Resource): @setup_required @login_required @@ -453,6 +460,7 @@ class RagPipelineDraftDatasourceNodeRunApi(Resource): ) +@console_ns.route("/rag/pipelines//workflows/draft/nodes//run") class RagPipelineDraftNodeRunApi(Resource): @setup_required @login_required @@ -486,6 +494,7 @@ class RagPipelineDraftNodeRunApi(Resource): return workflow_node_execution +@console_ns.route("/rag/pipelines//workflow-runs/tasks//stop") class RagPipelineTaskStopApi(Resource): @setup_required @login_required @@ -504,6 +513,7 @@ class RagPipelineTaskStopApi(Resource): return {"result": "success"} +@console_ns.route("/rag/pipelines//workflows/publish") class PublishedRagPipelineApi(Resource): @setup_required @login_required @@ -559,6 +569,7 @@ class PublishedRagPipelineApi(Resource): } +@console_ns.route("/rag/pipelines//workflows/default-workflow-block-configs") class DefaultRagPipelineBlockConfigsApi(Resource): @setup_required @login_required @@ -577,6 +588,7 @@ class DefaultRagPipelineBlockConfigsApi(Resource): return rag_pipeline_service.get_default_block_configs() +@console_ns.route("/rag/pipelines//workflows/default-workflow-block-configs/") class DefaultRagPipelineBlockConfigApi(Resource): @setup_required @login_required @@ -608,6 +620,7 @@ class DefaultRagPipelineBlockConfigApi(Resource): return rag_pipeline_service.get_default_block_config(node_type=block_type, filters=filters) +@console_ns.route("/rag/pipelines//workflows") class PublishedAllRagPipelineApi(Resource): @setup_required @login_required @@ -656,6 +669,7 @@ class PublishedAllRagPipelineApi(Resource): } +@console_ns.route("/rag/pipelines//workflows/") class RagPipelineByIdApi(Resource): @setup_required @login_required @@ -713,6 +727,7 @@ class RagPipelineByIdApi(Resource): return workflow +@console_ns.route("/rag/pipelines//workflows/published/processing/parameters") class PublishedRagPipelineSecondStepApi(Resource): @setup_required @login_required @@ -738,6 +753,7 @@ class PublishedRagPipelineSecondStepApi(Resource): } +@console_ns.route("/rag/pipelines//workflows/published/pre-processing/parameters") class PublishedRagPipelineFirstStepApi(Resource): @setup_required @login_required @@ -763,6 +779,7 @@ class PublishedRagPipelineFirstStepApi(Resource): } +@console_ns.route("/rag/pipelines//workflows/draft/pre-processing/parameters") class DraftRagPipelineFirstStepApi(Resource): @setup_required @login_required @@ -788,6 +805,7 @@ class DraftRagPipelineFirstStepApi(Resource): } +@console_ns.route("/rag/pipelines//workflows/draft/processing/parameters") class DraftRagPipelineSecondStepApi(Resource): @setup_required @login_required @@ -814,6 +832,7 @@ class DraftRagPipelineSecondStepApi(Resource): } +@console_ns.route("/rag/pipelines//workflow-runs") class RagPipelineWorkflowRunListApi(Resource): @setup_required @login_required @@ -835,6 +854,7 @@ class RagPipelineWorkflowRunListApi(Resource): return result +@console_ns.route("/rag/pipelines//workflow-runs/") class RagPipelineWorkflowRunDetailApi(Resource): @setup_required @login_required @@ -853,6 +873,7 @@ class RagPipelineWorkflowRunDetailApi(Resource): return workflow_run +@console_ns.route("/rag/pipelines//workflow-runs//node-executions") class RagPipelineWorkflowRunNodeExecutionListApi(Resource): @setup_required @login_required @@ -876,6 +897,7 @@ class RagPipelineWorkflowRunNodeExecutionListApi(Resource): return {"data": node_executions} +@console_ns.route("/rag/pipelines/datasource-plugins") class DatasourceListApi(Resource): @setup_required @login_required @@ -891,6 +913,7 @@ class DatasourceListApi(Resource): return jsonable_encoder(RagPipelineManageService.list_rag_pipeline_datasources(tenant_id)) +@console_ns.route("/rag/pipelines//workflows/draft/nodes//last-run") class RagPipelineWorkflowLastRunApi(Resource): @setup_required @login_required @@ -912,6 +935,7 @@ class RagPipelineWorkflowLastRunApi(Resource): return node_exec +@console_ns.route("/rag/pipelines/transform/datasets/") class RagPipelineTransformApi(Resource): @setup_required @login_required @@ -929,6 +953,7 @@ class RagPipelineTransformApi(Resource): return result +@console_ns.route("/rag/pipelines//workflows/draft/datasource/variables-inspect") class RagPipelineDatasourceVariableApi(Resource): @setup_required @login_required @@ -958,6 +983,7 @@ class RagPipelineDatasourceVariableApi(Resource): return workflow_node_execution +@console_ns.route("/rag/pipelines/recommended-plugins") class RagPipelineRecommendedPluginApi(Resource): @setup_required @login_required @@ -966,114 +992,3 @@ class RagPipelineRecommendedPluginApi(Resource): rag_pipeline_service = RagPipelineService() recommended_plugins = rag_pipeline_service.get_recommended_plugins() return recommended_plugins - - -api.add_resource( - DraftRagPipelineApi, - "/rag/pipelines//workflows/draft", -) -api.add_resource( - DraftRagPipelineRunApi, - "/rag/pipelines//workflows/draft/run", -) -api.add_resource( - PublishedRagPipelineRunApi, - "/rag/pipelines//workflows/published/run", -) -api.add_resource( - RagPipelineTaskStopApi, - "/rag/pipelines//workflow-runs/tasks//stop", -) -api.add_resource( - RagPipelineDraftNodeRunApi, - "/rag/pipelines//workflows/draft/nodes//run", -) -api.add_resource( - RagPipelinePublishedDatasourceNodeRunApi, - "/rag/pipelines//workflows/published/datasource/nodes//run", -) - -api.add_resource( - RagPipelineDraftDatasourceNodeRunApi, - "/rag/pipelines//workflows/draft/datasource/nodes//run", -) - -api.add_resource( - RagPipelineDraftRunIterationNodeApi, - "/rag/pipelines//workflows/draft/iteration/nodes//run", -) - -api.add_resource( - RagPipelineDraftRunLoopNodeApi, - "/rag/pipelines//workflows/draft/loop/nodes//run", -) - -api.add_resource( - PublishedRagPipelineApi, - "/rag/pipelines//workflows/publish", -) -api.add_resource( - PublishedAllRagPipelineApi, - "/rag/pipelines//workflows", -) -api.add_resource( - DefaultRagPipelineBlockConfigsApi, - "/rag/pipelines//workflows/default-workflow-block-configs", -) -api.add_resource( - DefaultRagPipelineBlockConfigApi, - "/rag/pipelines//workflows/default-workflow-block-configs/", -) -api.add_resource( - RagPipelineByIdApi, - "/rag/pipelines//workflows/", -) -api.add_resource( - RagPipelineWorkflowRunListApi, - "/rag/pipelines//workflow-runs", -) -api.add_resource( - RagPipelineWorkflowRunDetailApi, - "/rag/pipelines//workflow-runs/", -) -api.add_resource( - RagPipelineWorkflowRunNodeExecutionListApi, - "/rag/pipelines//workflow-runs//node-executions", -) -api.add_resource( - DatasourceListApi, - "/rag/pipelines/datasource-plugins", -) -api.add_resource( - PublishedRagPipelineSecondStepApi, - "/rag/pipelines//workflows/published/processing/parameters", -) -api.add_resource( - PublishedRagPipelineFirstStepApi, - "/rag/pipelines//workflows/published/pre-processing/parameters", -) -api.add_resource( - DraftRagPipelineSecondStepApi, - "/rag/pipelines//workflows/draft/processing/parameters", -) -api.add_resource( - DraftRagPipelineFirstStepApi, - "/rag/pipelines//workflows/draft/pre-processing/parameters", -) -api.add_resource( - RagPipelineWorkflowLastRunApi, - "/rag/pipelines//workflows/draft/nodes//last-run", -) -api.add_resource( - RagPipelineTransformApi, - "/rag/pipelines/transform/datasets/", -) -api.add_resource( - RagPipelineDatasourceVariableApi, - "/rag/pipelines//workflows/draft/datasource/variables-inspect", -) - -api.add_resource( - RagPipelineRecommendedPluginApi, - "/rag/pipelines/recommended-plugins", -) diff --git a/api/controllers/service_api/dataset/dataset.py b/api/controllers/service_api/dataset/dataset.py index 6a70345f7c..72ab05cec0 100644 --- a/api/controllers/service_api/dataset/dataset.py +++ b/api/controllers/service_api/dataset/dataset.py @@ -1,10 +1,10 @@ -from typing import Literal +from typing import Any, Literal, cast from flask import request from flask_restx import marshal, reqparse from werkzeug.exceptions import Forbidden, NotFound -import services.dataset_service +import services from controllers.service_api import service_api_ns from controllers.service_api.dataset.error import DatasetInUseError, DatasetNameDuplicateError, InvalidActionError from controllers.service_api.wraps import ( @@ -254,19 +254,21 @@ class DatasetListApi(DatasetApiResource): """Resource for creating datasets.""" args = dataset_create_parser.parse_args() - if args.get("embedding_model_provider"): - DatasetService.check_embedding_model_setting( - tenant_id, args.get("embedding_model_provider"), args.get("embedding_model") - ) + embedding_model_provider = args.get("embedding_model_provider") + embedding_model = args.get("embedding_model") + if embedding_model_provider and embedding_model: + DatasetService.check_embedding_model_setting(tenant_id, embedding_model_provider, embedding_model) + + retrieval_model = args.get("retrieval_model") if ( - args.get("retrieval_model") - and args.get("retrieval_model").get("reranking_model") - and args.get("retrieval_model").get("reranking_model").get("reranking_provider_name") + retrieval_model + and retrieval_model.get("reranking_model") + and retrieval_model.get("reranking_model").get("reranking_provider_name") ): DatasetService.check_reranking_model_setting( tenant_id, - args.get("retrieval_model").get("reranking_model").get("reranking_provider_name"), - args.get("retrieval_model").get("reranking_model").get("reranking_model_name"), + retrieval_model.get("reranking_model").get("reranking_provider_name"), + retrieval_model.get("reranking_model").get("reranking_model_name"), ) try: @@ -317,7 +319,7 @@ class DatasetApi(DatasetApiResource): DatasetService.check_dataset_permission(dataset, current_user) except services.errors.account.NoPermissionError as e: raise Forbidden(str(e)) - data = marshal(dataset, dataset_detail_fields) + data = cast(dict[str, Any], marshal(dataset, dataset_detail_fields)) # check embedding setting provider_manager = ProviderManager() assert isinstance(current_user, Account) @@ -331,8 +333,8 @@ class DatasetApi(DatasetApiResource): for embedding_model in embedding_models: model_names.append(f"{embedding_model.model}:{embedding_model.provider.provider}") - if data["indexing_technique"] == "high_quality": - item_model = f"{data['embedding_model']}:{data['embedding_model_provider']}" + if data.get("indexing_technique") == "high_quality": + item_model = f"{data.get('embedding_model')}:{data.get('embedding_model_provider')}" if item_model in model_names: data["embedding_available"] = True else: @@ -341,7 +343,9 @@ class DatasetApi(DatasetApiResource): data["embedding_available"] = True # force update search method to keyword_search if indexing_technique is economic - data["retrieval_model_dict"]["search_method"] = "keyword_search" + retrieval_model_dict = data.get("retrieval_model_dict") + if retrieval_model_dict: + retrieval_model_dict["search_method"] = "keyword_search" if data.get("permission") == "partial_members": part_users_list = DatasetPermissionService.get_dataset_partial_member_list(dataset_id_str) @@ -372,19 +376,24 @@ class DatasetApi(DatasetApiResource): data = request.get_json() # check embedding model setting - if data.get("indexing_technique") == "high_quality" or data.get("embedding_model_provider"): - DatasetService.check_embedding_model_setting( - dataset.tenant_id, data.get("embedding_model_provider"), data.get("embedding_model") - ) + embedding_model_provider = data.get("embedding_model_provider") + embedding_model = data.get("embedding_model") + if data.get("indexing_technique") == "high_quality" or embedding_model_provider: + if embedding_model_provider and embedding_model: + DatasetService.check_embedding_model_setting( + dataset.tenant_id, embedding_model_provider, embedding_model + ) + + retrieval_model = data.get("retrieval_model") if ( - data.get("retrieval_model") - and data.get("retrieval_model").get("reranking_model") - and data.get("retrieval_model").get("reranking_model").get("reranking_provider_name") + retrieval_model + and retrieval_model.get("reranking_model") + and retrieval_model.get("reranking_model").get("reranking_provider_name") ): DatasetService.check_reranking_model_setting( dataset.tenant_id, - data.get("retrieval_model").get("reranking_model").get("reranking_provider_name"), - data.get("retrieval_model").get("reranking_model").get("reranking_model_name"), + retrieval_model.get("reranking_model").get("reranking_provider_name"), + retrieval_model.get("reranking_model").get("reranking_model_name"), ) # The role of the current user in the ta table must be admin, owner, editor, or dataset_operator @@ -397,7 +406,7 @@ class DatasetApi(DatasetApiResource): if dataset is None: raise NotFound("Dataset not found.") - result_data = marshal(dataset, dataset_detail_fields) + result_data = cast(dict[str, Any], marshal(dataset, dataset_detail_fields)) assert isinstance(current_user, Account) tenant_id = current_user.current_tenant_id @@ -591,9 +600,10 @@ class DatasetTagsApi(DatasetApiResource): args = tag_update_parser.parse_args() args["type"] = "knowledge" - tag = TagService.update_tags(args, args.get("tag_id")) + tag_id = args["tag_id"] + tag = TagService.update_tags(args, tag_id) - binding_count = TagService.get_tag_binding_count(args.get("tag_id")) + binding_count = TagService.get_tag_binding_count(tag_id) response = {"id": tag.id, "name": tag.name, "type": tag.type, "binding_count": binding_count} @@ -616,7 +626,7 @@ class DatasetTagsApi(DatasetApiResource): if not current_user.has_edit_permission: raise Forbidden() args = tag_delete_parser.parse_args() - TagService.delete_tag(args.get("tag_id")) + TagService.delete_tag(args["tag_id"]) return 204 diff --git a/api/controllers/service_api/dataset/document.py b/api/controllers/service_api/dataset/document.py index e01bc8940c..c1122acd7b 100644 --- a/api/controllers/service_api/dataset/document.py +++ b/api/controllers/service_api/dataset/document.py @@ -108,19 +108,21 @@ class DocumentAddByTextApi(DatasetApiResource): if text is None or name is None: raise ValueError("Both 'text' and 'name' must be non-null values.") - if args.get("embedding_model_provider"): - DatasetService.check_embedding_model_setting( - tenant_id, args.get("embedding_model_provider"), args.get("embedding_model") - ) + embedding_model_provider = args.get("embedding_model_provider") + embedding_model = args.get("embedding_model") + if embedding_model_provider and embedding_model: + DatasetService.check_embedding_model_setting(tenant_id, embedding_model_provider, embedding_model) + + retrieval_model = args.get("retrieval_model") if ( - args.get("retrieval_model") - and args.get("retrieval_model").get("reranking_model") - and args.get("retrieval_model").get("reranking_model").get("reranking_provider_name") + retrieval_model + and retrieval_model.get("reranking_model") + and retrieval_model.get("reranking_model").get("reranking_provider_name") ): DatasetService.check_reranking_model_setting( tenant_id, - args.get("retrieval_model").get("reranking_model").get("reranking_provider_name"), - args.get("retrieval_model").get("reranking_model").get("reranking_model_name"), + retrieval_model.get("reranking_model").get("reranking_provider_name"), + retrieval_model.get("reranking_model").get("reranking_model_name"), ) if not current_user: @@ -187,15 +189,16 @@ class DocumentUpdateByTextApi(DatasetApiResource): if not dataset: raise ValueError("Dataset does not exist.") + retrieval_model = args.get("retrieval_model") if ( - args.get("retrieval_model") - and args.get("retrieval_model").get("reranking_model") - and args.get("retrieval_model").get("reranking_model").get("reranking_provider_name") + retrieval_model + and retrieval_model.get("reranking_model") + and retrieval_model.get("reranking_model").get("reranking_provider_name") ): DatasetService.check_reranking_model_setting( tenant_id, - args.get("retrieval_model").get("reranking_model").get("reranking_provider_name"), - args.get("retrieval_model").get("reranking_model").get("reranking_model_name"), + retrieval_model.get("reranking_model").get("reranking_provider_name"), + retrieval_model.get("reranking_model").get("reranking_model_name"), ) # indexing_technique is already set in dataset since this is an update diff --git a/api/controllers/service_api/dataset/metadata.py b/api/controllers/service_api/dataset/metadata.py index c6032048e6..e01659dc68 100644 --- a/api/controllers/service_api/dataset/metadata.py +++ b/api/controllers/service_api/dataset/metadata.py @@ -106,7 +106,7 @@ class DatasetMetadataServiceApi(DatasetApiResource): raise NotFound("Dataset not found.") DatasetService.check_dataset_permission(dataset, current_user) - metadata = MetadataService.update_metadata_name(dataset_id_str, metadata_id_str, args.get("name")) + metadata = MetadataService.update_metadata_name(dataset_id_str, metadata_id_str, args["name"]) return marshal(metadata, dataset_metadata_fields), 200 @service_api_ns.doc("delete_dataset_metadata") diff --git a/api/core/workflow/entities/variable_pool.py b/api/core/workflow/entities/variable_pool.py index 8ceabde7e6..2dc00fd70b 100644 --- a/api/core/workflow/entities/variable_pool.py +++ b/api/core/workflow/entities/variable_pool.py @@ -184,11 +184,22 @@ class VariablePool(BaseModel): """Extract the actual value from an ObjectSegment.""" return obj.value if isinstance(obj, ObjectSegment) else obj - def _get_nested_attribute(self, obj: Mapping[str, Any], attr: str): - """Get a nested attribute from a dictionary-like object.""" - if not isinstance(obj, dict): + def _get_nested_attribute(self, obj: Mapping[str, Any], attr: str) -> Segment | None: + """ + Get a nested attribute from a dictionary-like object. + + Args: + obj: The dictionary-like object to search. + attr: The key to look up. + + Returns: + Segment | None: + The corresponding Segment built from the attribute value if the key exists, + otherwise None. + """ + if not isinstance(obj, dict) or attr not in obj: return None - return obj.get(attr) + return variable_factory.build_segment(obj.get(attr)) def remove(self, selector: Sequence[str], /): """ diff --git a/api/core/workflow/nodes/iteration/iteration_node.py b/api/core/workflow/nodes/iteration/iteration_node.py index 1a417b5739..a05a6b1b96 100644 --- a/api/core/workflow/nodes/iteration/iteration_node.py +++ b/api/core/workflow/nodes/iteration/iteration_node.py @@ -10,6 +10,8 @@ from typing_extensions import TypeIs from core.variables import IntegerVariable, NoneSegment from core.variables.segments import ArrayAnySegment, ArraySegment +from core.variables.variables import VariableUnion +from core.workflow.constants import CONVERSATION_VARIABLE_NODE_ID from core.workflow.entities import VariablePool from core.workflow.enums import ( ErrorStrategy, @@ -217,6 +219,13 @@ class IterationNode(Node): graph_engine=graph_engine, ) + # Sync conversation variables after each iteration completes + self._sync_conversation_variables_from_snapshot( + self._extract_conversation_variable_snapshot( + variable_pool=graph_engine.graph_runtime_state.variable_pool + ) + ) + # Update the total tokens from this iteration self.graph_runtime_state.total_tokens += graph_engine.graph_runtime_state.total_tokens iter_run_map[str(index)] = (datetime.now(UTC).replace(tzinfo=None) - iter_start_at).total_seconds() @@ -235,7 +244,10 @@ class IterationNode(Node): with ThreadPoolExecutor(max_workers=max_workers) as executor: # Submit all iteration tasks - future_to_index: dict[Future[tuple[datetime, list[GraphNodeEventBase], object | None, int]], int] = {} + future_to_index: dict[ + Future[tuple[datetime, list[GraphNodeEventBase], object | None, int, dict[str, VariableUnion]]], + int, + ] = {} for index, item in enumerate(iterator_list_value): yield IterationNextEvent(index=index) future = executor.submit( @@ -252,7 +264,7 @@ class IterationNode(Node): index = future_to_index[future] try: result = future.result() - iter_start_at, events, output_value, tokens_used = result + iter_start_at, events, output_value, tokens_used, conversation_snapshot = result # Update outputs at the correct index outputs[index] = output_value @@ -264,6 +276,9 @@ class IterationNode(Node): self.graph_runtime_state.total_tokens += tokens_used iter_run_map[str(index)] = (datetime.now(UTC).replace(tzinfo=None) - iter_start_at).total_seconds() + # Sync conversation variables after iteration completion + self._sync_conversation_variables_from_snapshot(conversation_snapshot) + except Exception as e: # Handle errors based on error_handle_mode match self._node_data.error_handle_mode: @@ -288,7 +303,7 @@ class IterationNode(Node): item: object, flask_app: Flask, context_vars: contextvars.Context, - ) -> tuple[datetime, list[GraphNodeEventBase], object | None, int]: + ) -> tuple[datetime, list[GraphNodeEventBase], object | None, int, dict[str, VariableUnion]]: """Execute a single iteration in parallel mode and return results.""" with preserve_flask_contexts(flask_app=flask_app, context_vars=context_vars): iter_start_at = datetime.now(UTC).replace(tzinfo=None) @@ -307,8 +322,17 @@ class IterationNode(Node): # Get the output value from the temporary outputs list output_value = outputs_temp[0] if outputs_temp else None + conversation_snapshot = self._extract_conversation_variable_snapshot( + variable_pool=graph_engine.graph_runtime_state.variable_pool + ) - return iter_start_at, events, output_value, graph_engine.graph_runtime_state.total_tokens + return ( + iter_start_at, + events, + output_value, + graph_engine.graph_runtime_state.total_tokens, + conversation_snapshot, + ) def _handle_iteration_success( self, @@ -430,6 +454,23 @@ class IterationNode(Node): return variable_mapping + def _extract_conversation_variable_snapshot(self, *, variable_pool: VariablePool) -> dict[str, VariableUnion]: + conversation_variables = variable_pool.variable_dictionary.get(CONVERSATION_VARIABLE_NODE_ID, {}) + return {name: variable.model_copy(deep=True) for name, variable in conversation_variables.items()} + + def _sync_conversation_variables_from_snapshot(self, snapshot: dict[str, VariableUnion]) -> None: + parent_pool = self.graph_runtime_state.variable_pool + parent_conversations = parent_pool.variable_dictionary.get(CONVERSATION_VARIABLE_NODE_ID, {}) + + current_keys = set(parent_conversations.keys()) + snapshot_keys = set(snapshot.keys()) + + for removed_key in current_keys - snapshot_keys: + parent_pool.remove((CONVERSATION_VARIABLE_NODE_ID, removed_key)) + + for name, variable in snapshot.items(): + parent_pool.add((CONVERSATION_VARIABLE_NODE_ID, name), variable) + def _append_iteration_info_to_event( self, event: GraphNodeEventBase, diff --git a/api/extensions/ext_celery.py b/api/extensions/ext_celery.py index 585539e2ce..6d7d81ed87 100644 --- a/api/extensions/ext_celery.py +++ b/api/extensions/ext_celery.py @@ -145,6 +145,7 @@ def init_app(app: DifyApp) -> Celery: } if dify_config.ENABLE_CHECK_UPGRADABLE_PLUGIN_TASK and dify_config.MARKETPLACE_ENABLED: imports.append("schedule.check_upgradable_plugin_task") + imports.append("tasks.process_tenant_plugin_autoupgrade_check_task") beat_schedule["check_upgradable_plugin_task"] = { "task": "schedule.check_upgradable_plugin_task.check_upgradable_plugin_task", "schedule": crontab(minute="*/15"), diff --git a/api/factories/variable_factory.py b/api/factories/variable_factory.py index 2104e66254..494194369a 100644 --- a/api/factories/variable_factory.py +++ b/api/factories/variable_factory.py @@ -142,6 +142,8 @@ def build_segment(value: Any, /) -> Segment: # below if value is None: return NoneSegment() + if isinstance(value, Segment): + return value if isinstance(value, str): return StringSegment(value=value) if isinstance(value, bool): diff --git a/api/pyproject.toml b/api/pyproject.toml index 012702edd2..485fb918ed 100644 --- a/api/pyproject.toml +++ b/api/pyproject.toml @@ -180,7 +180,7 @@ dev = [ storage = [ "azure-storage-blob==12.13.0", "bce-python-sdk~=0.9.23", - "cos-python-sdk-v5==1.9.30", + "cos-python-sdk-v5==1.9.38", "esdk-obs-python==3.24.6.1", "google-cloud-storage==2.16.0", "opendal~=0.46.0", @@ -207,7 +207,7 @@ vdb = [ "couchbase~=4.3.0", "elasticsearch==8.14.0", "opensearch-py==2.4.0", - "oracledb==3.0.0", + "oracledb==3.3.0", "pgvecto-rs[sqlalchemy]~=0.2.1", "pgvector==0.2.5", "pymilvus~=2.5.0", diff --git a/api/pyrightconfig.json b/api/pyrightconfig.json index 61ed3ac3b4..f3d56cfbee 100644 --- a/api/pyrightconfig.json +++ b/api/pyrightconfig.json @@ -8,7 +8,6 @@ "extensions", "libs", "controllers/console/datasets", - "controllers/service_api/dataset", "core/ops", "core/tools", "core/model_runtime", diff --git a/api/schedule/check_upgradable_plugin_task.py b/api/schedule/check_upgradable_plugin_task.py index a9ad27b059..0712100c01 100644 --- a/api/schedule/check_upgradable_plugin_task.py +++ b/api/schedule/check_upgradable_plugin_task.py @@ -6,7 +6,7 @@ import click import app from extensions.ext_database import db from models.account import TenantPluginAutoUpgradeStrategy -from tasks.process_tenant_plugin_autoupgrade_check_task import process_tenant_plugin_autoupgrade_check_task +from tasks import process_tenant_plugin_autoupgrade_check_task as check_task AUTO_UPGRADE_MINIMAL_CHECKING_INTERVAL = 15 * 60 # 15 minutes MAX_CONCURRENT_CHECK_TASKS = 20 @@ -43,7 +43,7 @@ def check_upgradable_plugin_task(): for i in range(0, total_strategies, MAX_CONCURRENT_CHECK_TASKS): batch_strategies = strategies[i : i + MAX_CONCURRENT_CHECK_TASKS] for strategy in batch_strategies: - process_tenant_plugin_autoupgrade_check_task.delay( + check_task.process_tenant_plugin_autoupgrade_check_task.delay( strategy.tenant_id, strategy.strategy_setting, strategy.upgrade_time_of_day, diff --git a/api/tasks/process_tenant_plugin_autoupgrade_check_task.py b/api/tasks/process_tenant_plugin_autoupgrade_check_task.py index bae8f1c4db..124971e8e2 100644 --- a/api/tasks/process_tenant_plugin_autoupgrade_check_task.py +++ b/api/tasks/process_tenant_plugin_autoupgrade_check_task.py @@ -1,5 +1,5 @@ +import json import operator -import traceback import typing import click @@ -9,38 +9,106 @@ from core.helper import marketplace from core.helper.marketplace import MarketplacePluginDeclaration from core.plugin.entities.plugin import PluginInstallationSource from core.plugin.impl.plugin import PluginInstaller +from extensions.ext_redis import redis_client from models.account import TenantPluginAutoUpgradeStrategy RETRY_TIMES_OF_ONE_PLUGIN_IN_ONE_TENANT = 3 +CACHE_REDIS_KEY_PREFIX = "plugin_autoupgrade_check_task:cached_plugin_manifests:" +CACHE_REDIS_TTL = 60 * 15 # 15 minutes -cached_plugin_manifests: dict[str, typing.Union[MarketplacePluginDeclaration, None]] = {} +def _get_redis_cache_key(plugin_id: str) -> str: + """Generate Redis cache key for plugin manifest.""" + return f"{CACHE_REDIS_KEY_PREFIX}{plugin_id}" + + +def _get_cached_manifest(plugin_id: str) -> typing.Union[MarketplacePluginDeclaration, None, bool]: + """ + Get cached plugin manifest from Redis. + Returns: + - MarketplacePluginDeclaration: if found in cache + - None: if cached as not found (marketplace returned no result) + - False: if not in cache at all + """ + try: + key = _get_redis_cache_key(plugin_id) + cached_data = redis_client.get(key) + if cached_data is None: + return False + + cached_json = json.loads(cached_data) + if cached_json is None: + return None + + return MarketplacePluginDeclaration.model_validate(cached_json) + except Exception: + return False + + +def _set_cached_manifest(plugin_id: str, manifest: typing.Union[MarketplacePluginDeclaration, None]) -> None: + """ + Cache plugin manifest in Redis. + Args: + plugin_id: The plugin ID + manifest: The manifest to cache, or None if not found in marketplace + """ + try: + key = _get_redis_cache_key(plugin_id) + if manifest is None: + # Cache the fact that this plugin was not found + redis_client.setex(key, CACHE_REDIS_TTL, json.dumps(None)) + else: + # Cache the manifest data + redis_client.setex(key, CACHE_REDIS_TTL, manifest.model_dump_json()) + except Exception: + # If Redis fails, continue without caching + # traceback.print_exc() + pass def marketplace_batch_fetch_plugin_manifests( plugin_ids_plain_list: list[str], ) -> list[MarketplacePluginDeclaration]: - global cached_plugin_manifests - # return marketplace.batch_fetch_plugin_manifests(plugin_ids_plain_list) - not_included_plugin_ids = [ - plugin_id for plugin_id in plugin_ids_plain_list if plugin_id not in cached_plugin_manifests - ] - if not_included_plugin_ids: - manifests = marketplace.batch_fetch_plugin_manifests_ignore_deserialization_error(not_included_plugin_ids) + """Fetch plugin manifests with Redis caching support.""" + cached_manifests: dict[str, typing.Union[MarketplacePluginDeclaration, None]] = {} + not_cached_plugin_ids: list[str] = [] + + # Check Redis cache for each plugin + for plugin_id in plugin_ids_plain_list: + cached_result = _get_cached_manifest(plugin_id) + if cached_result is False: + # Not in cache, need to fetch + not_cached_plugin_ids.append(plugin_id) + else: + # Either found manifest or cached as None (not found in marketplace) + # At this point, cached_result is either MarketplacePluginDeclaration or None + if isinstance(cached_result, bool): + # This should never happen due to the if condition above, but for type safety + continue + cached_manifests[plugin_id] = cached_result + + # Fetch uncached plugins from marketplace + if not_cached_plugin_ids: + manifests = marketplace.batch_fetch_plugin_manifests_ignore_deserialization_error(not_cached_plugin_ids) + + # Cache the fetched manifests for manifest in manifests: - cached_plugin_manifests[manifest.plugin_id] = manifest + cached_manifests[manifest.plugin_id] = manifest + _set_cached_manifest(manifest.plugin_id, manifest) - if ( - len(manifests) == 0 - ): # this indicates that the plugin not found in marketplace, should set None in cache to prevent future check - for plugin_id in not_included_plugin_ids: - cached_plugin_manifests[plugin_id] = None + # Cache plugins that were not found in marketplace + fetched_plugin_ids = {manifest.plugin_id for manifest in manifests} + for plugin_id in not_cached_plugin_ids: + if plugin_id not in fetched_plugin_ids: + cached_manifests[plugin_id] = None + _set_cached_manifest(plugin_id, None) + # Build result list from cached manifests result: list[MarketplacePluginDeclaration] = [] for plugin_id in plugin_ids_plain_list: - final_manifest = cached_plugin_manifests.get(plugin_id) - if final_manifest is not None: - result.append(final_manifest) + cached_manifest: typing.Union[MarketplacePluginDeclaration, None] = cached_manifests.get(plugin_id) + if cached_manifest is not None: + result.append(cached_manifest) return result @@ -157,10 +225,10 @@ def process_tenant_plugin_autoupgrade_check_task( ) except Exception as e: click.echo(click.style(f"Error when upgrading plugin: {e}", fg="red")) - traceback.print_exc() + # traceback.print_exc() break except Exception as e: click.echo(click.style(f"Error when checking upgradable plugin: {e}", fg="red")) - traceback.print_exc() + # traceback.print_exc() return diff --git a/api/tests/fixtures/workflow/test-answer-order.yml b/api/tests/fixtures/workflow/test-answer-order.yml new file mode 100644 index 0000000000..3c6631aebb --- /dev/null +++ b/api/tests/fixtures/workflow/test-answer-order.yml @@ -0,0 +1,222 @@ +app: + description: 'this is a chatflow with 2 answer nodes. + + + it''s outouts should like: + + + ``` + + --- answer 1 --- + + + foo + + --- answer 2 --- + + + + + ```' + icon: 🤖 + icon_background: '#FFEAD5' + mode: advanced-chat + name: test-answer-order + use_icon_as_answer_icon: false +dependencies: +- current_identifier: null + type: marketplace + value: + marketplace_plugin_unique_identifier: langgenius/openai:0.2.6@e2665624a156f52160927bceac9e169bd7e5ae6b936ae82575e14c90af390e6e + version: null +kind: app +version: 0.4.0 +workflow: + conversation_variables: [] + environment_variables: [] + features: + file_upload: + allowed_file_extensions: + - .JPG + - .JPEG + - .PNG + - .GIF + - .WEBP + - .SVG + allowed_file_types: + - image + allowed_file_upload_methods: + - local_file + - remote_url + enabled: false + fileUploadConfig: + audio_file_size_limit: 50 + batch_count_limit: 5 + file_size_limit: 15 + image_file_size_limit: 10 + video_file_size_limit: 100 + workflow_file_upload_limit: 10 + image: + enabled: false + number_limits: 3 + transfer_methods: + - local_file + - remote_url + number_limits: 3 + opening_statement: '' + retriever_resource: + enabled: true + sensitive_word_avoidance: + enabled: false + speech_to_text: + enabled: false + suggested_questions: [] + suggested_questions_after_answer: + enabled: false + text_to_speech: + enabled: false + language: '' + voice: '' + graph: + edges: + - data: + isInIteration: false + isInLoop: false + sourceType: answer + targetType: answer + id: 1759052466526-source-1759052469368-target + source: '1759052466526' + sourceHandle: source + target: '1759052469368' + targetHandle: target + type: custom + zIndex: 0 + - data: + isInIteration: false + isInLoop: false + sourceType: start + targetType: llm + id: 1759052439553-source-1759052580454-target + source: '1759052439553' + sourceHandle: source + target: '1759052580454' + targetHandle: target + type: custom + zIndex: 0 + - data: + isInIteration: false + isInLoop: false + sourceType: llm + targetType: answer + id: 1759052580454-source-1759052466526-target + source: '1759052580454' + sourceHandle: source + target: '1759052466526' + targetHandle: target + type: custom + zIndex: 0 + nodes: + - data: + selected: false + title: Start + type: start + variables: [] + height: 52 + id: '1759052439553' + position: + x: 30 + y: 242 + positionAbsolute: + x: 30 + y: 242 + sourcePosition: right + targetPosition: left + type: custom + width: 242 + - data: + answer: '--- answer 1 --- + + + foo + + ' + selected: false + title: Answer + type: answer + variables: [] + height: 100 + id: '1759052466526' + position: + x: 632 + y: 242 + positionAbsolute: + x: 632 + y: 242 + selected: true + sourcePosition: right + targetPosition: left + type: custom + width: 242 + - data: + answer: '--- answer 2 --- + + + {{#1759052580454.text#}} + + ' + selected: false + title: Answer 2 + type: answer + variables: [] + height: 103 + id: '1759052469368' + position: + x: 934 + y: 242 + positionAbsolute: + x: 934 + y: 242 + selected: false + sourcePosition: right + targetPosition: left + type: custom + width: 242 + - data: + context: + enabled: false + variable_selector: [] + model: + completion_params: + temperature: 0.7 + mode: chat + name: gpt-4o + provider: langgenius/openai/openai + prompt_template: + - id: 5c1d873b-06b2-4dce-939e-672882bbd7c0 + role: system + text: '' + - role: user + text: '{{#sys.query#}}' + selected: false + title: LLM + type: llm + vision: + enabled: false + height: 88 + id: '1759052580454' + position: + x: 332 + y: 242 + positionAbsolute: + x: 332 + y: 242 + selected: false + sourcePosition: right + targetPosition: left + type: custom + width: 242 + viewport: + x: 126.2797574512839 + y: 289.55932160537446 + zoom: 1.0743222672006216 + rag_pipeline_variables: [] diff --git a/api/tests/fixtures/workflow/update-conversation-variable-in-iteration.yml b/api/tests/fixtures/workflow/update-conversation-variable-in-iteration.yml new file mode 100644 index 0000000000..ffc6eb9120 --- /dev/null +++ b/api/tests/fixtures/workflow/update-conversation-variable-in-iteration.yml @@ -0,0 +1,316 @@ +app: + description: 'This chatflow receives a sys.query, writes it into the `answer` variable, + and then outputs the `answer` variable. + + + `answer` is a conversation variable with a blank default value; it will be updated + in an iteration node. + + + if this chatflow works correctly, it will output the `sys.query` as the same.' + icon: 🤖 + icon_background: '#FFEAD5' + mode: advanced-chat + name: update-conversation-variable-in-iteration + use_icon_as_answer_icon: false +dependencies: [] +kind: app +version: 0.4.0 +workflow: + conversation_variables: + - description: '' + id: c30af82d-b2ec-417d-a861-4dd78584faa4 + name: answer + selector: + - conversation + - answer + value: '' + value_type: string + environment_variables: [] + features: + file_upload: + allowed_file_extensions: + - .JPG + - .JPEG + - .PNG + - .GIF + - .WEBP + - .SVG + allowed_file_types: + - image + allowed_file_upload_methods: + - local_file + - remote_url + enabled: false + fileUploadConfig: + audio_file_size_limit: 50 + batch_count_limit: 5 + file_size_limit: 15 + image_file_size_limit: 10 + video_file_size_limit: 100 + workflow_file_upload_limit: 10 + image: + enabled: false + number_limits: 3 + transfer_methods: + - local_file + - remote_url + number_limits: 3 + opening_statement: '' + retriever_resource: + enabled: true + sensitive_word_avoidance: + enabled: false + speech_to_text: + enabled: false + suggested_questions: [] + suggested_questions_after_answer: + enabled: false + text_to_speech: + enabled: false + language: '' + voice: '' + graph: + edges: + - data: + isInIteration: false + isInLoop: false + sourceType: start + targetType: code + id: 1759032354471-source-1759032363865-target + source: '1759032354471' + sourceHandle: source + target: '1759032363865' + targetHandle: target + type: custom + zIndex: 0 + - data: + isInIteration: false + isInLoop: false + sourceType: code + targetType: iteration + id: 1759032363865-source-1759032379989-target + source: '1759032363865' + sourceHandle: source + target: '1759032379989' + targetHandle: target + type: custom + zIndex: 0 + - data: + isInIteration: true + isInLoop: false + iteration_id: '1759032379989' + sourceType: iteration-start + targetType: assigner + id: 1759032379989start-source-1759032394460-target + source: 1759032379989start + sourceHandle: source + target: '1759032394460' + targetHandle: target + type: custom + zIndex: 1002 + - data: + isInIteration: false + isInLoop: false + sourceType: iteration + targetType: answer + id: 1759032379989-source-1759032410331-target + source: '1759032379989' + sourceHandle: source + target: '1759032410331' + targetHandle: target + type: custom + zIndex: 0 + - data: + isInIteration: true + isInLoop: false + iteration_id: '1759032379989' + sourceType: assigner + targetType: code + id: 1759032394460-source-1759032476318-target + source: '1759032394460' + sourceHandle: source + target: '1759032476318' + targetHandle: target + type: custom + zIndex: 1002 + nodes: + - data: + selected: false + title: Start + type: start + variables: [] + height: 52 + id: '1759032354471' + position: + x: 30 + y: 302 + positionAbsolute: + x: 30 + y: 302 + selected: false + sourcePosition: right + targetPosition: left + type: custom + width: 242 + - data: + code: "\ndef main():\n return {\n \"result\": [1],\n }\n" + code_language: python3 + outputs: + result: + children: null + type: array[number] + selected: false + title: Code + type: code + variables: [] + height: 52 + id: '1759032363865' + position: + x: 332 + y: 302 + positionAbsolute: + x: 332 + y: 302 + sourcePosition: right + targetPosition: left + type: custom + width: 242 + - data: + error_handle_mode: terminated + height: 204 + is_parallel: false + iterator_input_type: array[number] + iterator_selector: + - '1759032363865' + - result + output_selector: + - '1759032476318' + - result + output_type: array[string] + parallel_nums: 10 + selected: false + start_node_id: 1759032379989start + title: Iteration + type: iteration + width: 808 + height: 204 + id: '1759032379989' + position: + x: 634 + y: 302 + positionAbsolute: + x: 634 + y: 302 + selected: true + sourcePosition: right + targetPosition: left + type: custom + width: 808 + zIndex: 1 + - data: + desc: '' + isInIteration: true + selected: false + title: '' + type: iteration-start + draggable: false + height: 48 + id: 1759032379989start + parentId: '1759032379989' + position: + x: 60 + y: 78 + positionAbsolute: + x: 694 + y: 380 + selectable: false + sourcePosition: right + targetPosition: left + type: custom-iteration-start + width: 44 + zIndex: 1002 + - data: + isInIteration: true + isInLoop: false + items: + - input_type: variable + operation: over-write + value: + - sys + - query + variable_selector: + - conversation + - answer + write_mode: over-write + iteration_id: '1759032379989' + selected: false + title: Variable Assigner + type: assigner + version: '2' + height: 84 + id: '1759032394460' + parentId: '1759032379989' + position: + x: 204 + y: 60 + positionAbsolute: + x: 838 + y: 362 + sourcePosition: right + targetPosition: left + type: custom + width: 242 + zIndex: 1002 + - data: + answer: '{{#conversation.answer#}}' + selected: false + title: Answer + type: answer + variables: [] + height: 104 + id: '1759032410331' + position: + x: 1502 + y: 302 + positionAbsolute: + x: 1502 + y: 302 + selected: false + sourcePosition: right + targetPosition: left + type: custom + width: 242 + - data: + code: "\ndef main():\n return {\n \"result\": '',\n }\n" + code_language: python3 + isInIteration: true + isInLoop: false + iteration_id: '1759032379989' + outputs: + result: + children: null + type: string + selected: false + title: Code 2 + type: code + variables: [] + height: 52 + id: '1759032476318' + parentId: '1759032379989' + position: + x: 506 + y: 76 + positionAbsolute: + x: 1140 + y: 378 + sourcePosition: right + targetPosition: left + type: custom + width: 242 + zIndex: 1002 + viewport: + x: 120.39999999999998 + y: 85.20000000000005 + zoom: 0.7 + rag_pipeline_variables: [] diff --git a/api/tests/unit_tests/core/workflow/entities/test_variable_pool.py b/api/tests/unit_tests/core/workflow/entities/test_variable_pool.py new file mode 100644 index 0000000000..68fe82d05e --- /dev/null +++ b/api/tests/unit_tests/core/workflow/entities/test_variable_pool.py @@ -0,0 +1,113 @@ +from core.variables.segments import ( + BooleanSegment, + IntegerSegment, + NoneSegment, + StringSegment, +) +from core.workflow.entities.variable_pool import VariablePool + + +class TestVariablePoolGetAndNestedAttribute: + # + # _get_nested_attribute tests + # + def test__get_nested_attribute_existing_key(self): + pool = VariablePool.empty() + obj = {"a": 123} + segment = pool._get_nested_attribute(obj, "a") + assert segment is not None + assert segment.value == 123 + + def test__get_nested_attribute_missing_key(self): + pool = VariablePool.empty() + obj = {"a": 123} + segment = pool._get_nested_attribute(obj, "b") + assert segment is None + + def test__get_nested_attribute_non_dict(self): + pool = VariablePool.empty() + obj = ["not", "a", "dict"] + segment = pool._get_nested_attribute(obj, "a") + assert segment is None + + def test__get_nested_attribute_with_none_value(self): + pool = VariablePool.empty() + obj = {"a": None} + segment = pool._get_nested_attribute(obj, "a") + assert segment is not None + assert isinstance(segment, NoneSegment) + + def test__get_nested_attribute_with_empty_string(self): + pool = VariablePool.empty() + obj = {"a": ""} + segment = pool._get_nested_attribute(obj, "a") + assert segment is not None + assert isinstance(segment, StringSegment) + assert segment.value == "" + + # + # get tests + # + def test_get_simple_variable(self): + pool = VariablePool.empty() + pool.add(("node1", "var1"), "value1") + segment = pool.get(("node1", "var1")) + assert segment is not None + assert segment.value == "value1" + + def test_get_missing_variable(self): + pool = VariablePool.empty() + result = pool.get(("node1", "unknown")) + assert result is None + + def test_get_with_too_short_selector(self): + pool = VariablePool.empty() + result = pool.get(("only_node",)) + assert result is None + + def test_get_nested_object_attribute(self): + pool = VariablePool.empty() + obj_value = {"inner": "hello"} + pool.add(("node1", "obj"), obj_value) + + # simulate selector with nested attr + segment = pool.get(("node1", "obj", "inner")) + assert segment is not None + assert segment.value == "hello" + + def test_get_nested_object_missing_attribute(self): + pool = VariablePool.empty() + obj_value = {"inner": "hello"} + pool.add(("node1", "obj"), obj_value) + + result = pool.get(("node1", "obj", "not_exist")) + assert result is None + + def test_get_nested_object_attribute_with_falsy_values(self): + pool = VariablePool.empty() + obj_value = { + "inner_none": None, + "inner_empty": "", + "inner_zero": 0, + "inner_false": False, + } + pool.add(("node1", "obj"), obj_value) + + segment_none = pool.get(("node1", "obj", "inner_none")) + assert segment_none is not None + assert isinstance(segment_none, NoneSegment) + + segment_empty = pool.get(("node1", "obj", "inner_empty")) + assert segment_empty is not None + assert isinstance(segment_empty, StringSegment) + assert segment_empty.value == "" + + segment_zero = pool.get(("node1", "obj", "inner_zero")) + assert segment_zero is not None + assert isinstance(segment_zero, IntegerSegment) + assert segment_zero.value == 0 + + segment_false = pool.get(("node1", "obj", "inner_false")) + assert segment_false is not None + assert isinstance(segment_false, BooleanSegment) + assert segment_false.value is False diff --git a/api/tests/unit_tests/core/workflow/graph_engine/test_answer_order_workflow.py b/api/tests/unit_tests/core/workflow/graph_engine/test_answer_order_workflow.py new file mode 100644 index 0000000000..6569439b56 --- /dev/null +++ b/api/tests/unit_tests/core/workflow/graph_engine/test_answer_order_workflow.py @@ -0,0 +1,28 @@ +from .test_mock_config import MockConfigBuilder +from .test_table_runner import TableTestRunner, WorkflowTestCase + +LLM_NODE_ID = "1759052580454" + + +def test_answer_nodes_emit_in_order() -> None: + mock_config = ( + MockConfigBuilder() + .with_llm_response("unused default") + .with_node_output(LLM_NODE_ID, {"text": "mocked llm text"}) + .build() + ) + + expected_answer = "--- answer 1 ---\n\nfoo\n--- answer 2 ---\n\nmocked llm text\n" + + case = WorkflowTestCase( + fixture_path="test-answer-order", + query="", + expected_outputs={"answer": expected_answer}, + use_auto_mock=True, + mock_config=mock_config, + ) + + runner = TableTestRunner() + result = runner.run_test_case(case) + + assert result.success, result.error diff --git a/api/tests/unit_tests/core/workflow/graph_engine/test_update_conversation_variable_iteration.py b/api/tests/unit_tests/core/workflow/graph_engine/test_update_conversation_variable_iteration.py new file mode 100644 index 0000000000..a7309f64de --- /dev/null +++ b/api/tests/unit_tests/core/workflow/graph_engine/test_update_conversation_variable_iteration.py @@ -0,0 +1,41 @@ +"""Validate conversation variable updates inside an iteration workflow. + +This test uses the ``update-conversation-variable-in-iteration`` fixture, which +routes ``sys.query`` into the conversation variable ``answer`` from within an +iteration container. The workflow should surface that updated conversation +variable in the final answer output. + +Code nodes in the fixture are mocked because their concrete outputs are not +relevant to verifying variable propagation semantics. +""" + +from .test_mock_config import MockConfigBuilder +from .test_table_runner import TableTestRunner, WorkflowTestCase + + +def test_update_conversation_variable_in_iteration(): + fixture_name = "update-conversation-variable-in-iteration" + user_query = "ensure conversation variable syncs" + + mock_config = ( + MockConfigBuilder() + .with_node_output("1759032363865", {"result": [1]}) + .with_node_output("1759032476318", {"result": ""}) + .build() + ) + + case = WorkflowTestCase( + fixture_path=fixture_name, + use_auto_mock=True, + mock_config=mock_config, + query=user_query, + expected_outputs={"answer": user_query}, + description="Conversation variable updated within iteration should flow to answer output.", + ) + + runner = TableTestRunner() + result = runner.run_test_case(case) + + assert result.success, f"Workflow execution failed: {result.error}" + assert result.actual_outputs is not None + assert result.actual_outputs.get("answer") == user_query diff --git a/api/uv.lock b/api/uv.lock index 7ce71cd215..87799977ca 100644 --- a/api/uv.lock +++ b/api/uv.lock @@ -1076,7 +1076,7 @@ wheels = [ [[package]] name = "cos-python-sdk-v5" -version = "1.9.30" +version = "1.9.38" source = { registry = "https://pypi.org/simple" } dependencies = [ { name = "crcmod" }, @@ -1085,7 +1085,10 @@ dependencies = [ { name = "six" }, { name = "xmltodict" }, ] -sdist = { url = "https://files.pythonhosted.org/packages/c4/f2/be99b41433b33a76896680920fca621f191875ca410a66778015e47a501b/cos-python-sdk-v5-1.9.30.tar.gz", hash = "sha256:a23fd090211bf90883066d90cd74317860aa67c6d3aa80fe5e44b18c7e9b2a81", size = 108384, upload-time = "2024-06-14T08:02:37.063Z" } +sdist = { url = "https://files.pythonhosted.org/packages/24/3c/d208266fec7cc3221b449e236b87c3fc1999d5ac4379d4578480321cfecc/cos_python_sdk_v5-1.9.38.tar.gz", hash = "sha256:491a8689ae2f1a6f04dacba66a877b2c8d361456f9cfd788ed42170a1cbf7a9f", size = 98092, upload-time = "2025-07-22T07:56:20.34Z" } +wheels = [ + { url = "https://files.pythonhosted.org/packages/ab/c8/c9c156aa3bc7caba9b4f8a2b6abec3da6263215988f3fec0ea843f137a10/cos_python_sdk_v5-1.9.38-py3-none-any.whl", hash = "sha256:1d3dd3be2bd992b2e9c2dcd018e2596aa38eab022dbc86b4a5d14c8fc88370e6", size = 92601, upload-time = "2025-08-17T05:12:30.867Z" }, +] [[package]] name = "couchbase" @@ -1624,7 +1627,7 @@ dev = [ storage = [ { name = "azure-storage-blob", specifier = "==12.13.0" }, { name = "bce-python-sdk", specifier = "~=0.9.23" }, - { name = "cos-python-sdk-v5", specifier = "==1.9.30" }, + { name = "cos-python-sdk-v5", specifier = "==1.9.38" }, { name = "esdk-obs-python", specifier = "==3.24.6.1" }, { name = "google-cloud-storage", specifier = "==2.16.0" }, { name = "opendal", specifier = "~=0.46.0" }, @@ -1646,7 +1649,7 @@ vdb = [ { name = "elasticsearch", specifier = "==8.14.0" }, { name = "mo-vector", specifier = "~=0.1.13" }, { name = "opensearch-py", specifier = "==2.4.0" }, - { name = "oracledb", specifier = "==3.0.0" }, + { name = "oracledb", specifier = "==3.3.0" }, { name = "pgvecto-rs", extras = ["sqlalchemy"], specifier = "~=0.2.1" }, { name = "pgvector", specifier = "==0.2.5" }, { name = "pymilvus", specifier = "~=2.5.0" }, @@ -4079,23 +4082,23 @@ numpy = [ [[package]] name = "oracledb" -version = "3.0.0" +version = "3.3.0" source = { registry = "https://pypi.org/simple" } dependencies = [ { name = "cryptography" }, ] -sdist = { url = "https://files.pythonhosted.org/packages/bf/39/712f797b75705c21148fa1d98651f63c2e5cc6876e509a0a9e2f5b406572/oracledb-3.0.0.tar.gz", hash = "sha256:64dc86ee5c032febc556798b06e7b000ef6828bb0252084f6addacad3363db85", size = 840431, upload-time = "2025-03-03T19:36:12.223Z" } +sdist = { url = "https://files.pythonhosted.org/packages/51/c9/fae18fa5d803712d188486f8e86ad4f4e00316793ca19745d7c11092c360/oracledb-3.3.0.tar.gz", hash = "sha256:e830d3544a1578296bcaa54c6e8c8ae10a58c7db467c528c4b27adbf9c8b4cb0", size = 811776, upload-time = "2025-07-29T22:34:10.489Z" } wheels = [ - { url = "https://files.pythonhosted.org/packages/fa/bf/d872c4b3fc15cd3261fe0ea72b21d181700c92dbc050160e161654987062/oracledb-3.0.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:52daa9141c63dfa75c07d445e9bb7f69f43bfb3c5a173ecc48c798fe50288d26", size = 4312963, upload-time = "2025-03-03T19:36:32.576Z" }, - { url = "https://files.pythonhosted.org/packages/b1/ea/01ee29e76a610a53bb34fdc1030f04b7669c3f80b25f661e07850fc6160e/oracledb-3.0.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:af98941789df4c6aaaf4338f5b5f6b7f2c8c3fe6f8d6a9382f177f350868747a", size = 2661536, upload-time = "2025-03-03T19:36:34.904Z" }, - { url = "https://files.pythonhosted.org/packages/3d/8e/ad380e34a46819224423b4773e58c350bc6269643c8969604097ced8c3bc/oracledb-3.0.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9812bb48865aaec35d73af54cd1746679f2a8a13cbd1412ab371aba2e39b3943", size = 2867461, upload-time = "2025-03-03T19:36:36.508Z" }, - { url = "https://files.pythonhosted.org/packages/96/09/ecc4384a27fd6e1e4de824ae9c160e4ad3aaebdaade5b4bdcf56a4d1ff63/oracledb-3.0.0-cp311-cp311-win32.whl", hash = "sha256:6c27fe0de64f2652e949eb05b3baa94df9b981a4a45fa7f8a991e1afb450c8e2", size = 1752046, upload-time = "2025-03-03T19:36:38.313Z" }, - { url = "https://files.pythonhosted.org/packages/62/e8/f34bde24050c6e55eeba46b23b2291f2dd7fd272fa8b322dcbe71be55778/oracledb-3.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:f922709672002f0b40997456f03a95f03e5712a86c61159951c5ce09334325e0", size = 2101210, upload-time = "2025-03-03T19:36:40.669Z" }, - { url = "https://files.pythonhosted.org/packages/6f/fc/24590c3a3d41e58494bd3c3b447a62835138e5f9b243d9f8da0cfb5da8dc/oracledb-3.0.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:acd0e747227dea01bebe627b07e958bf36588a337539f24db629dc3431d3f7eb", size = 4351993, upload-time = "2025-03-03T19:36:42.577Z" }, - { url = "https://files.pythonhosted.org/packages/b7/b6/1f3b0b7bb94d53e8857d77b2e8dbdf6da091dd7e377523e24b79dac4fd71/oracledb-3.0.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f8b402f77c22af031cd0051aea2472ecd0635c1b452998f511aa08b7350c90a4", size = 2532640, upload-time = "2025-03-03T19:36:45.066Z" }, - { url = "https://files.pythonhosted.org/packages/72/1a/1815f6c086ab49c00921cf155ff5eede5267fb29fcec37cb246339a5ce4d/oracledb-3.0.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:378a27782e9a37918bd07a5a1427a77cb6f777d0a5a8eac9c070d786f50120ef", size = 2765949, upload-time = "2025-03-03T19:36:47.47Z" }, - { url = "https://files.pythonhosted.org/packages/33/8d/208900f8d372909792ee70b2daad3f7361181e55f2217c45ed9dff658b54/oracledb-3.0.0-cp312-cp312-win32.whl", hash = "sha256:54a28c2cb08316a527cd1467740a63771cc1c1164697c932aa834c0967dc4efc", size = 1709373, upload-time = "2025-03-03T19:36:49.67Z" }, - { url = "https://files.pythonhosted.org/packages/0c/5e/c21754f19c896102793c3afec2277e2180aa7d505e4d7fcca24b52d14e4f/oracledb-3.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:8289bad6d103ce42b140e40576cf0c81633e344d56e2d738b539341eacf65624", size = 2056452, upload-time = "2025-03-03T19:36:51.363Z" }, + { url = "https://files.pythonhosted.org/packages/3f/35/95d9a502fdc48ce1ef3a513ebd027488353441e15aa0448619abb3d09d32/oracledb-3.3.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:d9adb74f837838e21898d938e3a725cf73099c65f98b0b34d77146b453e945e0", size = 3963945, upload-time = "2025-07-29T22:34:28.633Z" }, + { url = "https://files.pythonhosted.org/packages/16/a7/8f1ef447d995bb51d9fdc36356697afeceb603932f16410c12d52b2df1a4/oracledb-3.3.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4b063d1007882570f170ebde0f364e78d4a70c8f015735cc900663278b9ceef7", size = 2449385, upload-time = "2025-07-29T22:34:30.592Z" }, + { url = "https://files.pythonhosted.org/packages/b3/fa/6a78480450bc7d256808d0f38ade3385735fb5a90dab662167b4257dcf94/oracledb-3.3.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:187728f0a2d161676b8c581a9d8f15d9631a8fea1e628f6d0e9fa2f01280cd22", size = 2634943, upload-time = "2025-07-29T22:34:33.142Z" }, + { url = "https://files.pythonhosted.org/packages/5b/90/ea32b569a45fb99fac30b96f1ac0fb38b029eeebb78357bc6db4be9dde41/oracledb-3.3.0-cp311-cp311-win32.whl", hash = "sha256:920f14314f3402c5ab98f2efc5932e0547e9c0a4ca9338641357f73844e3e2b1", size = 1483549, upload-time = "2025-07-29T22:34:35.015Z" }, + { url = "https://files.pythonhosted.org/packages/81/55/ae60f72836eb8531b630299f9ed68df3fe7868c6da16f820a108155a21f9/oracledb-3.3.0-cp311-cp311-win_amd64.whl", hash = "sha256:825edb97976468db1c7e52c78ba38d75ce7e2b71a2e88f8629bcf02be8e68a8a", size = 1834737, upload-time = "2025-07-29T22:34:36.824Z" }, + { url = "https://files.pythonhosted.org/packages/08/a8/f6b7809d70e98e113786d5a6f1294da81c046d2fa901ad656669fc5d7fae/oracledb-3.3.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:9d25e37d640872731ac9b73f83cbc5fc4743cd744766bdb250488caf0d7696a8", size = 3943512, upload-time = "2025-07-29T22:34:39.237Z" }, + { url = "https://files.pythonhosted.org/packages/df/b9/8145ad8991f4864d3de4a911d439e5bc6cdbf14af448f3ab1e846a54210c/oracledb-3.3.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b0bf7cdc2b668f939aa364f552861bc7a149d7cd3f3794730d43ef07613b2bf9", size = 2276258, upload-time = "2025-07-29T22:34:41.547Z" }, + { url = "https://files.pythonhosted.org/packages/56/bf/f65635ad5df17d6e4a2083182750bb136ac663ff0e9996ce59d77d200f60/oracledb-3.3.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2fe20540fde64a6987046807ea47af93be918fd70b9766b3eb803c01e6d4202e", size = 2458811, upload-time = "2025-07-29T22:34:44.648Z" }, + { url = "https://files.pythonhosted.org/packages/7d/30/e0c130b6278c10b0e6cd77a3a1a29a785c083c549676cf701c5d180b8e63/oracledb-3.3.0-cp312-cp312-win32.whl", hash = "sha256:db080be9345cbf9506ffdaea3c13d5314605355e76d186ec4edfa49960ffb813", size = 1445525, upload-time = "2025-07-29T22:34:46.603Z" }, + { url = "https://files.pythonhosted.org/packages/1a/5c/7254f5e1a33a5d6b8bf6813d4f4fdcf5c4166ec8a7af932d987879d5595c/oracledb-3.3.0-cp312-cp312-win_amd64.whl", hash = "sha256:be81e3afe79f6c8ece79a86d6067ad1572d2992ce1c590a086f3755a09535eb4", size = 1789976, upload-time = "2025-07-29T22:34:48.5Z" }, ] [[package]] diff --git a/docker/.env.example b/docker/.env.example index c0f084796e..8c23bfc9b7 100644 --- a/docker/.env.example +++ b/docker/.env.example @@ -45,7 +45,7 @@ APP_WEB_URL= # Recommendation: use a dedicated domain (e.g., https://upload.example.com). # Alternatively, use http://:5001 or http://api:5001, # ensuring port 5001 is externally accessible (see docker-compose.yaml). -FILES_URL=http://api:5001 +FILES_URL= # INTERNAL_FILES_URL is used for plugin daemon communication within Docker network. # Set this to the internal Docker service URL for proper plugin file access. diff --git a/docker/docker-compose.yaml b/docker/docker-compose.yaml index 2617f84e7d..cc66f69550 100644 --- a/docker/docker-compose.yaml +++ b/docker/docker-compose.yaml @@ -10,7 +10,7 @@ x-shared-env: &shared-api-worker-env SERVICE_API_URL: ${SERVICE_API_URL:-} APP_API_URL: ${APP_API_URL:-} APP_WEB_URL: ${APP_WEB_URL:-} - FILES_URL: ${FILES_URL:-http://api:5001} + FILES_URL: ${FILES_URL:-} INTERNAL_FILES_URL: ${INTERNAL_FILES_URL:-} LANG: ${LANG:-en_US.UTF-8} LC_ALL: ${LC_ALL:-en_US.UTF-8} diff --git a/sdks/python-client/dify_client/client.py b/sdks/python-client/dify_client/client.py index 201391eae9..2154741e91 100644 --- a/sdks/python-client/dify_client/client.py +++ b/sdks/python-client/dify_client/client.py @@ -1,5 +1,5 @@ import json -from typing import Literal +from typing import IO, Literal import requests diff --git a/web/app/components/app/configuration/dataset-config/index.tsx b/web/app/components/app/configuration/dataset-config/index.tsx index 6165cfdeec..65ef74bc27 100644 --- a/web/app/components/app/configuration/dataset-config/index.tsx +++ b/web/app/components/app/configuration/dataset-config/index.tsx @@ -65,13 +65,40 @@ const DatasetConfig: FC = () => { const onRemove = (id: string) => { const filteredDataSets = dataSet.filter(item => item.id !== id) setDataSet(filteredDataSets) - const retrievalConfig = getMultipleRetrievalConfig(datasetConfigs as any, filteredDataSets, dataSet, { + const { datasets, retrieval_model, score_threshold_enabled, ...restConfigs } = datasetConfigs + const { + top_k, + score_threshold, + reranking_model, + reranking_mode, + weights, + reranking_enable, + } = restConfigs + const oldRetrievalConfig = { + top_k, + score_threshold, + reranking_model: (reranking_model.reranking_provider_name && reranking_model.reranking_model_name) ? { + provider: reranking_model.reranking_provider_name, + model: reranking_model.reranking_model_name, + } : undefined, + reranking_mode, + weights, + reranking_enable, + } + const retrievalConfig = getMultipleRetrievalConfig(oldRetrievalConfig, filteredDataSets, dataSet, { provider: currentRerankProvider?.provider, model: currentRerankModel?.model, }) setDatasetConfigs({ - ...(datasetConfigs as any), + ...datasetConfigsRef.current, ...retrievalConfig, + reranking_model: { + reranking_provider_name: retrievalConfig?.reranking_model?.provider || '', + reranking_model_name: retrievalConfig?.reranking_model?.model || '', + }, + retrieval_model, + score_threshold_enabled, + datasets, }) const { allExternal, diff --git a/web/app/components/app/configuration/dataset-config/params-config/config-content.tsx b/web/app/components/app/configuration/dataset-config/params-config/config-content.tsx index cb61b927bc..1558d32fc6 100644 --- a/web/app/components/app/configuration/dataset-config/params-config/config-content.tsx +++ b/web/app/components/app/configuration/dataset-config/params-config/config-content.tsx @@ -30,11 +30,11 @@ import { noop } from 'lodash-es' type Props = { datasetConfigs: DatasetConfigs onChange: (configs: DatasetConfigs, isRetrievalModeChange?: boolean) => void + selectedDatasets?: DataSet[] isInWorkflow?: boolean singleRetrievalModelConfig?: ModelConfig onSingleRetrievalModelChange?: (config: ModelConfig) => void onSingleRetrievalModelParamsChange?: (config: ModelConfig) => void - selectedDatasets?: DataSet[] } const ConfigContent: FC = ({ @@ -61,22 +61,28 @@ const ConfigContent: FC = ({ const { modelList: rerankModelList, + currentModel: validDefaultRerankModel, + currentProvider: validDefaultRerankProvider, } = useModelListAndDefaultModelAndCurrentProviderAndModel(ModelTypeEnum.rerank) + /** + * If reranking model is set and is valid, use the reranking model + * Otherwise, check if the default reranking model is valid + */ const { currentModel: currentRerankModel, } = useCurrentProviderAndModel( rerankModelList, { - provider: datasetConfigs.reranking_model?.reranking_provider_name, - model: datasetConfigs.reranking_model?.reranking_model_name, + provider: datasetConfigs.reranking_model?.reranking_provider_name || validDefaultRerankProvider?.provider || '', + model: datasetConfigs.reranking_model?.reranking_model_name || validDefaultRerankModel?.model || '', }, ) const rerankModel = useMemo(() => { return { - provider_name: datasetConfigs?.reranking_model?.reranking_provider_name ?? '', - model_name: datasetConfigs?.reranking_model?.reranking_model_name ?? '', + provider_name: datasetConfigs.reranking_model?.reranking_provider_name ?? '', + model_name: datasetConfigs.reranking_model?.reranking_model_name ?? '', } }, [datasetConfigs.reranking_model]) @@ -135,7 +141,7 @@ const ConfigContent: FC = ({ }) } - const model = singleRetrievalConfig + const model = singleRetrievalConfig // Legacy code, for compatibility, have to keep it const rerankingModeOptions = [ { @@ -158,7 +164,7 @@ const ConfigContent: FC = ({ const canManuallyToggleRerank = useMemo(() => { return (selectedDatasetsMode.allInternal && selectedDatasetsMode.allEconomic) - || selectedDatasetsMode.allExternal + || selectedDatasetsMode.allExternal }, [selectedDatasetsMode.allEconomic, selectedDatasetsMode.allExternal, selectedDatasetsMode.allInternal]) const showRerankModel = useMemo(() => { @@ -168,7 +174,7 @@ const ConfigContent: FC = ({ return datasetConfigs.reranking_enable }, [datasetConfigs.reranking_enable, canManuallyToggleRerank]) - const handleDisabledSwitchClick = useCallback((enable: boolean) => { + const handleManuallyToggleRerank = useCallback((enable: boolean) => { if (!currentRerankModel && enable) Toast.notify({ type: 'error', message: t('workflow.errorMsg.rerankModelRequired') }) onChange({ @@ -255,12 +261,11 @@ const ConfigContent: FC = ({
{ - selectedDatasetsMode.allEconomic && !selectedDatasetsMode.mixtureInternalAndExternal && ( + canManuallyToggleRerank && ( ) } diff --git a/web/app/components/app/configuration/index.tsx b/web/app/components/app/configuration/index.tsx index 091900642a..f1f81ebf97 100644 --- a/web/app/components/app/configuration/index.tsx +++ b/web/app/components/app/configuration/index.tsx @@ -284,18 +284,28 @@ const Configuration: FC = () => { setRerankSettingModalOpen(true) const { datasets, retrieval_model, score_threshold_enabled, ...restConfigs } = datasetConfigs + const { + top_k, + score_threshold, + reranking_model, + reranking_mode, + weights, + reranking_enable, + } = restConfigs - const retrievalConfig = getMultipleRetrievalConfig({ - top_k: restConfigs.top_k, - score_threshold: restConfigs.score_threshold, - reranking_model: restConfigs.reranking_model && { - provider: restConfigs.reranking_model.reranking_provider_name, - model: restConfigs.reranking_model.reranking_model_name, - }, - reranking_mode: restConfigs.reranking_mode, - weights: restConfigs.weights, - reranking_enable: restConfigs.reranking_enable, - }, newDatasets, dataSets, { + const oldRetrievalConfig = { + top_k, + score_threshold, + reranking_model: (reranking_model.reranking_provider_name && reranking_model.reranking_model_name) ? { + provider: reranking_model.reranking_provider_name, + model: reranking_model.reranking_model_name, + } : undefined, + reranking_mode, + weights, + reranking_enable, + } + + const retrievalConfig = getMultipleRetrievalConfig(oldRetrievalConfig, newDatasets, dataSets, { provider: currentRerankProvider?.provider, model: currentRerankModel?.model, }) diff --git a/web/app/components/base/chat/chat/index.tsx b/web/app/components/base/chat/chat/index.tsx index bee37cf2cd..a362f4dc99 100644 --- a/web/app/components/base/chat/chat/index.tsx +++ b/web/app/components/base/chat/chat/index.tsx @@ -160,8 +160,13 @@ const Chat: FC = ({ }) useEffect(() => { - window.addEventListener('resize', debounce(handleWindowResize)) - return () => window.removeEventListener('resize', handleWindowResize) + const debouncedHandler = debounce(handleWindowResize, 200) + window.addEventListener('resize', debouncedHandler) + + return () => { + window.removeEventListener('resize', debouncedHandler) + debouncedHandler.cancel() + } }, [handleWindowResize]) useEffect(() => { diff --git a/web/app/components/base/chat/chat/loading-anim/style.module.css b/web/app/components/base/chat/chat/loading-anim/style.module.css index b1371ec82a..d5a373df6f 100644 --- a/web/app/components/base/chat/chat/loading-anim/style.module.css +++ b/web/app/components/base/chat/chat/loading-anim/style.module.css @@ -1,6 +1,6 @@ .dot-flashing { position: relative; - animation: 1s infinite linear alternate; + animation: dot-flashing 1s infinite linear alternate; animation-delay: 0.5s; } @@ -10,7 +10,7 @@ display: inline-block; position: absolute; top: 0; - animation: 1s infinite linear alternate; + animation: dot-flashing 1s infinite linear alternate; } .dot-flashing::before { @@ -51,15 +51,21 @@ border-radius: 50%; background-color: #667085; color: #667085; - animation-name: dot-flashing; + animation: dot-flashing 1s infinite linear alternate; +} + +.text { + animation-delay: 0.5s; } .text::before { left: -7px; + animation-delay: 0s; } .text::after { left: 7px; + animation-delay: 1s; } .avatar, @@ -70,13 +76,19 @@ border-radius: 50%; background-color: #155EEF; color: #155EEF; - animation-name: dot-flashing-avatar; + animation: dot-flashing-avatar 1s infinite linear alternate; +} + +.avatar { + animation-delay: 0.5s; } .avatar::before { left: -5px; + animation-delay: 0s; } .avatar::after { left: 5px; + animation-delay: 1s; } diff --git a/web/app/components/datasets/common/retrieval-method-config/index.tsx b/web/app/components/datasets/common/retrieval-method-config/index.tsx index 57d357442f..ed230c52ce 100644 --- a/web/app/components/datasets/common/retrieval-method-config/index.tsx +++ b/web/app/components/datasets/common/retrieval-method-config/index.tsx @@ -40,7 +40,7 @@ const RetrievalMethodConfig: FC = ({ onChange({ ...value, search_method: retrieveMethod, - ...(!value.reranking_model.reranking_model_name + ...((!value.reranking_model.reranking_model_name || !value.reranking_model.reranking_provider_name) ? { reranking_model: { reranking_provider_name: isRerankDefaultModelValid ? rerankDefaultModel?.provider?.provider ?? '' : '', @@ -57,7 +57,7 @@ const RetrievalMethodConfig: FC = ({ onChange({ ...value, search_method: retrieveMethod, - ...(!value.reranking_model.reranking_model_name + ...((!value.reranking_model.reranking_model_name || !value.reranking_model.reranking_provider_name) ? { reranking_model: { reranking_provider_name: isRerankDefaultModelValid ? rerankDefaultModel?.provider?.provider ?? '' : '', diff --git a/web/app/components/datasets/common/retrieval-param-config/index.tsx b/web/app/components/datasets/common/retrieval-param-config/index.tsx index 216a56ab16..0c28149d56 100644 --- a/web/app/components/datasets/common/retrieval-param-config/index.tsx +++ b/web/app/components/datasets/common/retrieval-param-config/index.tsx @@ -54,7 +54,7 @@ const RetrievalParamConfig: FC = ({ }, ) - const handleDisabledSwitchClick = useCallback((enable: boolean) => { + const handleToggleRerankEnable = useCallback((enable: boolean) => { if (enable && !currentModel) Toast.notify({ type: 'error', message: t('workflow.errorMsg.rerankModelRequired') }) onChange({ @@ -119,7 +119,7 @@ const RetrievalParamConfig: FC = ({ )}
diff --git a/web/app/components/header/account-setting/data-source-page-new/install-from-marketplace.tsx b/web/app/components/header/account-setting/data-source-page-new/install-from-marketplace.tsx index 4c0de924d1..f4f7749f7f 100644 --- a/web/app/components/header/account-setting/data-source-page-new/install-from-marketplace.tsx +++ b/web/app/components/header/account-setting/data-source-page-new/install-from-marketplace.tsx @@ -52,7 +52,7 @@ const InstallFromMarketplace = ({
setCollapse(!collapse)}> - {t('common.modelProvider.installProvider')} + {t('common.modelProvider.installDataSourceProvider')}
{t('common.modelProvider.discoverMore')} diff --git a/web/app/components/header/account-setting/model-provider-page/hooks.ts b/web/app/components/header/account-setting/model-provider-page/hooks.ts index b10aeeb47e..48dc609795 100644 --- a/web/app/components/header/account-setting/model-provider-page/hooks.ts +++ b/web/app/components/header/account-setting/model-provider-page/hooks.ts @@ -323,15 +323,18 @@ export const useRefreshModel = () => { const { eventEmitter } = useEventEmitterContextContext() const updateModelProviders = useUpdateModelProviders() const updateModelList = useUpdateModelList() - const handleRefreshModel = useCallback((provider: ModelProvider, configurationMethod: ConfigurationMethodEnum, CustomConfigurationModelFixedFields?: CustomConfigurationModelFixedFields) => { + const handleRefreshModel = useCallback(( + provider: ModelProvider, + CustomConfigurationModelFixedFields?: CustomConfigurationModelFixedFields, + refreshModelList?: boolean, + ) => { updateModelProviders() provider.supported_model_types.forEach((type) => { updateModelList(type) }) - if (configurationMethod === ConfigurationMethodEnum.customizableModel - && provider.custom_configuration.status === CustomConfigurationStatusEnum.active) { + if (refreshModelList && provider.custom_configuration.status === CustomConfigurationStatusEnum.active) { eventEmitter?.emit({ type: UPDATE_MODEL_PROVIDER_CUSTOM_MODEL_LIST, payload: provider.provider, diff --git a/web/app/components/header/account-setting/model-provider-page/model-auth/hooks/use-auth.ts b/web/app/components/header/account-setting/model-provider-page/model-auth/hooks/use-auth.ts index 14b21be7f7..3136a70563 100644 --- a/web/app/components/header/account-setting/model-provider-page/model-auth/hooks/use-auth.ts +++ b/web/app/components/header/account-setting/model-provider-page/model-auth/hooks/use-auth.ts @@ -90,7 +90,7 @@ export const useAuth = ( type: 'success', message: t('common.api.actionSuccess'), }) - handleRefreshModel(provider, configurationMethod, undefined) + handleRefreshModel(provider, undefined, true) } finally { handleSetDoingAction(false) @@ -125,7 +125,7 @@ export const useAuth = ( type: 'success', message: t('common.api.actionSuccess'), }) - handleRefreshModel(provider, configurationMethod, undefined) + handleRefreshModel(provider, undefined, true) onRemove?.(pendingOperationCredentialId.current ?? '') closeConfirmDelete() } @@ -147,7 +147,7 @@ export const useAuth = ( if (res.result === 'success') { notify({ type: 'success', message: t('common.actionMsg.modifiedSuccessfully') }) - handleRefreshModel(provider, configurationMethod, undefined) + handleRefreshModel(provider, undefined, !payload.credential_id) } } finally { diff --git a/web/app/components/header/account-setting/model-provider-page/provider-added-card/model-load-balancing-modal.tsx b/web/app/components/header/account-setting/model-provider-page/provider-added-card/model-load-balancing-modal.tsx index 070c2ee90f..090147897b 100644 --- a/web/app/components/header/account-setting/model-provider-page/provider-added-card/model-load-balancing-modal.tsx +++ b/web/app/components/header/account-setting/model-provider-page/provider-added-card/model-load-balancing-modal.tsx @@ -159,7 +159,7 @@ const ModelLoadBalancingModal = ({ ) if (res.result === 'success') { notify({ type: 'success', message: t('common.actionMsg.modifiedSuccessfully') }) - handleRefreshModel(provider, configurateMethod, currentCustomConfigurationModelFixedFields) + handleRefreshModel(provider, currentCustomConfigurationModelFixedFields, false) onSave?.(provider.provider) onClose?.() } diff --git a/web/app/components/tools/provider-list.tsx b/web/app/components/tools/provider-list.tsx index d267b49c79..08a4aa0b5d 100644 --- a/web/app/components/tools/provider-list.tsx +++ b/web/app/components/tools/provider-list.tsx @@ -17,7 +17,7 @@ import CardMoreInfo from '@/app/components/plugins/card/card-more-info' import PluginDetailPanel from '@/app/components/plugins/plugin-detail-panel' import MCPList from './mcp' import { useAllToolProviders } from '@/service/use-tools' -import { useInstalledPluginList, useInvalidateInstalledPluginList } from '@/service/use-plugins' +import { useCheckInstalled, useInvalidateInstalledPluginList } from '@/service/use-plugins' import { useGlobalPublicStore } from '@/context/global-public-context' import { ToolTypeEnum } from '../workflow/block-selector/types' import { useMarketplace } from './marketplace/hooks' @@ -77,12 +77,14 @@ const ProviderList = () => { const currentProvider = useMemo(() => { return filteredCollectionList.find(collection => collection.id === currentProviderId) }, [currentProviderId, filteredCollectionList]) - const { data: pluginList } = useInstalledPluginList() + const { data: checkedInstalledData } = useCheckInstalled({ + pluginIds: currentProvider?.plugin_id ? [currentProvider.plugin_id] : [], + enabled: !!currentProvider?.plugin_id, + }) const invalidateInstalledPluginList = useInvalidateInstalledPluginList() const currentPluginDetail = useMemo(() => { - const detail = pluginList?.plugins.find(plugin => plugin.plugin_id === currentProvider?.plugin_id) - return detail - }, [currentProvider?.plugin_id, pluginList?.plugins]) + return checkedInstalledData?.plugins?.[0] + }, [checkedInstalledData]) const toolListTailRef = useRef(null) const showMarketplacePanel = useCallback(() => { diff --git a/web/app/components/workflow/nodes/knowledge-base/components/option-card.tsx b/web/app/components/workflow/nodes/knowledge-base/components/option-card.tsx index c15157fc5c..789e24835f 100644 --- a/web/app/components/workflow/nodes/knowledge-base/components/option-card.tsx +++ b/web/app/components/workflow/nodes/knowledge-base/components/option-card.tsx @@ -86,7 +86,10 @@ const OptionCard = memo(({ readonly && 'cursor-not-allowed', wrapperClassName && (typeof wrapperClassName === 'function' ? wrapperClassName(isActive) : wrapperClassName), )} - onClick={() => !readonly && enableSelect && id && onClick?.(id)} + onClick={(e) => { + e.stopPropagation() + !readonly && enableSelect && id && onClick?.(id) + }} >
= { chunk_structure, indexing_technique, retrieval_model, + embedding_model, + embedding_model_provider, + index_chunk_variable_selector, } = payload + const { + search_method, + reranking_enable, + reranking_model, + } = retrieval_model || {} + if (!chunk_structure) { return { isValid: false, @@ -36,6 +46,13 @@ const nodeDefault: NodeDefault = { } } + if (index_chunk_variable_selector.length === 0) { + return { + isValid: false, + errorMessage: t('workflow.nodes.knowledgeBase.chunksVariableIsRequired'), + } + } + if (!indexing_technique) { return { isValid: false, @@ -43,13 +60,27 @@ const nodeDefault: NodeDefault = { } } - if (!retrieval_model || !retrieval_model.search_method) { + if (indexing_technique === IndexingType.QUALIFIED && (!embedding_model || !embedding_model_provider)) { + return { + isValid: false, + errorMessage: t('workflow.nodes.knowledgeBase.embeddingModelIsRequired'), + } + } + + if (!retrieval_model || !search_method) { return { isValid: false, errorMessage: t('workflow.nodes.knowledgeBase.retrievalSettingIsRequired'), } } + if (reranking_enable && (!reranking_model || !reranking_model.reranking_provider_name || !reranking_model.reranking_model_name)) { + return { + isValid: false, + errorMessage: t('workflow.nodes.knowledgeBase.rerankingModelIsRequired'), + } + } + return { isValid: true, errorMessage: '', diff --git a/web/app/components/workflow/nodes/knowledge-base/hooks/use-config.ts b/web/app/components/workflow/nodes/knowledge-base/hooks/use-config.ts index 365722feba..8b22704c5a 100644 --- a/web/app/components/workflow/nodes/knowledge-base/hooks/use-config.ts +++ b/web/app/components/workflow/nodes/knowledge-base/hooks/use-config.ts @@ -9,13 +9,17 @@ import { ChunkStructureEnum, IndexMethodEnum, RetrievalSearchMethodEnum, + WeightedScoreEnum, } from '../types' import type { - HybridSearchModeEnum, KnowledgeBaseNodeType, RerankingModel, } from '../types' +import { + HybridSearchModeEnum, +} from '../types' import { isHighQualitySearchMethod } from '../utils' +import { DEFAULT_WEIGHTED_SCORE, RerankingModeEnum } from '@/models/datasets' export const useConfig = (id: string) => { const store = useStoreApi() @@ -35,6 +39,25 @@ export const useConfig = (id: string) => { }) }, [id, handleNodeDataUpdateWithSyncDraft]) + const getDefaultWeights = useCallback(({ + embeddingModel, + embeddingModelProvider, + }: { + embeddingModel: string + embeddingModelProvider: string + }) => { + return { + vector_setting: { + vector_weight: DEFAULT_WEIGHTED_SCORE.other.semantic, + embedding_provider_name: embeddingModelProvider || '', + embedding_model_name: embeddingModel, + }, + keyword_setting: { + keyword_weight: DEFAULT_WEIGHTED_SCORE.other.keyword, + }, + } + }, []) + const handleChunkStructureChange = useCallback((chunkStructure: ChunkStructureEnum) => { const nodeData = getNodeData() const { @@ -80,39 +103,72 @@ export const useConfig = (id: string) => { embeddingModelProvider: string }) => { const nodeData = getNodeData() - handleNodeDataUpdate({ + const defaultWeights = getDefaultWeights({ + embeddingModel, + embeddingModelProvider, + }) + const changeData = { embedding_model: embeddingModel, embedding_model_provider: embeddingModelProvider, retrieval_model: { ...nodeData?.data.retrieval_model, - vector_setting: { - ...nodeData?.data.retrieval_model.vector_setting, - embedding_provider_name: embeddingModelProvider, - embedding_model_name: embeddingModel, - }, }, - }) - }, [getNodeData, handleNodeDataUpdate]) + } + if (changeData.retrieval_model.weights) { + changeData.retrieval_model = { + ...changeData.retrieval_model, + weights: { + ...changeData.retrieval_model.weights, + vector_setting: { + ...changeData.retrieval_model.weights.vector_setting, + embedding_provider_name: embeddingModelProvider, + embedding_model_name: embeddingModel, + }, + }, + } + } + else { + changeData.retrieval_model = { + ...changeData.retrieval_model, + weights: defaultWeights, + } + } + handleNodeDataUpdate(changeData) + }, [getNodeData, getDefaultWeights, handleNodeDataUpdate]) const handleRetrievalSearchMethodChange = useCallback((searchMethod: RetrievalSearchMethodEnum) => { const nodeData = getNodeData() - handleNodeDataUpdate({ + const changeData = { retrieval_model: { ...nodeData?.data.retrieval_model, search_method: searchMethod, + reranking_mode: nodeData?.data.retrieval_model.reranking_mode || RerankingModeEnum.RerankingModel, }, - }) + } + if (searchMethod === RetrievalSearchMethodEnum.hybrid) { + changeData.retrieval_model = { + ...changeData.retrieval_model, + reranking_enable: changeData.retrieval_model.reranking_mode === RerankingModeEnum.RerankingModel, + } + } + handleNodeDataUpdate(changeData) }, [getNodeData, handleNodeDataUpdate]) const handleHybridSearchModeChange = useCallback((hybridSearchMode: HybridSearchModeEnum) => { const nodeData = getNodeData() + const defaultWeights = getDefaultWeights({ + embeddingModel: nodeData?.data.embedding_model || '', + embeddingModelProvider: nodeData?.data.embedding_model_provider || '', + }) handleNodeDataUpdate({ retrieval_model: { ...nodeData?.data.retrieval_model, reranking_mode: hybridSearchMode, + reranking_enable: hybridSearchMode === HybridSearchModeEnum.RerankingModel, + weights: nodeData?.data.retrieval_model.weights || defaultWeights, }, }) - }, [getNodeData, handleNodeDataUpdate]) + }, [getNodeData, getDefaultWeights, handleNodeDataUpdate]) const handleRerankingModelEnabledChange = useCallback((rerankingModelEnabled: boolean) => { const nodeData = getNodeData() @@ -130,11 +186,10 @@ export const useConfig = (id: string) => { retrieval_model: { ...nodeData?.data.retrieval_model, weights: { - weight_type: 'weighted_score', + weight_type: WeightedScoreEnum.Customized, vector_setting: { + ...nodeData?.data.retrieval_model.weights?.vector_setting, vector_weight: weightedScore.value[0], - embedding_provider_name: '', - embedding_model_name: '', }, keyword_setting: { keyword_weight: weightedScore.value[1], diff --git a/web/app/components/workflow/nodes/knowledge-base/node.tsx b/web/app/components/workflow/nodes/knowledge-base/node.tsx index 32810f4d3a..29de1bce9e 100644 --- a/web/app/components/workflow/nodes/knowledge-base/node.tsx +++ b/web/app/components/workflow/nodes/knowledge-base/node.tsx @@ -28,9 +28,9 @@ const Node: FC> = ({ data }) => {
- {settingsDisplay[data.retrieval_model.search_method as keyof typeof settingsDisplay]} + {settingsDisplay[data.retrieval_model?.search_method as keyof typeof settingsDisplay]}
diff --git a/web/app/components/workflow/nodes/knowledge-retrieval/components/retrieval-config.tsx b/web/app/components/workflow/nodes/knowledge-retrieval/components/retrieval-config.tsx index 8a3dc1efba..619216d672 100644 --- a/web/app/components/workflow/nodes/knowledge-retrieval/components/retrieval-config.tsx +++ b/web/app/components/workflow/nodes/knowledge-retrieval/components/retrieval-config.tsx @@ -1,6 +1,6 @@ 'use client' import type { FC } from 'react' -import React, { useCallback, useState } from 'react' +import React, { useCallback, useMemo } from 'react' import { RiEqualizer2Line } from '@remixicon/react' import { useTranslation } from 'react-i18next' import type { MultipleRetrievalConfig, SingleRetrievalConfig } from '../types' @@ -14,8 +14,6 @@ import { import ConfigRetrievalContent from '@/app/components/app/configuration/dataset-config/params-config/config-content' import { RETRIEVE_TYPE } from '@/types/app' import { DATASET_DEFAULT } from '@/config' -import { useModelListAndDefaultModelAndCurrentProviderAndModel } from '@/app/components/header/account-setting/model-provider-page/hooks' -import { ModelTypeEnum } from '@/app/components/header/account-setting/model-provider-page/declarations' import Button from '@/app/components/base/button' import type { DatasetConfigs } from '@/models/debug' import type { DataSet } from '@/models/datasets' @@ -32,8 +30,8 @@ type Props = { onSingleRetrievalModelChange?: (config: ModelConfig) => void onSingleRetrievalModelParamsChange?: (config: ModelConfig) => void readonly?: boolean - openFromProps?: boolean - onOpenFromPropsChange?: (openFromProps: boolean) => void + rerankModalOpen: boolean + onRerankModelOpenChange: (open: boolean) => void selectedDatasets: DataSet[] } @@ -45,26 +43,52 @@ const RetrievalConfig: FC = ({ onSingleRetrievalModelChange, onSingleRetrievalModelParamsChange, readonly, - openFromProps, - onOpenFromPropsChange, + rerankModalOpen, + onRerankModelOpenChange, selectedDatasets, }) => { const { t } = useTranslation() - const [open, setOpen] = useState(false) - const mergedOpen = openFromProps !== undefined ? openFromProps : open + const { retrieval_mode, multiple_retrieval_config } = payload const handleOpen = useCallback((newOpen: boolean) => { - setOpen(newOpen) - onOpenFromPropsChange?.(newOpen) - }, [onOpenFromPropsChange]) + onRerankModelOpenChange(newOpen) + }, [onRerankModelOpenChange]) - const { - currentProvider: validRerankDefaultProvider, - currentModel: validRerankDefaultModel, - } = useModelListAndDefaultModelAndCurrentProviderAndModel(ModelTypeEnum.rerank) + const datasetConfigs = useMemo(() => { + const { + reranking_model, + top_k, + score_threshold, + reranking_mode, + weights, + reranking_enable, + } = multiple_retrieval_config || {} + + return { + retrieval_model: retrieval_mode, + reranking_model: (reranking_model?.provider && reranking_model?.model) + ? { + reranking_provider_name: reranking_model?.provider, + reranking_model_name: reranking_model?.model, + } + : { + reranking_provider_name: '', + reranking_model_name: '', + }, + top_k: top_k || DATASET_DEFAULT.top_k, + score_threshold_enabled: !(score_threshold === undefined || score_threshold === null), + score_threshold, + datasets: { + datasets: [], + }, + reranking_mode, + weights, + reranking_enable, + } + }, [retrieval_mode, multiple_retrieval_config]) - const { multiple_retrieval_config } = payload const handleChange = useCallback((configs: DatasetConfigs, isRetrievalModeChange?: boolean) => { + // Legacy code, for compatibility, have to keep it if (isRetrievalModeChange) { onRetrievalModeChange(configs.retrieval_model) return @@ -72,13 +96,11 @@ const RetrievalConfig: FC = ({ onMultipleRetrievalConfigChange({ top_k: configs.top_k, score_threshold: configs.score_threshold_enabled ? (configs.score_threshold ?? DATASET_DEFAULT.score_threshold) : null, - reranking_model: payload.retrieval_mode === RETRIEVE_TYPE.oneWay + reranking_model: retrieval_mode === RETRIEVE_TYPE.oneWay ? undefined + // eslint-disable-next-line sonarjs/no-nested-conditional : (!configs.reranking_model?.reranking_provider_name - ? { - provider: validRerankDefaultProvider?.provider || '', - model: validRerankDefaultModel?.model || '', - } + ? undefined : { provider: configs.reranking_model?.reranking_provider_name, model: configs.reranking_model?.reranking_model_name, @@ -87,11 +109,11 @@ const RetrievalConfig: FC = ({ weights: configs.weights, reranking_enable: configs.reranking_enable, }) - }, [onMultipleRetrievalConfigChange, payload.retrieval_mode, validRerankDefaultProvider, validRerankDefaultModel, onRetrievalModeChange]) + }, [onMultipleRetrievalConfigChange, retrieval_mode, onRetrievalModeChange]) return ( = ({ onClick={() => { if (readonly) return - handleOpen(!mergedOpen) + handleOpen(!rerankModalOpen) }} >