Merge branch 'main' into feat/plugin

.github/DISCUSSION_TEMPLATE/general.yml

@@ -9,7 +9,7 @@ body:
           required: true
         - label: I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).
           required: true
-        - label: "请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
+        - label: "[FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
           required: true
         - label: "Please do not modify this template :) and fill in all the required fields."
           required: true

.github/DISCUSSION_TEMPLATE/help.yml

@@ -9,7 +9,7 @@ body:
           required: true
         - label: I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).
           required: true
-        - label: "请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
+        - label: "[FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
           required: true
         - label: "Please do not modify this template :) and fill in all the required fields."
           required: true

.github/DISCUSSION_TEMPLATE/suggestion.yml

@@ -9,7 +9,7 @@ body:
           required: true
         - label: I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).
           required: true
-        - label: "请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
+        - label: "[FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
           required: true
         - label: "Please do not modify this template :) and fill in all the required fields."
           required: true

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -14,7 +14,7 @@ body:
           required: true
         - label: I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).
           required: true
-        - label: "请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
+        - label: "[FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
           required: true
         - label: "Please do not modify this template :) and fill in all the required fields."
           required: true
@@ -22,7 +22,6 @@ body:
   - type: input
     attributes:
       label: Dify version
-      placeholder: 0.6.15
       description: See about section in Dify console
     validations:
       required: true

.github/ISSUE_TEMPLATE/document_issue.yml

@@ -12,7 +12,7 @@ body:
           required: true
         - label: I confirm that I am using English to submit report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).
           required: true
-        - label: "请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
+        - label: "[FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
           required: true
         - label: "Please do not modify this template :) and fill in all the required fields."
           required: true

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -12,7 +12,7 @@ body:
           required: true
         - label: I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).
           required: true
-        - label: "请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
+        - label: "[FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
           required: true
         - label: "Please do not modify this template :) and fill in all the required fields."
           required: true

.github/ISSUE_TEMPLATE/translation_issue.yml

@@ -12,14 +12,13 @@ body:
           required: true
         - label: I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).
           required: true
-        - label: "请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
+        - label: "[FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）"
           required: true
         - label: "Please do not modify this template :) and fill in all the required fields."
           required: true
   - type: input
     attributes:
       label: Dify version
-      placeholder: 0.3.21
       description: Hover over system tray icon or look at Settings
     validations:
       required: true

.gitignore

@@ -155,6 +155,7 @@ docker-legacy/volumes/milvus/*
 docker-legacy/volumes/chroma/*
 
 docker/volumes/app/storage/*
+docker/volumes/certbot/*
 docker/volumes/db/data/*
 docker/volumes/redis/data/*
 docker/volumes/weaviate/*

api/app.py

@@ -261,6 +261,7 @@ def after_request(response):
 @app.route('/health')
 def health():
     return Response(json.dumps({
+        'pid': os.getpid(),
         'status': 'ok',
         'version': app.config['CURRENT_VERSION']
     }), status=200, content_type="application/json")
@@ -284,6 +285,7 @@ def threads():
         })
 
     return {
+        'pid': os.getpid(),
         'thread_num': num_threads,
         'threads': thread_list
     }
@@ -293,6 +295,7 @@ def threads():
 def pool_stat():
     engine = db.engine
     return {
+        'pid': os.getpid(),
         'pool_size': engine.pool.size(),
         'checked_in_connections': engine.pool.checkedin(),
         'checked_out_connections': engine.pool.checkedout(),

api/commands.py

@@ -249,8 +249,7 @@ def migrate_knowledge_vector_database():
     create_count = 0
     skipped_count = 0
     total_count = 0
-    config = current_app.config
-    vector_type = config.get('VECTOR_STORE')
+    vector_type = dify_config.VECTOR_STORE
     page = 1
     while True:
         try:
@@ -484,8 +483,7 @@ def convert_to_agent_apps():
 @click.option('--field', default='metadata.doc_id', prompt=False, help='index field , default is metadata.doc_id.')
 def add_qdrant_doc_id_index(field: str):
     click.echo(click.style('Start add qdrant doc_id index.', fg='green'))
-    config = current_app.config
-    vector_type = config.get('VECTOR_STORE')
+    vector_type = dify_config.VECTOR_STORE
     if vector_type != "qdrant":
         click.echo(click.style('Sorry, only support qdrant vector store.', fg='red'))
         return
@@ -502,13 +500,15 @@ def add_qdrant_doc_id_index(field: str):
 
         from core.rag.datasource.vdb.qdrant.qdrant_vector import QdrantConfig
         for binding in bindings:
+            if dify_config.QDRANT_URL is None:
+                raise ValueError('Qdrant url is required.')
             qdrant_config = QdrantConfig(
-                endpoint=config.get('QDRANT_URL'),
-                api_key=config.get('QDRANT_API_KEY'),
+                endpoint=dify_config.QDRANT_URL,
+                api_key=dify_config.QDRANT_API_KEY,
                 root_path=current_app.root_path,
-                timeout=config.get('QDRANT_CLIENT_TIMEOUT'),
-                grpc_port=config.get('QDRANT_GRPC_PORT'),
-                prefer_grpc=config.get('QDRANT_GRPC_ENABLED')
+                timeout=dify_config.QDRANT_CLIENT_TIMEOUT,
+                grpc_port=dify_config.QDRANT_GRPC_PORT,
+                prefer_grpc=dify_config.QDRANT_GRPC_ENABLED
             )
             try:
                 client = qdrant_client.QdrantClient(**qdrant_config.to_qdrant_params())

api/configs/app_config.py

@@ -64,4 +64,6 @@ class DifyConfig(
         return f'{self.HTTP_REQUEST_NODE_MAX_TEXT_SIZE / 1024 / 1024:.2f}MB'
 
     SSRF_PROXY_HTTP_URL: str | None = None
-    SSRF_PROXY_HTTPS_URL: str | None = None
+    SSRF_PROXY_HTTPS_URL: str | None = None
+
+    MODERATION_BUFFER_SIZE: int = Field(default=300, description='The buffer size for moderation.')

api/controllers/console/auth/login.py

@@ -71,7 +71,7 @@ class ResetPasswordApi(Resource):
         # AccountService.update_password(account, new_password)
 
         # todo: Send email
-        # MAILCHIMP_API_KEY = current_app.config['MAILCHIMP_TRANSACTIONAL_API_KEY']
+        # MAILCHIMP_API_KEY = dify_config.MAILCHIMP_TRANSACTIONAL_API_KEY
         # mailchimp = MailchimpTransactional(MAILCHIMP_API_KEY)
 
         # message = {
@@ -92,7 +92,7 @@ class ResetPasswordApi(Resource):
         #     'message': message,
         #     # required for transactional email
         #     ' settings': {
-        #         'sandbox_mode': current_app.config['MAILCHIMP_SANDBOX_MODE'],
+        #         'sandbox_mode': dify_config.MAILCHIMP_SANDBOX_MODE,
         #     },
         # })
 

api/controllers/service_api/app/workflow.py

@@ -29,22 +29,21 @@ from services.app_generate_service import AppGenerateService
 
 logger = logging.getLogger(__name__)
 
-
-class WorkflowRunApi(Resource):
-    workflow_run_fields = {
-        'id': fields.String,
-        'workflow_id': fields.String,
-        'status': fields.String,
-        'inputs': fields.Raw,
-        'outputs': fields.Raw,
-        'error': fields.String,
-        'total_steps': fields.Integer,
-        'total_tokens': fields.Integer,
-        'created_at': fields.DateTime,
-        'finished_at': fields.DateTime,
-        'elapsed_time': fields.Float,
-    }
-
+workflow_run_fields = {
+    'id': fields.String,
+    'workflow_id': fields.String,
+    'status': fields.String,
+    'inputs': fields.Raw,
+    'outputs': fields.Raw,
+    'error': fields.String,
+    'total_steps': fields.Integer,
+    'total_tokens': fields.Integer,
+    'created_at': fields.DateTime,
+    'finished_at': fields.DateTime,
+    'elapsed_time': fields.Float,
+}
+
+class WorkflowRunDetailApi(Resource):
     @validate_app_token
     @marshal_with(workflow_run_fields)
     def get(self, app_model: App, workflow_id: str):
@@ -57,7 +56,7 @@ class WorkflowRunApi(Resource):
 
         workflow_run = db.session.query(WorkflowRun).filter(WorkflowRun.id == workflow_id).first()
         return workflow_run
-
+class WorkflowRunApi(Resource):
     @validate_app_token(fetch_user_arg=FetchUserArg(fetch_from=WhereisUserArg.JSON, required=True))
     def post(self, app_model: App, end_user: EndUser):
         """
@@ -117,5 +116,6 @@ class WorkflowTaskStopApi(Resource):
         }
 
 
-api.add_resource(WorkflowRunApi, '/workflows/run/<string:workflow_id>', '/workflows/run')
+api.add_resource(WorkflowRunApi, '/workflows/run')
+api.add_resource(WorkflowRunDetailApi, '/workflows/run/<string:workflow_id>')
 api.add_resource(WorkflowTaskStopApi, '/workflows/tasks/<string:task_id>/stop')

api/core/indexing_runner.py

@@ -12,6 +12,7 @@ from flask import Flask, current_app
 from flask_login import current_user
 from sqlalchemy.orm.exc import ObjectDeletedError
 
+from configs import dify_config
 from core.errors.error import ProviderTokenNotInitError
 from core.llm_generator.llm_generator import LLMGenerator
 from core.model_manager import ModelInstance, ModelManager
@@ -224,7 +225,7 @@ class IndexingRunner:
         features = FeatureService.get_features(tenant_id)
         if features.billing.enabled:
             count = len(extract_settings)
-            batch_upload_limit = int(current_app.config['BATCH_UPLOAD_LIMIT'])
+            batch_upload_limit = dify_config.BATCH_UPLOAD_LIMIT
             if count > batch_upload_limit:
                 raise ValueError(f"You have reached the batch upload limit of {batch_upload_limit}.")
 
@@ -427,7 +428,7 @@ class IndexingRunner:
             # The user-defined segmentation rule
             rules = json.loads(processing_rule.rules)
             segmentation = rules["segmentation"]
-            max_segmentation_tokens_length = int(current_app.config['INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH'])
+            max_segmentation_tokens_length = dify_config.INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH
             if segmentation["max_tokens"] < 50 or segmentation["max_tokens"] > max_segmentation_tokens_length:
                 raise ValueError(f"Custom segment length should be between 50 and {max_segmentation_tokens_length}.")
 

api/core/model_runtime/model_providers/bedrock/llm/_position.yaml

@@ -12,6 +12,7 @@
 - cohere.command-r-v1.0
 - meta.llama3-1-8b-instruct-v1:0
 - meta.llama3-1-70b-instruct-v1:0
+- meta.llama3-1-405b-instruct-v1:0
 - meta.llama3-8b-instruct-v1:0
 - meta.llama3-70b-instruct-v1:0
 - meta.llama2-13b-chat-v1

api/core/model_runtime/model_providers/bedrock/llm/cohere.command-r-plus-v1.0.yaml

@@ -3,8 +3,7 @@ label:
   en_US: Command R+
 model_type: llm
 features:
-  #- multi-tool-call
-  - agent-thought
+  - tool-call
   #- stream-tool-call
 model_properties:
   mode: chat

api/core/model_runtime/model_providers/bedrock/llm/cohere.command-r-v1.0.yaml

@@ -3,9 +3,7 @@ label:
   en_US: Command R
 model_type: llm
 features:
-  #- multi-tool-call
-  - agent-thought
-  #- stream-tool-call
+  - tool-call
 model_properties:
   mode: chat
   context_size: 128000

api/core/model_runtime/model_providers/bedrock/llm/llm.py

@@ -17,7 +17,6 @@ from botocore.exceptions import (
     ServiceNotInRegionError,
     UnknownServiceError,
 )
-from cohere import ChatMessage
 
 # local import
 from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta
@@ -42,7 +41,6 @@ from core.model_runtime.errors.invoke import (
 )
 from core.model_runtime.errors.validate import CredentialsValidateFailedError
 from core.model_runtime.model_providers.__base.large_language_model import LargeLanguageModel
-from core.model_runtime.model_providers.cohere.llm.llm import CohereLargeLanguageModel
 
 logger = logging.getLogger(__name__)
 
@@ -59,6 +57,7 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         {'prefix': 'mistral.mixtral-8x7b-instruct', 'support_system_prompts': False, 'support_tool_use': False},
         {'prefix': 'mistral.mistral-large', 'support_system_prompts': True, 'support_tool_use': True},
         {'prefix': 'mistral.mistral-small', 'support_system_prompts': True, 'support_tool_use': True},
+        {'prefix': 'cohere.command-r', 'support_system_prompts': True, 'support_tool_use': True},
         {'prefix': 'amazon.titan', 'support_system_prompts': False, 'support_tool_use': False}
     ]
 
@@ -94,86 +93,8 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
             model_info['model'] = model
             # invoke models via boto3 converse API
             return self._generate_with_converse(model_info, credentials, prompt_messages, model_parameters, stop, stream, user, tools)
-        # invoke Cohere models via boto3 client
-        if "cohere.command-r" in model:
-            return self._generate_cohere_chat(model, credentials, prompt_messages, model_parameters, stop, stream, user, tools)
         # invoke other models via boto3 client
         return self._generate(model, credentials, prompt_messages, model_parameters, stop, stream, user)
-    
-    def _generate_cohere_chat(
-            self, model: str, credentials: dict, prompt_messages: list[PromptMessage], model_parameters: dict,
-            stop: Optional[list[str]] = None, stream: bool = True, user: Optional[str] = None,
-            tools: Optional[list[PromptMessageTool]] = None,) -> Union[LLMResult, Generator]:
-        cohere_llm = CohereLargeLanguageModel()
-        client_config = Config(
-            region_name=credentials["aws_region"]
-        )
-
-        runtime_client = boto3.client(
-            service_name='bedrock-runtime',
-            config=client_config,
-            aws_access_key_id=credentials["aws_access_key_id"],
-            aws_secret_access_key=credentials["aws_secret_access_key"]
-        )
-
-        extra_model_kwargs = {}
-        if stop:
-            extra_model_kwargs['stop_sequences'] = stop
-
-        if tools:
-            tools = cohere_llm._convert_tools(tools)
-            model_parameters['tools'] = tools
-
-        message, chat_histories, tool_results \
-            = cohere_llm._convert_prompt_messages_to_message_and_chat_histories(prompt_messages)
-
-        if tool_results:
-            model_parameters['tool_results'] = tool_results
-
-        payload = {
-            **model_parameters,
-            "message": message,
-            "chat_history": chat_histories,
-        }
-
-        # need workaround for ai21 models which doesn't support streaming
-        if stream:
-            invoke = runtime_client.invoke_model_with_response_stream
-        else:
-            invoke = runtime_client.invoke_model
-
-        def serialize(obj):
-            if isinstance(obj, ChatMessage):
-                return obj.__dict__
-            raise TypeError(f"Type {type(obj)} not serializable")
-
-        try:
-            body_jsonstr=json.dumps(payload, default=serialize)
-            response = invoke(
-                modelId=model,
-                contentType="application/json",
-                accept="*/*",
-                body=body_jsonstr
-            )
-        except ClientError as ex:
-            error_code = ex.response['Error']['Code']
-            full_error_msg = f"{error_code}: {ex.response['Error']['Message']}"
-            raise self._map_client_to_invoke_error(error_code, full_error_msg)
-
-        except (EndpointConnectionError, NoRegionError, ServiceNotInRegionError) as ex:
-            raise InvokeConnectionError(str(ex))
-
-        except UnknownServiceError as ex:
-            raise InvokeServerUnavailableError(str(ex))
-
-        except Exception as ex:
-            raise InvokeError(str(ex))
-
-        if stream:
-            return self._handle_generate_stream_response(model, credentials, response, prompt_messages)
-
-        return self._handle_generate_response(model, credentials, response, prompt_messages)
-
 
     def _generate_with_converse(self, model_info: dict, credentials: dict, prompt_messages: list[PromptMessage], model_parameters: dict,
                 stop: Optional[list[str]] = None, stream: bool = True, user: Optional[str] = None, tools: Optional[list[PromptMessageTool]] = None,) -> Union[LLMResult, Generator]:
@@ -581,38 +502,9 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         :param message: PromptMessage to convert.
         :return: String representation of the message.
         """
-        
-        if model_prefix == "anthropic":
-            human_prompt_prefix = "\n\nHuman:"
-            human_prompt_postfix = ""
-            ai_prompt = "\n\nAssistant:"
-
-        elif model_prefix == "meta":
-            # LLAMA3
-            if model_name.startswith("llama3"):
-                human_prompt_prefix = "<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n"
-                human_prompt_postfix = "<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
-                ai_prompt = "\n\nAssistant:"
-            else:
-                # LLAMA2
-                human_prompt_prefix = "\n[INST]"
-                human_prompt_postfix = "[\\INST]\n"
-                ai_prompt = ""
-
-        elif model_prefix == "mistral":
-            human_prompt_prefix = "<s>[INST]"
-            human_prompt_postfix = "[\\INST]\n"
-            ai_prompt = "\n\nAssistant:"
-
-        elif model_prefix == "amazon":
-            human_prompt_prefix = "\n\nUser:"
-            human_prompt_postfix = ""
-            ai_prompt = "\n\nBot:"
-        
-        else:
-            human_prompt_prefix = ""
-            human_prompt_postfix = ""
-            ai_prompt = ""
+        human_prompt_prefix = ""
+        human_prompt_postfix = ""
+        ai_prompt = ""
 
         content = message.content
 
@@ -663,13 +555,7 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         model_prefix = model.split('.')[0]
         model_name = model.split('.')[1]
 
-        if model_prefix == "amazon":
-            payload["textGenerationConfig"] = { **model_parameters }
-            payload["textGenerationConfig"]["stopSequences"] = ["User:"]
-            
-            payload["inputText"] = self._convert_messages_to_prompt(prompt_messages, model_prefix)
-        
-        elif model_prefix == "ai21":
+        if model_prefix == "ai21":
             payload["temperature"] = model_parameters.get("temperature")
             payload["topP"] = model_parameters.get("topP")
             payload["maxTokens"] = model_parameters.get("maxTokens")
@@ -681,28 +567,12 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
                 payload["frequencyPenalty"] = {model_parameters.get("frequencyPenalty")}
             if model_parameters.get("countPenalty"):
                 payload["countPenalty"] = {model_parameters.get("countPenalty")}
-        
-        elif model_prefix == "mistral":
-            payload["temperature"] = model_parameters.get("temperature")
-            payload["top_p"] = model_parameters.get("top_p")
-            payload["max_tokens"] = model_parameters.get("max_tokens")
-            payload["prompt"] = self._convert_messages_to_prompt(prompt_messages, model_prefix)
-            payload["stop"] = stop[:10] if stop else []
-
-        elif model_prefix == "anthropic":
-            payload = { **model_parameters }
-            payload["prompt"] = self._convert_messages_to_prompt(prompt_messages, model_prefix)
-            payload["stop_sequences"] = ["\n\nHuman:"] + (stop if stop else [])
-            
+                    
         elif model_prefix == "cohere":
             payload = { **model_parameters }
             payload["prompt"] = prompt_messages[0].content
             payload["stream"] = stream
         
-        elif model_prefix == "meta":
-            payload = { **model_parameters }
-            payload["prompt"] = self._convert_messages_to_prompt(prompt_messages, model_prefix, model_name)
-
         else:
             raise ValueError(f"Got unknown model prefix {model_prefix}")
         
@@ -793,36 +663,16 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         # get output text and calculate num tokens based on model / provider
         model_prefix = model.split('.')[0]
 
-        if model_prefix == "amazon":
-            output = response_body.get("results")[0].get("outputText").strip('\n')
-            prompt_tokens = response_body.get("inputTextTokenCount")
-            completion_tokens = response_body.get("results")[0].get("tokenCount")
-
-        elif model_prefix == "ai21":
+        if model_prefix == "ai21":
             output = response_body.get('completions')[0].get('data').get('text')
             prompt_tokens = len(response_body.get("prompt").get("tokens"))
             completion_tokens = len(response_body.get('completions')[0].get('data').get('tokens'))
-
-        elif model_prefix == "anthropic":
-            output = response_body.get("completion")
-            prompt_tokens = self.get_num_tokens(model, credentials, prompt_messages)
-            completion_tokens = self.get_num_tokens(model, credentials, output if output else '')
             
         elif model_prefix == "cohere":
             output = response_body.get("generations")[0].get("text")
             prompt_tokens = self.get_num_tokens(model, credentials, prompt_messages)
             completion_tokens = self.get_num_tokens(model, credentials, output if output else '')
-            
-        elif model_prefix == "meta":
-            output = response_body.get("generation").strip('\n')
-            prompt_tokens = response_body.get("prompt_token_count")
-            completion_tokens = response_body.get("generation_token_count")
-        
-        elif model_prefix == "mistral":
-            output = response_body.get("outputs")[0].get("text")
-            prompt_tokens = response.get('ResponseMetadata').get('HTTPHeaders').get('x-amzn-bedrock-input-token-count')
-            completion_tokens = response.get('ResponseMetadata').get('HTTPHeaders').get('x-amzn-bedrock-output-token-count')
-
+                        
         else:
             raise ValueError(f"Got unknown model prefix {model_prefix} when handling block response")
 
@@ -893,26 +743,10 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
             payload = json.loads(chunk.get('bytes').decode())
 
             model_prefix = model.split('.')[0]
-            if model_prefix == "amazon":
-                content_delta = payload.get("outputText").strip('\n')
-                finish_reason = payload.get("completion_reason")
- 
-            elif model_prefix == "anthropic":
-                content_delta = payload.get("completion")
-                finish_reason = payload.get("stop_reason")
-
-            elif model_prefix == "cohere":
+            if model_prefix == "cohere":
                 content_delta = payload.get("text")
                 finish_reason = payload.get("finish_reason")
             
-            elif model_prefix == "mistral":
-                content_delta = payload.get('outputs')[0].get("text")
-                finish_reason = payload.get('outputs')[0].get("stop_reason")
-
-            elif model_prefix == "meta":
-                content_delta = payload.get("generation").strip('\n')
-                finish_reason = payload.get("stop_reason")
-            
             else:
                 raise ValueError(f"Got unknown model prefix {model_prefix} when handling stream response")
 

api/core/model_runtime/model_providers/bedrock/llm/meta.llama3-1-405b-instruct-v1.0.yaml

@@ -0,0 +1,25 @@
+model: meta.llama3-1-405b-instruct-v1:0
+label:
+  en_US: Llama 3.1 405B Instruct
+model_type: llm
+model_properties:
+  mode: completion
+  context_size: 128000
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+    default: 0.5
+  - name: top_p
+    use_template: top_p
+    default: 0.9
+  - name: max_gen_len
+    use_template: max_tokens
+    required: true
+    default: 512
+    min: 1
+    max: 2048
+pricing:
+  input: '0.00532'
+  output: '0.016'
+  unit: '0.001'
+  currency: USD

api/core/model_runtime/model_providers/deepseek/llm/deepseek-chat.yaml

@@ -23,7 +23,7 @@ parameter_rules:
     type: int
     default: 4096
     min: 1
-    max: 4096
+    max: 8192
     help:
       zh_Hans: 指定生成结果长度的上限。如果生成结果截断，可以调大该参数。
       en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter.
@@ -57,6 +57,18 @@ parameter_rules:
     help:
       zh_Hans: 介于 -2.0 和 2.0 之间的数字。如果该值为正，那么新 token 会根据其在已有文本中的出现频率受到相应的惩罚，降低模型重复相同内容的可能性。
       en_US: A number between -2.0 and 2.0. If the value is positive, new tokens are penalized based on their frequency of occurrence in existing text, reducing the likelihood that the model will repeat the same content.
+  - name: response_format
+    label:
+      zh_Hans: 回复格式
+      en_US: response_format
+    type: string
+    help:
+      zh_Hans: 指定模型必须输出的格式
+      en_US: specifying the format that the model must output
+    required: false
+    options:
+      - text
+      - json_object
 pricing:
   input: '1'
   output: '2'

api/core/model_runtime/model_providers/hunyuan/hunyuan.yaml

@@ -18,6 +18,7 @@ help:
     en_US: https://console.cloud.tencent.com/cam/capi
 supported_model_types:
   - llm
+  - text-embedding
 configurate_methods:
   - predefined-model
 provider_credential_schema:

api/core/model_runtime/model_providers/hunyuan/text_embedding/hunyuan-text-embedding.yaml

@@ -0,0 +1,5 @@
+model: hunyuan-embedding
+model_type: text-embedding
+model_properties:
+  context_size: 1024
+  max_chunks: 1

api/core/model_runtime/model_providers/hunyuan/text_embedding/text_embedding.py

@@ -0,0 +1,173 @@
+import json
+import logging
+import time
+from typing import Optional
+
+from tencentcloud.common import credential
+from tencentcloud.common.exception import TencentCloudSDKException
+from tencentcloud.common.profile.client_profile import ClientProfile
+from tencentcloud.common.profile.http_profile import HttpProfile
+from tencentcloud.hunyuan.v20230901 import hunyuan_client, models
+
+from core.model_runtime.entities.model_entities import PriceType
+from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
+from core.model_runtime.errors.invoke import (
+    InvokeError,
+)
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.__base.text_embedding_model import TextEmbeddingModel
+
+logger = logging.getLogger(__name__)
+
+class HunyuanTextEmbeddingModel(TextEmbeddingModel):
+    """
+    Model class for Hunyuan text embedding model.
+    """
+
+    def _invoke(self, model: str, credentials: dict,
+                texts: list[str], user: Optional[str] = None) \
+            -> TextEmbeddingResult:
+        """
+        Invoke text embedding model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param texts: texts to embed
+        :param user: unique user id
+        :return: embeddings result
+        """
+
+        if model != 'hunyuan-embedding':
+            raise ValueError('Invalid model name')
+        
+        client = self._setup_hunyuan_client(credentials)
+
+        embeddings = []
+        token_usage = 0
+
+        for input in texts:
+            request = models.GetEmbeddingRequest()
+            params = {
+                "Input": input
+            }
+            request.from_json_string(json.dumps(params))
+            response = client.GetEmbedding(request)
+            usage = response.Usage.TotalTokens
+
+            embeddings.extend([data.Embedding for data in response.Data])
+            token_usage += usage
+
+        result = TextEmbeddingResult(
+            model=model,
+            embeddings=embeddings,
+            usage=self._calc_response_usage(
+                model=model,
+                credentials=credentials,
+                tokens=token_usage
+            )
+        )
+
+        return result
+
+    def validate_credentials(self, model: str, credentials: dict) -> None:
+        """
+        Validate credentials
+        """
+        try:
+            client = self._setup_hunyuan_client(credentials)
+
+            req = models.ChatCompletionsRequest()
+            params = {
+                "Model": model,
+                "Messages": [{
+                    "Role": "user",
+                    "Content": "hello"
+                }],
+                "TopP": 1,
+                "Temperature": 0,
+                "Stream": False
+            }
+            req.from_json_string(json.dumps(params))
+            client.ChatCompletions(req)
+        except Exception as e:
+            raise CredentialsValidateFailedError(f'Credentials validation failed: {e}')
+
+    def _setup_hunyuan_client(self, credentials):
+        secret_id = credentials['secret_id']
+        secret_key = credentials['secret_key']
+        cred = credential.Credential(secret_id, secret_key)
+        httpProfile = HttpProfile()
+        httpProfile.endpoint = "hunyuan.tencentcloudapi.com"
+        clientProfile = ClientProfile()
+        clientProfile.httpProfile = httpProfile
+        client = hunyuan_client.HunyuanClient(cred, "", clientProfile)
+        return client
+    
+    def _calc_response_usage(self, model: str, credentials: dict, tokens: int) -> EmbeddingUsage:
+        """
+        Calculate response usage
+
+        :param model: model name
+        :param credentials: model credentials
+        :param tokens: input tokens
+        :return: usage
+        """
+        # get input price info
+        input_price_info = self.get_price(
+            model=model,
+            credentials=credentials,
+            price_type=PriceType.INPUT,
+            tokens=tokens
+        )
+
+        # transform usage
+        usage = EmbeddingUsage(
+            tokens=tokens,
+            total_tokens=tokens,
+            unit_price=input_price_info.unit_price,
+            price_unit=input_price_info.unit,
+            total_price=input_price_info.total_amount,
+            currency=input_price_info.currency,
+            latency=time.perf_counter() - self.started_at
+        )
+
+        return usage
+    
+    @property
+    def _invoke_error_mapping(self) -> dict[type[InvokeError], list[type[Exception]]]:
+        """
+        Map model invoke error to unified error
+        The key is the error type thrown to the caller
+        The value is the error type thrown by the model,
+        which needs to be converted into a unified error type for the caller.
+
+        :return: Invoke error mapping
+        """
+        return {
+            InvokeError: [TencentCloudSDKException],
+        }
+    
+    def get_num_tokens(self, model: str, credentials: dict, texts: list[str]) -> int:
+        """
+        Get number of tokens for given prompt messages
+
+        :param model: model name
+        :param credentials: model credentials
+        :param texts: texts to embed
+        :return:
+        """
+        # client = self._setup_hunyuan_client(credentials)
+
+        num_tokens = 0
+        for text in texts:
+            num_tokens += self._get_num_tokens_by_gpt2(text)
+            # use client.GetTokenCount to get num tokens
+            # request = models.GetTokenCountRequest()
+            # params = {
+            #     "Prompt": text
+            # }
+            # request.from_json_string(json.dumps(params))
+            # response = client.GetTokenCount(request)
+            # num_tokens += response.TokenCount
+
+        return num_tokens

api/core/model_runtime/model_providers/siliconflow/llm/deepdeek-coder-v2-instruct.yaml

@@ -0,0 +1,30 @@
+model: deepseek-ai/DeepSeek-Coder-V2-Instruct
+label:
+  en_US: deepseek-ai/DeepSeek-Coder-V2-Instruct
+model_type: llm
+features:
+  - agent-thought
+model_properties:
+  mode: chat
+  context_size: 32768
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: max_tokens
+    use_template: max_tokens
+    type: int
+    default: 512
+    min: 1
+    max: 4096
+    help:
+      zh_Hans: 指定生成结果长度的上限。如果生成结果截断，可以调大该参数。
+      en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter.
+  - name: top_p
+    use_template: top_p
+  - name: frequency_penalty
+    use_template: frequency_penalty
+pricing:
+  input: '1.33'
+  output: '1.33'
+  unit: '0.000001'
+  currency: RMB

api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2-chat.yaml

@@ -1,11 +1,9 @@
 model: deepseek-ai/deepseek-v2-chat
 label:
-  en_US: deepseek-ai/deepseek-v2-chat
+  en_US: deepseek-ai/DeepSeek-V2-Chat
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 32768

api/core/model_runtime/model_providers/siliconflow/llm/glm4-9b-chat.yaml

@@ -1,11 +1,9 @@
 model: zhipuai/glm4-9B-chat
 label:
-  en_US: zhipuai/glm4-9B-chat
+  en_US: THUDM/glm-4-9b-chat
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 32768

api/core/model_runtime/model_providers/siliconflow/llm/qwen2-57b-a14b-instruct.yaml

@@ -1,11 +1,9 @@
 model: alibaba/Qwen2-57B-A14B-Instruct
 label:
-  en_US: alibaba/Qwen2-57B-A14B-Instruct
+  en_US: Qwen/Qwen2-57B-A14B-Instruct
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 32768

api/core/model_runtime/model_providers/siliconflow/llm/qwen2-72b-instruct.yaml

@@ -1,11 +1,9 @@
 model: alibaba/Qwen2-72B-Instruct
 label:
-  en_US: alibaba/Qwen2-72B-Instruct
+  en_US: Qwen/Qwen2-72B-Instruct
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 32768

api/core/model_runtime/model_providers/siliconflow/llm/qwen2-7b-instruct.yaml

@@ -1,11 +1,9 @@
 model: alibaba/Qwen2-7B-Instruct
 label:
-  en_US: alibaba/Qwen2-7B-Instruct
+  en_US: Qwen/Qwen2-7B-Instruct
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 32768

api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-34b-chat.yaml

@@ -1,11 +1,9 @@
 model: 01-ai/Yi-1.5-34B-Chat
 label:
-  en_US: 01-ai/Yi-1.5-34B-Chat
+  en_US: 01-ai/Yi-1.5-34B-Chat-16K
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 16384

api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-6b-chat.yaml

@@ -3,9 +3,7 @@ label:
   en_US: 01-ai/Yi-1.5-6B-Chat
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 4096

api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-9b-chat.yaml

@@ -1,11 +1,9 @@
 model: 01-ai/Yi-1.5-9B-Chat
 label:
-  en_US: 01-ai/Yi-1.5-9B-Chat
+  en_US: 01-ai/Yi-1.5-9B-Chat-16K
 model_type: llm
 features:
-  - multi-tool-call
   - agent-thought
-  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 16384

api/core/model_runtime/model_providers/siliconflow/siliconflow.py

@@ -19,7 +19,7 @@ class SiliconflowProvider(ModelProvider):
             model_instance = self.get_model_instance(ModelType.LLM)
 
             model_instance.validate_credentials(
-                model='deepseek-ai/deepseek-v2-chat',
+                model='deepseek-ai/DeepSeek-V2-Chat',
                 credentials=credentials
             )
         except CredentialsValidateFailedError as ex:

api/core/moderation/output_moderation.py

@@ -21,8 +21,6 @@ class ModerationRule(BaseModel):
 
 
 class OutputModeration(BaseModel):
-    DEFAULT_BUFFER_SIZE: int = 300
-
     tenant_id: str
     app_id: str
 
@@ -77,10 +75,10 @@ class OutputModeration(BaseModel):
         return final_output
 
     def start_thread(self) -> threading.Thread:
-        buffer_size = int(dify_config.config.MODERATION_BUFFER_SIZE)
+        buffer_size = dify_config.MODERATION_BUFFER_SIZE
         thread = threading.Thread(target=self.worker, kwargs={
             'flask_app': current_app._get_current_object(),
-            'buffer_size': buffer_size if buffer_size > 0 else self.DEFAULT_BUFFER_SIZE
+            'buffer_size': buffer_size if buffer_size > 0 else dify_config.MODERATION_BUFFER_SIZE
         })
 
         thread.start()

api/core/ops/ops_trace_manager.py

@@ -298,34 +298,29 @@ class TraceTask:
         self.file_base_url = os.getenv("FILES_URL", "http://127.0.0.1:5001")
 
     def execute(self):
-        method_name, trace_info = self.preprocess()
-        return trace_info
+        return self.preprocess()
 
     def preprocess(self):
-        if self.trace_type == TraceTaskName.CONVERSATION_TRACE:
-            return TraceTaskName.CONVERSATION_TRACE, self.conversation_trace(**self.kwargs)
-        if self.trace_type == TraceTaskName.WORKFLOW_TRACE:
-            return TraceTaskName.WORKFLOW_TRACE, self.workflow_trace(self.workflow_run, self.conversation_id)
-        elif self.trace_type == TraceTaskName.MESSAGE_TRACE:
-            return TraceTaskName.MESSAGE_TRACE, self.message_trace(self.message_id)
-        elif self.trace_type == TraceTaskName.MODERATION_TRACE:
-            return TraceTaskName.MODERATION_TRACE, self.moderation_trace(self.message_id, self.timer, **self.kwargs)
-        elif self.trace_type == TraceTaskName.SUGGESTED_QUESTION_TRACE:
-            return TraceTaskName.SUGGESTED_QUESTION_TRACE, self.suggested_question_trace(
+        preprocess_map = {
+            TraceTaskName.CONVERSATION_TRACE: lambda: self.conversation_trace(**self.kwargs),
+            TraceTaskName.WORKFLOW_TRACE: lambda: self.workflow_trace(self.workflow_run, self.conversation_id),
+            TraceTaskName.MESSAGE_TRACE: lambda: self.message_trace(self.message_id),
+            TraceTaskName.MODERATION_TRACE: lambda: self.moderation_trace(
                 self.message_id, self.timer, **self.kwargs
-            )
-        elif self.trace_type == TraceTaskName.DATASET_RETRIEVAL_TRACE:
-            return TraceTaskName.DATASET_RETRIEVAL_TRACE, self.dataset_retrieval_trace(
+            ),
+            TraceTaskName.SUGGESTED_QUESTION_TRACE: lambda: self.suggested_question_trace(
                 self.message_id, self.timer, **self.kwargs
-            )
-        elif self.trace_type == TraceTaskName.TOOL_TRACE:
-            return TraceTaskName.TOOL_TRACE, self.tool_trace(self.message_id, self.timer, **self.kwargs)
-        elif self.trace_type == TraceTaskName.GENERATE_NAME_TRACE:
-            return TraceTaskName.GENERATE_NAME_TRACE, self.generate_name_trace(
+            ),
+            TraceTaskName.DATASET_RETRIEVAL_TRACE: lambda: self.dataset_retrieval_trace(
+                self.message_id, self.timer, **self.kwargs
+            ),
+            TraceTaskName.TOOL_TRACE: lambda: self.tool_trace(self.message_id, self.timer, **self.kwargs),
+            TraceTaskName.GENERATE_NAME_TRACE: lambda: self.generate_name_trace(
                 self.conversation_id, self.timer, **self.kwargs
-            )
-        else:
-            return '', {}
+            ),
+        }
+
+        return preprocess_map.get(self.trace_type, lambda: None)()
 
     # process methods for different trace types
     def conversation_trace(self, **kwargs):

api/core/rag/datasource/keyword/jieba/jieba.py

@@ -2,9 +2,9 @@ import json
 from collections import defaultdict
 from typing import Any, Optional
 
-from flask import current_app
 from pydantic import BaseModel
 
+from configs import dify_config
 from core.rag.datasource.keyword.jieba.jieba_keyword_table_handler import JiebaKeywordTableHandler
 from core.rag.datasource.keyword.keyword_base import BaseKeyword
 from core.rag.models.document import Document
@@ -139,7 +139,7 @@ class Jieba(BaseKeyword):
             if keyword_table_dict:
                 return keyword_table_dict['__data__']['table']
         else:
-            keyword_data_source_type = current_app.config['KEYWORD_DATA_SOURCE_TYPE']
+            keyword_data_source_type = dify_config.KEYWORD_DATA_SOURCE_TYPE
             dataset_keyword_table = DatasetKeywordTable(
                 dataset_id=self.dataset.id,
                 keyword_table='',

api/core/rag/datasource/vdb/milvus/milvus_vector.py

@@ -5,6 +5,7 @@ from uuid import uuid4
 
 from pydantic import BaseModel, model_validator
 from pymilvus import MilvusClient, MilvusException, connections
+from pymilvus.milvus_client import IndexParams
 
 from configs import dify_config
 from core.rag.datasource.entity.embedding import Embeddings
@@ -250,11 +251,15 @@ class MilvusVector(BaseVector):
                 # Since primary field is auto-id, no need to track it
                 self._fields.remove(Field.PRIMARY_KEY.value)
 
+                # Create Index params for the collection
+                index_params_obj = IndexParams()
+                index_params_obj.add_index(field_name=Field.VECTOR.value, **index_params)
+
                 # Create the collection
                 collection_name = self._collection_name
-                self._client.create_collection_with_schema(collection_name=collection_name,
-                                                           schema=schema, index_param=index_params,
-                                                           consistency_level=self._consistency_level)
+                self._client.create_collection(collection_name=collection_name,
+                                               schema=schema, index_params=index_params_obj,
+                                               consistency_level=self._consistency_level)
             redis_client.set(collection_exist_cache_key, 1, ex=3600)
 
     def _init_client(self, config) -> MilvusClient:

api/core/tools/provider/builtin/aws/_assets/icon.svg

@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="utf-8"?>
+
<!-- Uploaded to: SVG Repo, www.svgrepo.com, Generator: SVG Repo Mixer Tools -->
+<svg width="800px" height="800px" viewBox="0 0 16 16" xmlns="http://www.w3.org/2000/svg" fill="none">
+
<path fill="#252F3E" d="M4.51 7.687c0 .197.02.357.058.475.042.117.096.245.17.384a.233.233 0 01.037.123c0 .053-.032.107-.1.16l-.336.224a.255.255 0 01-.138.048c-.054 0-.107-.026-.16-.074a1.652 1.652 0 01-.192-.251 4.137 4.137 0 01-.165-.315c-.415.491-.936.737-1.564.737-.447 0-.804-.129-1.064-.385-.261-.256-.394-.598-.394-1.025 0-.454.16-.822.484-1.1.325-.278.756-.416 1.304-.416.18 0 .367.016.564.042.197.027.4.07.612.118v-.39c0-.406-.085-.689-.25-.854-.17-.166-.458-.246-.868-.246-.186 0-.377.022-.574.07a4.23 4.23 0 00-.575.181 1.525 1.525 0 01-.186.07.326.326 0 01-.085.016c-.075 0-.112-.054-.112-.166v-.262c0-.085.01-.15.037-.186a.399.399 0 01.15-.113c.185-.096.409-.176.67-.24.26-.07.537-.101.83-.101.633 0 1.096.144 1.394.432.293.288.442.726.442 1.314v1.73h.01zm-2.161.811c.175 0 .356-.032.548-.096.191-.064.362-.182.505-.342a.848.848 0 00.181-.341c.032-.129.054-.283.054-.465V7.03a4.43 4.43 0 00-.49-.09 3.996 3.996 0 00-.5-.033c-.357 0-.618.07-.793.214-.176.144-.26.347-.26.614 0 .25.063.437.196.566.128.133.314.197.559.197zm4.273.577c-.096 0-.16-.016-.202-.054-.043-.032-.08-.106-.112-.208l-1.25-4.127a.938.938 0 01-.049-.214c0-.085.043-.133.128-.133h.522c.1 0 .17.016.207.053.043.032.075.107.107.208l.894 3.535.83-3.535c.026-.106.058-.176.1-.208a.365.365 0 01.214-.053h.425c.102 0 .17.016.213.053.043.032.08.107.101.208l.841 3.578.92-3.578a.458.458 0 01.107-.208.346.346 0 01.208-.053h.495c.085 0 .133.043.133.133 0 .027-.006.054-.01.086a.76.76 0 01-.038.133l-1.283 4.127c-.032.107-.069.177-.111.209a.34.34 0 01-.203.053h-.457c-.101 0-.17-.016-.213-.053-.043-.038-.08-.107-.101-.214L8.213 5.37l-.82 3.439c-.026.107-.058.176-.1.213-.043.038-.118.054-.213.054h-.458zm6.838.144a3.51 3.51 0 01-.82-.096c-.266-.064-.473-.134-.612-.214-.085-.048-.143-.101-.165-.15a.378.378 0 01-.031-.149v-.272c0-.112.042-.166.122-.166a.3.3 0 01.096.016c.032.011.08.032.133.054.18.08.378.144.585.187.213.042.42.064.633.064.336 0 .596-.059.777-.176a.575.575 0 00.277-.508.52.52 0 00-.144-.373c-.095-.102-.276-.193-.537-.278l-.772-.24c-.388-.123-.676-.305-.851-.545a1.275 1.275 0 01-.266-.774c0-.224.048-.422.143-.593.096-.17.224-.32.384-.438.16-.122.34-.213.553-.277.213-.064.436-.091.67-.091.118 0 .24.005.357.021.122.016.234.038.346.06.106.026.208.052.303.085.096.032.17.064.224.096a.46.46 0 01.16.133.289.289 0 01.047.176v.251c0 .112-.042.171-.122.171a.552.552 0 01-.202-.064 2.427 2.427 0 00-1.022-.208c-.303 0-.543.048-.708.15-.165.1-.25.256-.25.475 0 .149.053.277.16.379.106.101.303.202.585.293l.756.24c.383.123.66.294.825.513.165.219.244.47.244.748 0 .23-.047.437-.138.619a1.436 1.436 0 01-.388.47c-.165.133-.362.23-.591.299-.24.075-.49.112-.761.112z"/>
+
<g fill="#F90" fill-rule="evenodd" clip-rule="evenodd">
+
<path d="M14.465 11.813c-1.75 1.297-4.294 1.986-6.481 1.986-3.065 0-5.827-1.137-7.913-3.027-.165-.15-.016-.353.18-.235 2.257 1.313 5.04 2.109 7.92 2.109 1.941 0 4.075-.406 6.039-1.239.293-.133.543.192.255.406z"/>
+
<path d="M15.194 10.98c-.223-.287-1.479-.138-2.048-.069-.17.022-.197-.128-.043-.24 1-.705 2.645-.502 2.836-.267.192.24-.053 1.89-.99 2.68-.143.123-.281.06-.218-.1.213-.53.687-1.72.463-2.003z"/>
+
</g>
+
</svg>

api/core/tools/provider/builtin/aws/aws.py

@@ -0,0 +1,25 @@
+from core.tools.errors import ToolProviderCredentialValidationError
+from core.tools.provider.builtin.aws.tools.sagemaker_text_rerank import SageMakerReRankTool
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+
+
+class SageMakerProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        try:
+            SageMakerReRankTool().fork_tool_runtime(
+                runtime={
+                    "credentials": credentials,
+                }
+            ).invoke(
+                user_id='',
+                tool_parameters={
+                    "sagemaker_endpoint" : "",
+                    "query": "misaka mikoto",
+                    "candidate_texts" : "hello$$$hello world",
+                    "topk" : 5,
+                    "aws_region" : ""
+                },
+            )
+        except Exception as e:
+            raise ToolProviderCredentialValidationError(str(e))
+        

api/core/tools/provider/builtin/aws/aws.yaml

@@ -0,0 +1,15 @@
+identity:
+  author: AWS
+  name: aws
+  label:
+    en_US: AWS
+    zh_Hans: 亚马逊云科技
+    pt_BR: AWS
+  description:
+    en_US: Services on AWS.
+    zh_Hans: 亚马逊云科技的各类服务
+    pt_BR: Services on AWS.
+  icon: icon.svg
+  tags:
+    - search
+credentials_for_provider:

api/core/tools/provider/builtin/aws/tools/apply_guardrail.py

@@ -0,0 +1,83 @@
+import json
+import logging
+from typing import Any, Union
+
+import boto3
+from pydantic import BaseModel, Field
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+class GuardrailParameters(BaseModel):
+    guardrail_id: str = Field(..., description="The identifier of the guardrail")
+    guardrail_version: str = Field(..., description="The version of the guardrail")
+    source: str = Field(..., description="The source of the content")
+    text: str = Field(..., description="The text to apply the guardrail to")
+    aws_region: str = Field(default="us-east-1", description="AWS region for the Bedrock client")
+
+class ApplyGuardrailTool(BuiltinTool):
+    def _invoke(self,
+                user_id: str,
+                tool_parameters: dict[str, Any]
+                ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        """
+        Invoke the ApplyGuardrail tool
+        """
+        try:
+            # Validate and parse input parameters
+            params = GuardrailParameters(**tool_parameters)
+            
+            # Initialize AWS client
+            bedrock_client = boto3.client('bedrock-runtime', region_name=params.aws_region)
+
+            # Apply guardrail
+            response = bedrock_client.apply_guardrail(
+                guardrailIdentifier=params.guardrail_id,
+                guardrailVersion=params.guardrail_version,
+                source=params.source,
+                content=[{"text": {"text": params.text}}]
+            )
+
+            # Check for empty response
+            if not response:
+                return self.create_text_message(text="Received empty response from AWS Bedrock.")
+           
+            # Process the result
+            action = response.get("action", "No action specified")
+            outputs = response.get("outputs", [])
+            output = outputs[0].get("text", "No output received") if outputs else "No output received"
+            assessments = response.get("assessments", [])
+
+            # Format assessments
+            formatted_assessments = []
+            for assessment in assessments:
+                for policy_type, policy_data in assessment.items():
+                    if isinstance(policy_data, dict) and 'topics' in policy_data:
+                        for topic in policy_data['topics']:
+                            formatted_assessments.append(f"Policy: {policy_type}, Topic: {topic['name']}, Type: {topic['type']}, Action: {topic['action']}")
+                    else:
+                        formatted_assessments.append(f"Policy: {policy_type}, Data: {policy_data}")
+
+            result = f"Action: {action}\n "
+            result += f"Output: {output}\n "
+            if formatted_assessments:
+                result += "Assessments:\n " + "\n ".join(formatted_assessments) + "\n "
+#           result += f"Full response: {json.dumps(response, indent=2, ensure_ascii=False)}"
+
+            return self.create_text_message(text=result)
+
+        except boto3.exceptions.BotoCoreError as e:
+            error_message = f'AWS service error: {str(e)}'
+            logger.error(error_message, exc_info=True)
+            return self.create_text_message(text=error_message)
+        except json.JSONDecodeError as e:
+            error_message = f'JSON parsing error: {str(e)}'
+            logger.error(error_message, exc_info=True)
+            return self.create_text_message(text=error_message)
+        except Exception as e:
+            error_message = f'An unexpected error occurred: {str(e)}'
+            logger.error(error_message, exc_info=True)
+            return self.create_text_message(text=error_message)

api/core/tools/provider/builtin/aws/tools/apply_guardrail.yaml

@@ -0,0 +1,56 @@
+identity:
+  name: apply_guardrail
+  author: AWS
+  label:
+    en_US: Content Moderation Guardrails
+    zh_Hans: 内容审查护栏
+description:
+  human:
+    en_US: Content Moderation Guardrails utilizes the ApplyGuardrail API, a feature of Guardrails for Amazon Bedrock. This API is capable of evaluating input prompts and model responses for all Foundation Models (FMs), including those on Amazon Bedrock, custom FMs, and third-party FMs. By implementing this functionality, organizations can achieve centralized governance across all their generative AI applications, thereby enhancing control and consistency in content moderation.
+    zh_Hans: 内容审查护栏采用 Guardrails for Amazon Bedrock 功能中的 ApplyGuardrail API 。ApplyGuardrail 可以评估所有基础模型(FMs)的输入提示和模型响应，包括 Amazon Bedrock 上的 FMs、自定义 FMs 和第三方 FMs。通过实施这一功能, 组织可以在所有生成式 AI 应用程序中实现集中化的治理，从而增强内容审核的控制力和一致性。
+  llm: Content Moderation Guardrails utilizes the ApplyGuardrail API, a feature of Guardrails for Amazon Bedrock. This API is capable of evaluating input prompts and model responses for all Foundation Models (FMs), including those on Amazon Bedrock, custom FMs, and third-party FMs. By implementing this functionality, organizations can achieve centralized governance across all their generative AI applications, thereby enhancing control and consistency in content moderation.
+parameters:
+  - name: guardrail_id
+    type: string
+    required: true
+    label:
+      en_US: Guardrail ID
+      zh_Hans: Guardrail ID
+    human_description:
+      en_US: Please enter the ID of the Guardrail that has already been created on Amazon Bedrock, for example 'qk5nk0e4b77b'.
+      zh_Hans: 请输入已经在 Amazon Bedrock 上创建好的 Guardrail ID, 例如 'qk5nk0e4b77b'.
+    llm_description: Please enter the ID of the Guardrail that has already been created on Amazon Bedrock, for example 'qk5nk0e4b77b'.
+    form: form
+  - name: guardrail_version
+    type: string
+    required: true
+    label:
+      en_US: Guardrail Version Number
+      zh_Hans: Guardrail 版本号码
+    human_description:
+      en_US: Please enter the published version of the Guardrail ID that has already been created on Amazon Bedrock. This is typically a version number, such as 2.
+      zh_Hans: 请输入已经在Amazon Bedrock 上创建好的Guardrail ID发布的版本, 通常使用版本号, 例如2.
+    llm_description: Please enter the published version of the Guardrail ID that has already been created on Amazon Bedrock. This is typically a version number, such as 2.
+    form: form
+  - name: source
+    type: string
+    required: true
+    label:
+      en_US: Content Source (INPUT or OUTPUT)
+      zh_Hans: 内容来源 (INPUT or OUTPUT)
+    human_description:
+      en_US: The source of data used in the request to apply the guardrail. Valid Values "INPUT | OUTPUT"
+      zh_Hans: 用于应用护栏的请求中所使用的数据来源。有效值为 "INPUT | OUTPUT"
+    llm_description: The source of data used in the request to apply the guardrail. Valid Values "INPUT | OUTPUT"
+    form: form
+  - name: text
+    type: string
+    required: true
+    label:
+      en_US: Content to be reviewed
+      zh_Hans: 待审查内容
+    human_description:
+      en_US: The content used for requesting guardrail review, which can be either user input or LLM output.
+      zh_Hans: 用于请求护栏审查的内容，可以是用户输入或 LLM 输出。
+    llm_description: The content used for requesting guardrail review, which can be either user input or LLM output.
+    form: llm

api/core/tools/provider/builtin/aws/tools/lambda_translate_utils.py

@@ -0,0 +1,88 @@
+import json
+from typing import Any, Union
+
+import boto3
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class LambdaTranslateUtilsTool(BuiltinTool):
+    lambda_client: Any = None
+
+    def _invoke_lambda(self, text_content, src_lang, dest_lang, model_id, dictionary_name, request_type, lambda_name):
+        msg = { 
+            "src_content":text_content, 
+            "src_lang": src_lang, 
+            "dest_lang":dest_lang, 
+            "dictionary_id": dictionary_name,
+            "request_type" : request_type, 
+            "model_id" : model_id
+        }
+
+        invoke_response = self.lambda_client.invoke(FunctionName=lambda_name,
+                                               InvocationType='RequestResponse',
+                                               Payload=json.dumps(msg))
+        response_body = invoke_response['Payload']
+
+        response_str = response_body.read().decode("unicode_escape")
+
+        return response_str
+
+    def _invoke(self, 
+                user_id: str, 
+               tool_parameters: dict[str, Any], 
+        ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        """
+            invoke tools
+        """
+        line = 0
+        try:
+            if not self.lambda_client:
+                aws_region = tool_parameters.get('aws_region')
+                if aws_region:
+                    self.lambda_client = boto3.client("lambda", region_name=aws_region)
+                else:
+                    self.lambda_client = boto3.client("lambda")
+
+            line = 1
+            text_content = tool_parameters.get('text_content', '')
+            if not text_content:
+                return self.create_text_message('Please input text_content')
+            
+            line = 2
+            src_lang = tool_parameters.get('src_lang', '')
+            if not src_lang:
+                return self.create_text_message('Please input src_lang')
+            
+            line = 3
+            dest_lang = tool_parameters.get('dest_lang', '')
+            if not dest_lang:
+                return self.create_text_message('Please input dest_lang')
+            
+            line = 4
+            lambda_name = tool_parameters.get('lambda_name', '')
+            if not lambda_name:
+                return self.create_text_message('Please input lambda_name')
+            
+            line = 5
+            request_type = tool_parameters.get('request_type', '')
+            if not request_type:
+                return self.create_text_message('Please input request_type')
+            
+            line = 6
+            model_id = tool_parameters.get('model_id', '')
+            if not model_id:
+                return self.create_text_message('Please input model_id')
+
+            line = 7
+            dictionary_name = tool_parameters.get('dictionary_name', '')
+            if not dictionary_name:
+                return self.create_text_message('Please input dictionary_name')
+            
+            result = self._invoke_lambda(text_content, src_lang, dest_lang, model_id, dictionary_name, request_type, lambda_name)
+
+            return self.create_text_message(text=result)
+
+        except Exception as e:
+            return self.create_text_message(f'Exception {str(e)}, line : {line}')

api/core/tools/provider/builtin/aws/tools/lambda_translate_utils.yaml

@@ -0,0 +1,134 @@
+identity:
+  name: lambda_translate_utils
+  author: AWS
+  label:
+    en_US: TranslateTool
+    zh_Hans: 翻译工具
+    pt_BR: TranslateTool
+  icon: icon.svg
+description:
+  human:
+    en_US: A util tools for LLM translation, extra deployment is needed on AWS. Please refer Github Repo - https://github.com/ybalbert001/dynamodb-rag
+    zh_Hans: 大语言模型翻译工具(专词映射获取)，需要在AWS上进行额外部署，可参考Github Repo - https://github.com/ybalbert001/dynamodb-rag
+    pt_BR: A util tools for LLM translation, specfic Lambda Function deployment is needed on AWS. Please refer Github Repo - https://github.com/ybalbert001/dynamodb-rag
+  llm: A util tools for translation.
+parameters:
+  - name: text_content
+    type: string
+    required: true
+    label:
+      en_US: source content for translation
+      zh_Hans: 待翻译原文
+      pt_BR: source content for translation
+    human_description:
+      en_US: source content for translation
+      zh_Hans: 待翻译原文
+      pt_BR: source content for translation
+    llm_description: source content for translation
+    form: llm
+  - name: src_lang
+    type: string
+    required: true
+    label:
+      en_US: source language code
+      zh_Hans: 原文语言代号
+      pt_BR: source language code
+    human_description:
+      en_US: source language code
+      zh_Hans: 原文语言代号
+      pt_BR: source language code
+    llm_description: source language code
+    form: llm
+  - name: dest_lang
+    type: string
+    required: true
+    label:
+      en_US: target language code
+      zh_Hans: 目标语言代号
+      pt_BR: target language code
+    human_description:
+      en_US: target language code
+      zh_Hans: 目标语言代号
+      pt_BR: target language code
+    llm_description: target language code
+    form: llm
+  - name: aws_region
+    type: string
+    required: false
+    label:
+      en_US: region of Lambda
+      zh_Hans: Lambda 所在的region
+      pt_BR: region of Lambda
+    human_description:
+      en_US: region of Lambda
+      zh_Hans: Lambda 所在的region
+      pt_BR: region of Lambda
+    llm_description: region of Lambda
+    form: form
+  - name: model_id
+    type: string
+    required: false
+    default: anthropic.claude-3-sonnet-20240229-v1:0
+    label:
+      en_US: LLM model_id in bedrock
+      zh_Hans: bedrock上的大语言模型model_id
+      pt_BR: LLM model_id in bedrock
+    human_description:
+      en_US: LLM model_id in bedrock
+      zh_Hans: bedrock上的大语言模型model_id
+      pt_BR: LLM model_id in bedrock
+    llm_description: LLM model_id in bedrock
+    form: form
+  - name: dictionary_name
+    type: string
+    required: false
+    label:
+      en_US: dictionary name for term mapping
+      zh_Hans: 专词映射表名称
+      pt_BR: dictionary name for term mapping
+    human_description:
+      en_US: dictionary name for term mapping
+      zh_Hans: 专词映射表名称
+      pt_BR: dictionary name for term mapping
+    llm_description: dictionary name for term mapping
+    form: form
+  - name: request_type
+    type: select
+    required: false
+    label:
+      en_US: request type
+      zh_Hans: 请求类型
+      pt_BR: request type
+    human_description:
+      en_US: request type
+      zh_Hans: 请求类型
+      pt_BR: request type
+    default: term_mapping
+    options:
+      - value: term_mapping
+        label:
+          en_US: term_mapping
+          zh_Hans: 专词映射
+      - value: segment_only
+        label:
+          en_US: segment_only
+          zh_Hans: 仅切词
+      - value: translate
+        label:
+          en_US: translate
+          zh_Hans: 翻译内容
+    form: form
+  - name: lambda_name
+    type: string
+    default: "translate_tool"
+    required: true
+    label:
+      en_US: AWS Lambda for term mapping retrieval
+      zh_Hans: 专词召回映射 - AWS Lambda
+      pt_BR: lambda name for term mapping retrieval
+    human_description:
+      en_US: AWS Lambda  for term mapping retrieval
+      zh_Hans: 专词召回映射 - AWS Lambda
+      pt_BR: AWS Lambda  for term mapping retrieval
+    llm_description: AWS Lambda  for term mapping retrieval
+    form: form

api/core/tools/provider/builtin/aws/tools/sagemaker_text_rerank.py

@@ -0,0 +1,86 @@
+import json
+from typing import Any, Union
+
+import boto3
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class SageMakerReRankTool(BuiltinTool):
+    sagemaker_client: Any = None
+    sagemaker_endpoint:str = None
+    topk:int = None
+
+    def _sagemaker_rerank(self, query_input: str, docs: list[str], rerank_endpoint:str):
+        inputs = [query_input]*len(docs)
+        response_model = self.sagemaker_client.invoke_endpoint(
+            EndpointName=rerank_endpoint,
+            Body=json.dumps(
+                {
+                    "inputs": inputs,
+                    "docs": docs
+                }
+            ),
+            ContentType="application/json",
+        )
+        json_str = response_model['Body'].read().decode('utf8')
+        json_obj = json.loads(json_str)
+        scores = json_obj['scores']
+        return scores if isinstance(scores, list) else [scores]
+
+    def _invoke(self, 
+                user_id: str, 
+               tool_parameters: dict[str, Any], 
+        ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        """
+            invoke tools
+        """
+        line = 0
+        try:
+            if not self.sagemaker_client:
+                aws_region = tool_parameters.get('aws_region')
+                if aws_region:
+                    self.sagemaker_client = boto3.client("sagemaker-runtime", region_name=aws_region)
+                else:
+                    self.sagemaker_client = boto3.client("sagemaker-runtime")
+
+            line = 1
+            if not self.sagemaker_endpoint:
+                self.sagemaker_endpoint = tool_parameters.get('sagemaker_endpoint')
+
+            line = 2
+            if not self.topk:
+                self.topk = tool_parameters.get('topk', 5)
+
+            line = 3
+            query = tool_parameters.get('query', '')
+            if not query:
+                return self.create_text_message('Please input query')
+            
+            line = 4
+            candidate_texts = tool_parameters.get('candidate_texts')
+            if not candidate_texts:
+                return self.create_text_message('Please input candidate_texts')
+            
+            line = 5
+            candidate_docs = json.loads(candidate_texts)
+            docs = [ item.get('content') for item in candidate_docs ]
+
+            line = 6
+            scores = self._sagemaker_rerank(query_input=query, docs=docs, rerank_endpoint=self.sagemaker_endpoint)
+
+            line = 7
+            for idx in range(len(candidate_docs)):
+                candidate_docs[idx]["score"] = scores[idx]
+
+            line = 8
+            sorted_candidate_docs = sorted(candidate_docs, key=lambda x: x['score'], reverse=True)
+
+            line = 9
+            results_str = json.dumps(sorted_candidate_docs[:self.topk], ensure_ascii=False)
+            return self.create_text_message(text=results_str)
+            
+        except Exception as e:
+            return self.create_text_message(f'Exception {str(e)}, line : {line}')
+    

api/core/tools/provider/builtin/aws/tools/sagemaker_text_rerank.yaml

@@ -0,0 +1,82 @@
+identity:
+  name: sagemaker_text_rerank
+  author: AWS
+  label:
+    en_US: SagemakerRerank
+    zh_Hans: Sagemaker重排序
+    pt_BR: SagemakerRerank
+  icon: icon.svg
+description:
+  human:
+    en_US: A tool for performing text similarity ranking. You can find deploy notebook on Github Repo - https://github.com/aws-samples/dify-aws-tool
+    zh_Hans: Sagemaker重排序工具, 请参考 Github Repo - https://github.com/aws-samples/dify-aws-tool上的部署脚本
+    pt_BR: A tool for performing text similarity ranking.
+  llm: A tool for performing text similarity ranking. You can find deploy notebook on Github Repo - https://github.com/aws-samples/dify-aws-tool
+parameters:
+  - name: sagemaker_endpoint
+    type: string
+    required: true
+    label:
+      en_US: sagemaker endpoint for reranking
+      zh_Hans: 重排序的SageMaker 端点
+      pt_BR: sagemaker endpoint for reranking
+    human_description:
+      en_US: sagemaker endpoint for reranking
+      zh_Hans: 重排序的SageMaker 端点
+      pt_BR: sagemaker endpoint for reranking
+    llm_description: sagemaker endpoint for reranking
+    form: form
+  - name: query
+    type: string
+    required: true
+    label:
+      en_US: Query string
+      zh_Hans: 查询语句
+      pt_BR: Query string
+    human_description:
+      en_US: key words for searching
+      zh_Hans: 查询关键词
+      pt_BR: key words for searching
+    llm_description: key words for searching
+    form: llm
+  - name: candidate_texts
+    type: string
+    required: true
+    label:
+      en_US: text candidates
+      zh_Hans: 候选文本
+      pt_BR: text candidates
+    human_description:
+      en_US: searched candidates by query
+      zh_Hans: 查询文本搜到候选文本
+      pt_BR: searched candidates by query
+    llm_description: searched candidates by query
+    form: llm
+  - name: topk
+    type: number
+    required: false
+    form: form
+    label:
+      en_US: Limit for results count
+      zh_Hans: 返回个数限制
+      pt_BR: Limit for results count
+    human_description:
+      en_US: Limit for results count
+      zh_Hans: 返回个数限制
+      pt_BR: Limit for results count
+    min: 1
+    max: 10
+    default: 5
+  - name: aws_region
+    type: string
+    required: false
+    label:
+      en_US: region of sagemaker endpoint
+      zh_Hans: SageMaker 端点所在的region
+      pt_BR: region of sagemaker endpoint
+    human_description:
+      en_US: region of sagemaker endpoint
+      zh_Hans: SageMaker 端点所在的region
+      pt_BR: region of sagemaker endpoint
+    llm_description: region of sagemaker endpoint
+    form: form

api/core/tools/provider/builtin/spider/spiderApp.py

@@ -116,6 +116,7 @@ class Spider:
         :param params: Optional dictionary of additional parameters for the scrape request.
         :return: JSON response containing the scraping results.
         """
+        params = params or {}
 
         # Add { "return_format": "markdown" } to the params if not already present
         if "return_format" not in params:
@@ -143,6 +144,7 @@ class Spider:
         :param stream: Boolean indicating if the response should be streamed. Defaults to False.
         :return: JSON response or the raw response stream if streaming enabled.
         """
+        params = params or {}
 
         # Add { "return_format": "markdown" } to the params if not already present
         if "return_format" not in params:

api/core/tools/utils/web_reader_tool.py

@@ -10,6 +10,7 @@ import unicodedata
 from contextlib import contextmanager
 from urllib.parse import unquote
 
+import chardet
 import cloudscraper
 from bs4 import BeautifulSoup, CData, Comment, NavigableString
 from regex import regex
@@ -75,7 +76,18 @@ def get_url(url: str, user_agent: str = None) -> str:
     if response.status_code != 200:
         return "URL returned status code {}.".format(response.status_code)
 
-    a = extract_using_readabilipy(response.text)
+    # Detect encoding using chardet
+    detected_encoding = chardet.detect(response.content)
+    encoding = detected_encoding['encoding']
+    if encoding:
+        try:
+            content = response.content.decode(encoding)
+        except (UnicodeDecodeError, TypeError):
+            content = response.text
+    else:
+        content = response.text
+
+    a = extract_using_readabilipy(content)
 
     if not a['plain_text'] or not a['plain_text'].strip():
         return ''

api/docker/entrypoint.sh

@@ -8,8 +8,21 @@ if [[ "${MIGRATION_ENABLED}" == "true" ]]; then
 fi
 
 if [[ "${MODE}" == "worker" ]]; then
-  exec celery -A app.celery worker -P ${CELERY_WORKER_CLASS:-gevent} -c ${CELERY_WORKER_AMOUNT:-1} --loglevel INFO \
+
+  # Get the number of available CPU cores
+  if [ "${CELERY_AUTO_SCALE,,}" = "true" ]; then
+    # Set MAX_WORKERS to the number of available cores if not specified
+    AVAILABLE_CORES=$(nproc)
+    MAX_WORKERS=${CELERY_MAX_WORKERS:-$AVAILABLE_CORES}
+    MIN_WORKERS=${CELERY_MIN_WORKERS:-1}
+    CONCURRENCY_OPTION="--autoscale=${MAX_WORKERS},${MIN_WORKERS}"
+  else
+    CONCURRENCY_OPTION="-c ${CELERY_WORKER_AMOUNT:-1}"
+  fi
+
+  exec celery -A app.celery worker -P ${CELERY_WORKER_CLASS:-gevent} $CONCURRENCY_OPTION --loglevel INFO \
     -Q ${CELERY_QUEUES:-dataset,generation,mail,ops_trace,app_deletion}
+
 elif [[ "${MODE}" == "beat" ]]; then
   exec celery -A app.celery beat --loglevel INFO
 else

api/libs/passport.py

@@ -1,15 +1,16 @@
 import jwt
-from flask import current_app
 from werkzeug.exceptions import Unauthorized
 
+from configs import dify_config
+
 
 class PassportService:
     def __init__(self):
-        self.sk = current_app.config.get('SECRET_KEY')
-    
+        self.sk = dify_config.SECRET_KEY
+
     def issue(self, payload):
         return jwt.encode(payload, self.sk, algorithm='HS256')
-    
+
     def verify(self, token):
         try:
             return jwt.decode(token, self.sk, algorithms=['HS256'])

api/poetry.lock

@@ -448,63 +448,6 @@ doc = ["Sphinx (>=7)", "packaging", "sphinx-autodoc-typehints (>=1.2.0)", "sphin
 test = ["anyio[trio]", "coverage[toml] (>=7)", "exceptiongroup (>=1.2.0)", "hypothesis (>=4.0)", "psutil (>=5.9)", "pytest (>=7.0)", "pytest-mock (>=3.6.1)", "trustme", "uvloop (>=0.17)"]
 trio = ["trio (>=0.23)"]
 
-[[package]]
-name = "argon2-cffi"
-version = "23.1.0"
-description = "Argon2 for Python"
-optional = false
-python-versions = ">=3.7"
-files = [
-    {file = "argon2_cffi-23.1.0-py3-none-any.whl", hash = "sha256:c670642b78ba29641818ab2e68bd4e6a78ba53b7eff7b4c3815ae16abf91c7ea"},
-    {file = "argon2_cffi-23.1.0.tar.gz", hash = "sha256:879c3e79a2729ce768ebb7d36d4609e3a78a4ca2ec3a9f12286ca057e3d0db08"},
-]
-
-[package.dependencies]
-argon2-cffi-bindings = "*"
-
-[package.extras]
-dev = ["argon2-cffi[tests,typing]", "tox (>4)"]
-docs = ["furo", "myst-parser", "sphinx", "sphinx-copybutton", "sphinx-notfound-page"]
-tests = ["hypothesis", "pytest"]
-typing = ["mypy"]
-
-[[package]]
-name = "argon2-cffi-bindings"
-version = "21.2.0"
-description = "Low-level CFFI bindings for Argon2"
-optional = false
-python-versions = ">=3.6"
-files = [
-    {file = "argon2-cffi-bindings-21.2.0.tar.gz", hash = "sha256:bb89ceffa6c791807d1305ceb77dbfacc5aa499891d2c55661c6459651fc39e3"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:ccb949252cb2ab3a08c02024acb77cfb179492d5701c7cbdbfd776124d4d2367"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9524464572e12979364b7d600abf96181d3541da11e23ddf565a32e70bd4dc0d"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b746dba803a79238e925d9046a63aa26bf86ab2a2fe74ce6b009a1c3f5c8f2ae"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:58ed19212051f49a523abb1dbe954337dc82d947fb6e5a0da60f7c8471a8476c"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-musllinux_1_1_aarch64.whl", hash = "sha256:bd46088725ef7f58b5a1ef7ca06647ebaf0eb4baff7d1d0d177c6cc8744abd86"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-musllinux_1_1_i686.whl", hash = "sha256:8cd69c07dd875537a824deec19f978e0f2078fdda07fd5c42ac29668dda5f40f"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-musllinux_1_1_x86_64.whl", hash = "sha256:f1152ac548bd5b8bcecfb0b0371f082037e47128653df2e8ba6e914d384f3c3e"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-win32.whl", hash = "sha256:603ca0aba86b1349b147cab91ae970c63118a0f30444d4bc80355937c950c082"},
-    {file = "argon2_cffi_bindings-21.2.0-cp36-abi3-win_amd64.whl", hash = "sha256:b2ef1c30440dbbcba7a5dc3e319408b59676e2e039e2ae11a8775ecf482b192f"},
-    {file = "argon2_cffi_bindings-21.2.0-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:e415e3f62c8d124ee16018e491a009937f8cf7ebf5eb430ffc5de21b900dad93"},
-    {file = "argon2_cffi_bindings-21.2.0-pp37-pypy37_pp73-macosx_10_9_x86_64.whl", hash = "sha256:3e385d1c39c520c08b53d63300c3ecc28622f076f4c2b0e6d7e796e9f6502194"},
-    {file = "argon2_cffi_bindings-21.2.0-pp37-pypy37_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2c3e3cc67fdb7d82c4718f19b4e7a87123caf8a93fde7e23cf66ac0337d3cb3f"},
-    {file = "argon2_cffi_bindings-21.2.0-pp37-pypy37_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6a22ad9800121b71099d0fb0a65323810a15f2e292f2ba450810a7316e128ee5"},
-    {file = "argon2_cffi_bindings-21.2.0-pp37-pypy37_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f9f8b450ed0547e3d473fdc8612083fd08dd2120d6ac8f73828df9b7d45bb351"},
-    {file = "argon2_cffi_bindings-21.2.0-pp37-pypy37_pp73-win_amd64.whl", hash = "sha256:93f9bf70084f97245ba10ee36575f0c3f1e7d7724d67d8e5b08e61787c320ed7"},
-    {file = "argon2_cffi_bindings-21.2.0-pp38-pypy38_pp73-macosx_10_9_x86_64.whl", hash = "sha256:3b9ef65804859d335dc6b31582cad2c5166f0c3e7975f324d9ffaa34ee7e6583"},
-    {file = "argon2_cffi_bindings-21.2.0-pp38-pypy38_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d4966ef5848d820776f5f562a7d45fdd70c2f330c961d0d745b784034bd9f48d"},
-    {file = "argon2_cffi_bindings-21.2.0-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:20ef543a89dee4db46a1a6e206cd015360e5a75822f76df533845c3cbaf72670"},
-    {file = "argon2_cffi_bindings-21.2.0-pp38-pypy38_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ed2937d286e2ad0cc79a7087d3c272832865f779430e0cc2b4f3718d3159b0cb"},
-    {file = "argon2_cffi_bindings-21.2.0-pp38-pypy38_pp73-win_amd64.whl", hash = "sha256:5e00316dabdaea0b2dd82d141cc66889ced0cdcbfa599e8b471cf22c620c329a"},
-]
-
-[package.dependencies]
-cffi = ">=1.0.1"
-
-[package.extras]
-dev = ["cogapp", "pre-commit", "pytest", "wheel"]
-tests = ["pytest"]
-
 [[package]]
 name = "arxiv"
 version = "2.1.0"
@@ -4616,22 +4559,20 @@ files = [
 ]
 
 [[package]]
-name = "minio"
-version = "7.2.7"
-description = "MinIO Python SDK for Amazon S3 Compatible Cloud Storage"
+name = "milvus-lite"
+version = "2.4.8"
+description = "A lightweight version of Milvus wrapped with Python."
 optional = false
-python-versions = "*"
+python-versions = ">=3.7"
 files = [
-    {file = "minio-7.2.7-py3-none-any.whl", hash = "sha256:59d1f255d852fe7104018db75b3bebbd987e538690e680f7c5de835e422de837"},
-    {file = "minio-7.2.7.tar.gz", hash = "sha256:473d5d53d79f340f3cd632054d0c82d2f93177ce1af2eac34a235bea55708d98"},
+    {file = "milvus_lite-2.4.8-py3-none-macosx_10_9_x86_64.whl", hash = "sha256:b7e90b34b214884cd44cdc112ab243d4cb197b775498355e2437b6cafea025fe"},
+    {file = "milvus_lite-2.4.8-py3-none-macosx_11_0_arm64.whl", hash = "sha256:519dfc62709d8f642d98a1c5b1dcde7080d107e6e312d677fef5a3412a40ac08"},
+    {file = "milvus_lite-2.4.8-py3-none-manylinux2014_aarch64.whl", hash = "sha256:b21f36d24cbb0e920b4faad607019bb28c1b2c88b4d04680ac8c7697a4ae8a4d"},
+    {file = "milvus_lite-2.4.8-py3-none-manylinux2014_x86_64.whl", hash = "sha256:08332a2b9abfe7c4e1d7926068937e46f8fb81f2707928b7bc02c9dc99cebe41"},
 ]
 
 [package.dependencies]
-argon2-cffi = "*"
-certifi = "*"
-pycryptodome = "*"
-typing-extensions = "*"
-urllib3 = "*"
+tqdm = "*"
 
 [[package]]
 name = "mmh3"
@@ -6078,6 +6019,19 @@ files = [
     {file = "pyarrow-17.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:392bc9feabc647338e6c89267635e111d71edad5fcffba204425a7c8d13610d7"},
     {file = "pyarrow-17.0.0-cp38-cp38-macosx_10_15_x86_64.whl", hash = "sha256:af5ff82a04b2171415f1410cff7ebb79861afc5dae50be73ce06d6e870615204"},
     {file = "pyarrow-17.0.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:edca18eaca89cd6382dfbcff3dd2d87633433043650c07375d095cd3517561d8"},
+    {file = "pyarrow-17.0.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7c7916bff914ac5d4a8fe25b7a25e432ff921e72f6f2b7547d1e325c1ad9d155"},
+    {file = "pyarrow-17.0.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f553ca691b9e94b202ff741bdd40f6ccb70cdd5fbf65c187af132f1317de6145"},
+    {file = "pyarrow-17.0.0-cp38-cp38-manylinux_2_28_aarch64.whl", hash = "sha256:0cdb0e627c86c373205a2f94a510ac4376fdc523f8bb36beab2e7f204416163c"},
+    {file = "pyarrow-17.0.0-cp38-cp38-manylinux_2_28_x86_64.whl", hash = "sha256:d7d192305d9d8bc9082d10f361fc70a73590a4c65cf31c3e6926cd72b76bc35c"},
+    {file = "pyarrow-17.0.0-cp38-cp38-win_amd64.whl", hash = "sha256:02dae06ce212d8b3244dd3e7d12d9c4d3046945a5933d28026598e9dbbda1fca"},
+    {file = "pyarrow-17.0.0-cp39-cp39-macosx_10_15_x86_64.whl", hash = "sha256:13d7a460b412f31e4c0efa1148e1d29bdf18ad1411eb6757d38f8fbdcc8645fb"},
+    {file = "pyarrow-17.0.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:9b564a51fbccfab5a04a80453e5ac6c9954a9c5ef2890d1bcf63741909c3f8df"},
+    {file = "pyarrow-17.0.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:32503827abbc5aadedfa235f5ece8c4f8f8b0a3cf01066bc8d29de7539532687"},
+    {file = "pyarrow-17.0.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a155acc7f154b9ffcc85497509bcd0d43efb80d6f733b0dc3bb14e281f131c8b"},
+    {file = "pyarrow-17.0.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:dec8d129254d0188a49f8a1fc99e0560dc1b85f60af729f47de4046015f9b0a5"},
+    {file = "pyarrow-17.0.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:a48ddf5c3c6a6c505904545c25a4ae13646ae1f8ba703c4df4a1bfe4f4006bda"},
+    {file = "pyarrow-17.0.0-cp39-cp39-win_amd64.whl", hash = "sha256:42bf93249a083aca230ba7e2786c5f673507fa97bbd9725a1e2754715151a204"},
+    {file = "pyarrow-17.0.0.tar.gz", hash = "sha256:4beca9521ed2c0921c1023e68d097d0299b62c362639ea315572a58f3f50fd28"},
 ]
 
 [package.dependencies]
@@ -6374,24 +6328,29 @@ tests = ["coverage[toml] (==5.0.4)", "pytest (>=6.0.0,<7.0.0)"]
 
 [[package]]
 name = "pymilvus"
-version = "2.3.1"
+version = "2.4.4"
 description = "Python Sdk for Milvus"
 optional = false
-python-versions = ">=3.7"
+python-versions = ">=3.8"
 files = [
-    {file = "pymilvus-2.3.1-py3-none-any.whl", hash = "sha256:ce65e1de8700f33bd9aade20f013291629702e25b05726773208f1f0b22548ff"},
-    {file = "pymilvus-2.3.1.tar.gz", hash = "sha256:d460f6204d7deb2cff93716bd65670c1b440694b77701fb0ab0ead791aa582c6"},
+    {file = "pymilvus-2.4.4-py3-none-any.whl", hash = "sha256:073b76bc36f6f4e70f0f0a0023a53324f0ba8ef9a60883f87cd30a44b6c6f2b5"},
+    {file = "pymilvus-2.4.4.tar.gz", hash = "sha256:50c53eb103e034fbffe936fe942751ea3dbd2452e18cf79acc52360ed4987fb7"},
 ]
 
 [package.dependencies]
 environs = "<=9.5.0"
-grpcio = ">=1.49.1,<=1.58.0"
-minio = "*"
+grpcio = ">=1.49.1,<=1.63.0"
+milvus-lite = {version = ">=2.4.0,<2.5.0", markers = "sys_platform != \"win32\""}
 pandas = ">=1.2.4"
 protobuf = ">=3.20.0"
-requests = "*"
+setuptools = ">=67"
 ujson = ">=2.0.0"
 
+[package.extras]
+bulk-writer = ["azure-storage-blob", "minio (>=7.0.0)", "pyarrow (>=12.0.0)", "requests"]
+dev = ["black", "grpcio (==1.62.2)", "grpcio-testing (==1.62.2)", "grpcio-tools (==1.62.2)", "pytest (>=5.3.4)", "pytest-cov (>=2.8.1)", "pytest-timeout (>=1.3.4)", "ruff (>0.4.0)"]
+model = ["milvus-model (>=0.1.0)"]
+
 [[package]]
 name = "pymysql"
 version = "1.1.1"
@@ -9543,4 +9502,4 @@ cffi = ["cffi (>=1.11)"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.10"
-content-hash = "9619ddabdd67710981c13dcfa3ddae0a48497c9f694afc81b820e882440c1265"
+content-hash = "a8b61d74d9322302b7447b6f8728ad606abc160202a8a122a05a8ef3cec7055b"

api/pyproject.toml

@@ -206,7 +206,7 @@ chromadb = "0.5.1"
 oracledb = "~2.2.1"
 pgvecto-rs = "0.1.4"
 pgvector = "0.2.5"
-pymilvus = "2.3.1"
+pymilvus = "~2.4.4"
 pymysql = "1.1.1"
 tcvectordb = "1.3.2"
 tidb-vector = "0.0.9"
@@ -216,18 +216,6 @@ alibabacloud_gpdb20160503 = "~3.8.0"
 alibabacloud_tea_openapi = "~0.3.9"
 clickhouse-connect = "~0.7.16"
 
-############################################################
-# Transparent dependencies required by main dependencies
-# for pinning versions
-############################################################
-
-[tool.poetry.group.transparent.dependencies]
-kaleido = "0.2.1"
-lxml = "5.1.0"
-sympy = "1.12"
-tenacity = "~8.3.0"
-xlrd = "~2.0.1"
-
 ############################################################
 # Dev dependencies for running tests
 ############################################################

api/tests/integration_tests/model_runtime/hunyuan/test_text_embedding.py

@@ -0,0 +1,104 @@
+import os
+
+import pytest
+
+from core.model_runtime.entities.text_embedding_entities import TextEmbeddingResult
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.hunyuan.text_embedding.text_embedding import HunyuanTextEmbeddingModel
+
+
+def test_validate_credentials():
+    model = HunyuanTextEmbeddingModel()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        model.validate_credentials(
+            model='hunyuan-embedding',
+            credentials={
+                'secret_id': 'invalid_key',
+                'secret_key': 'invalid_key'
+            }
+        )
+
+    model.validate_credentials(
+        model='hunyuan-embedding',
+        credentials={
+            'secret_id': os.environ.get('HUNYUAN_SECRET_ID'),
+            'secret_key': os.environ.get('HUNYUAN_SECRET_KEY')
+        }
+    )
+
+
+def test_invoke_model():
+    model = HunyuanTextEmbeddingModel()
+
+    result = model.invoke(
+        model='hunyuan-embedding',
+        credentials={
+            'secret_id': os.environ.get('HUNYUAN_SECRET_ID'),
+            'secret_key': os.environ.get('HUNYUAN_SECRET_KEY')
+        },
+        texts=[
+            "hello",
+            "world"
+        ],
+        user="abc-123"
+    )
+
+    assert isinstance(result, TextEmbeddingResult)
+    assert len(result.embeddings) == 2
+    assert result.usage.total_tokens == 6
+
+def test_get_num_tokens():
+    model = HunyuanTextEmbeddingModel()
+
+    num_tokens = model.get_num_tokens(
+        model='hunyuan-embedding',
+        credentials={
+            'secret_id': os.environ.get('HUNYUAN_SECRET_ID'),
+            'secret_key': os.environ.get('HUNYUAN_SECRET_KEY')
+        },
+        texts=[
+            "hello",
+            "world"
+        ]
+    )
+
+    assert num_tokens == 2
+
+def test_max_chunks():
+    model = HunyuanTextEmbeddingModel()
+
+    result = model.invoke(
+        model='hunyuan-embedding',
+        credentials={
+            'secret_id': os.environ.get('HUNYUAN_SECRET_ID'),
+            'secret_key': os.environ.get('HUNYUAN_SECRET_KEY')
+        },
+        texts=[
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+            "hello",
+            "world",
+        ]
+    )
+
+    assert isinstance(result, TextEmbeddingResult)
+    assert len(result.embeddings) == 22

api/tests/integration_tests/model_runtime/siliconflow/test_llm.py

@@ -0,0 +1,106 @@
+import os
+from collections.abc import Generator
+
+import pytest
+
+from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta
+from core.model_runtime.entities.message_entities import AssistantPromptMessage, SystemPromptMessage, UserPromptMessage
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.siliconflow.llm.llm import SiliconflowLargeLanguageModel
+
+
+def test_validate_credentials():
+    model = SiliconflowLargeLanguageModel()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        model.validate_credentials(
+            model='deepseek-ai/DeepSeek-V2-Chat',
+            credentials={
+                'api_key': 'invalid_key'
+            }
+        )
+
+    model.validate_credentials(
+        model='deepseek-ai/DeepSeek-V2-Chat',
+        credentials={
+            'api_key': os.environ.get('API_KEY')
+        }
+    )
+
+
+def test_invoke_model():
+    model = SiliconflowLargeLanguageModel()
+
+    response = model.invoke(
+        model='deepseek-ai/DeepSeek-V2-Chat',
+        credentials={
+            'api_key': os.environ.get('API_KEY')
+        },
+        prompt_messages=[
+            UserPromptMessage(
+                content='Who are you?'
+            )
+        ],
+        model_parameters={
+            'temperature': 0.5,
+            'max_tokens': 10
+        },
+        stop=['How'],
+        stream=False,
+        user="abc-123"
+    )
+
+    assert isinstance(response, LLMResult)
+    assert len(response.message.content) > 0
+
+
+def test_invoke_stream_model():
+    model = SiliconflowLargeLanguageModel()
+
+    response = model.invoke(
+        model='deepseek-ai/DeepSeek-V2-Chat',
+        credentials={
+            'api_key': os.environ.get('API_KEY')
+        },
+        prompt_messages=[
+            UserPromptMessage(
+                content='Hello World!'
+            )
+        ],
+        model_parameters={
+            'temperature': 0.5,
+            'max_tokens': 100,
+            'seed': 1234
+        },
+        stream=True,
+        user="abc-123"
+    )
+
+    assert isinstance(response, Generator)
+
+    for chunk in response:
+        assert isinstance(chunk, LLMResultChunk)
+        assert isinstance(chunk.delta, LLMResultChunkDelta)
+        assert isinstance(chunk.delta.message, AssistantPromptMessage)
+        assert len(chunk.delta.message.content) > 0 if chunk.delta.finish_reason is None else True
+
+
+def test_get_num_tokens():
+    model = SiliconflowLargeLanguageModel()
+
+    num_tokens = model.get_num_tokens(
+        model='deepseek-ai/DeepSeek-V2-Chat',
+        credentials={
+            'api_key': os.environ.get('API_KEY')
+        },
+        prompt_messages=[
+            SystemPromptMessage(
+                content='You are a helpful AI assistant.',
+            ),
+            UserPromptMessage(
+                content='Hello World!'
+            )
+        ]
+    )
+
+    assert num_tokens == 12

api/tests/integration_tests/model_runtime/siliconflow/test_provider.py

@@ -0,0 +1,21 @@
+import os
+
+import pytest
+
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.siliconflow.siliconflow import SiliconflowProvider
+
+
+def test_validate_provider_credentials():
+    provider = SiliconflowProvider()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        provider.validate_provider_credentials(
+            credentials={}
+        )
+
+    provider.validate_provider_credentials(
+        credentials={
+            'api_key': os.environ.get('API_KEY')
+        }
+    )

docker-legacy/docker-compose.milvus.yaml

@@ -38,7 +38,7 @@ services:
 
   milvus-standalone:
     container_name: milvus-standalone
-    image: milvusdb/milvus:v2.3.1
+    image: milvusdb/milvus:v2.4.6
     command: ["milvus", "run", "standalone"]
     environment:
       ETCD_ENDPOINTS: etcd:2379

docker/.env.example

@@ -124,10 +124,36 @@ GUNICORN_TIMEOUT=360
 # The number of Celery workers. The default is 1, and can be set as needed.
 CELERY_WORKER_AMOUNT=
 
+# Flag indicating whether to enable autoscaling of Celery workers.
+#
+# Autoscaling is useful when tasks are CPU intensive and can be dynamically
+# allocated and deallocated based on the workload.
+#
+# When autoscaling is enabled, the maximum and minimum number of workers can
+# be specified. The autoscaling algorithm will dynamically adjust the number
+# of workers within the specified range.
+#
+# Default is false (i.e., autoscaling is disabled).
+#
+# Example:
+# CELERY_AUTO_SCALE=true
+CELERY_AUTO_SCALE=false
+
+# The maximum number of Celery workers that can be autoscaled.
+# This is optional and only used when autoscaling is enabled.
+# Default is not set.
+CELERY_MAX_WORKERS=
+
+# The minimum number of Celery workers that can be autoscaled.
+# This is optional and only used when autoscaling is enabled.
+# Default is not set.
+CELERY_MIN_WORKERS=
+
 # API Tool configuration
 API_TOOL_DEFAULT_CONNECT_TIMEOUT=10
 API_TOOL_DEFAULT_READ_TIMEOUT=60
 
+
 # ------------------------------
 # Database Configuration
 # The database uses PostgreSQL. Please use the public schema.
@@ -601,6 +627,22 @@ NGINX_KEEPALIVE_TIMEOUT=65
 NGINX_PROXY_READ_TIMEOUT=3600s
 NGINX_PROXY_SEND_TIMEOUT=3600s
 
+NGINX_ENABLE_CERTBOT_CHALLENGE=false
+
+# ------------------------------
+# Certbot Configuration
+# ------------------------------
+
+# Email address (required to get certificates from Let's Encrypt)
+CERTBOT_EMAIL=your_email@example.com
+
+# Domain name
+CERTBOT_DOMAIN=your_domain.com
+
+# certbot command options
+# i.e: --force-renewal --dry-run --test-cert --debug
+CERTBOT_OPTIONS=
+
 # ------------------------------
 # Environment Variables for SSRF Proxy
 # ------------------------------
@@ -611,7 +653,7 @@ SSRF_SANDBOX_HOST=sandbox
 
 # ------------------------------
 # docker env var for specifying vector db type at startup
-# (based on the vector db type, the corresponding docker 
+# (based on the vector db type, the corresponding docker
 # compose profile will be used)
 # ------------------------------
 COMPOSE_PROFILES=${VECTOR_STORE:-weaviate}

docker/README.md

@@ -3,86 +3,105 @@
 Welcome to the new `docker` directory for deploying Dify using Docker Compose. This README outlines the updates, deployment instructions, and migration details for existing users.
 
 ### What's Updated
-- **Persistent Environment Variables**: Environment variables are now managed through a `.env` file, ensuring that your configurations persist across deployments.
 
-   > What is `.env`? </br> </br>
-   > The `.env` file is a crucial component in Docker and Docker Compose environments, serving as a centralized configuration file where you can define environment variables that are accessible to the containers at runtime. This file simplifies the management of environment settings across different stages of development, testing, and production, providing consistency and ease of configuration to deployments.
+- **Certbot Container**: `docker-compose.yaml` now contains `certbot` for managing SSL certificates. This container automatically renews certificates and ensures secure HTTPS connections.  
+  For more information, refer `docker/certbot/README.md`.
 
-- **Unified Vector Database Services**: All vector database services are now managed from a single Docker Compose file `docker-compose.yaml`. You can switch between different vector databases by setting the `VECTOR_STORE` environment variable in your `.env` file.
-- **Mandatory .env File**: A `.env` file is now required to run `docker compose up`. This file is crucial for configuring your deployment and for any custom settings to persist through upgrades.
-- **Legacy Support**: Previous deployment files are now located in the `docker-legacy` directory and will no longer be maintained.
+- **Persistent Environment Variables
+  **: Environment variables are now managed through a `.env` file, ensuring that your configurations persist across deployments.
+
+  > What is `.env`? </br> </br>
+  > The `.env` file is a crucial component in Docker and Docker Compose environments, serving as a centralized configuration file where you can define environment variables that are accessible to the containers at runtime. This file simplifies the management of environment settings across different stages of development, testing, and production, providing consistency and ease of configuration to deployments.
+
+- **Unified Vector Database Services
+  **: All vector database services are now managed from a single Docker Compose file `docker-compose.yaml`. You can switch between different vector databases by setting the `VECTOR_STORE` environment variable in your `.env` file.
+- **Mandatory .env File
+  **: A `.env` file is now required to run `docker compose up`. This file is crucial for configuring your deployment and for any custom settings to persist through upgrades.
+- **Legacy Support
+  **: Previous deployment files are now located in the `docker-legacy` directory and will no longer be maintained.
 
 ### How to Deploy Dify with `docker-compose.yaml`
+
 1. **Prerequisites**: Ensure Docker and Docker Compose are installed on your system.
 2. **Environment Setup**:
-   - Navigate to the `docker` directory.
-   - Copy the `.env.example` file to a new file named `.env` by running `cp .env.example .env`.
-   - Customize the `.env` file as needed. Refer to the `.env.example` file for detailed configuration options.
+    - Navigate to the `docker` directory.
+    - Copy the `.env.example` file to a new file named `.env` by running `cp .env.example .env`.
+    - Customize the `.env` file as needed. Refer to the `.env.example` file for detailed configuration options.
 3. **Running the Services**:
-   - Execute `docker compose up` from the `docker` directory to start the services.
-   - To specify a vector database, set the `VECTOR_store` variable in your `.env` file to your desired vector database service, such as `milvus`, `weaviate`, or `opensearch`.
+    - Execute `docker compose up` from the `docker` directory to start the services.
+    - To specify a vector database, set the `VECTOR_store` variable in your `.env` file to your desired vector database service, such as `milvus`, `weaviate`, or `opensearch`.
+4. **SSL Certificate Setup**:
+    - Rrefer `docker/certbot/README.md` to set up SSL certificates using Certbot.
 
 ### How to Deploy Middleware for Developing Dify
+
 1. **Middleware Setup**:
-   - Use the `docker-compose.middleware.yaml` for setting up essential middleware services like databases and caches.
-   - Navigate to the `docker` directory.
-   - Ensure the `middleware.env` file is created by running `cp middleware.env.example middleware.env` (refer to the `middleware.env.example` file).
+    - Use the `docker-compose.middleware.yaml` for setting up essential middleware services like databases and caches.
+    - Navigate to the `docker` directory.
+    - Ensure the `middleware.env` file is created by running `cp middleware.env.example middleware.env` (refer to the `middleware.env.example` file).
 2. **Running Middleware Services**:
-   - Execute `docker-compose -f docker-compose.middleware.yaml up -d` to start the middleware services.
+    - Execute `docker-compose -f docker-compose.middleware.yaml up -d` to start the middleware services.
 
 ### Migration for Existing Users
+
 For users migrating from the `docker-legacy` setup:
+
 1. **Review Changes**: Familiarize yourself with the new `.env` configuration and Docker Compose setup.
 2. **Transfer Customizations**:
-   - If you have customized configurations such as `docker-compose.yaml`, `ssrf_proxy/squid.conf`, or `nginx/conf.d/default.conf`, you will need to reflect these changes in the `.env` file you create.
+    - If you have customized configurations such as `docker-compose.yaml`, `ssrf_proxy/squid.conf`, or `nginx/conf.d/default.conf`, you will need to reflect these changes in the `.env` file you create.
 3. **Data Migration**:
-   - Ensure that data from services like databases and caches is backed up and migrated appropriately to the new structure if necessary.
+    - Ensure that data from services like databases and caches is backed up and migrated appropriately to the new structure if necessary.
 
-### Overview of `.env` 
+### Overview of `.env`
 
 #### Key Modules and Customization
 
-- **Vector Database Services**: Depending on the type of vector database used (`VECTOR_STORE`), users can set specific endpoints, ports, and authentication details.
-- **Storage Services**: Depending on the storage type (`STORAGE_TYPE`), users can configure specific settings for S3, Azure Blob, Google Storage, etc.
+- **Vector Database Services
+  **: Depending on the type of vector database used (`VECTOR_STORE`), users can set specific endpoints, ports, and authentication details.
+- **Storage Services
+  **: Depending on the storage type (`STORAGE_TYPE`), users can configure specific settings for S3, Azure Blob, Google Storage, etc.
 - **API and Web Services**: Users can define URLs and other settings that affect how the API and web frontends operate.
 
 #### Other notable variables
+
 The `.env.example` file provided in the Docker setup is extensive and covers a wide range of configuration options. It is structured into several sections, each pertaining to different aspects of the application and its services. Here are some of the key sections and variables:
 
 1. **Common Variables**:
-   - `CONSOLE_API_URL`, `SERVICE_API_URL`: URLs for different API services.
-   - `APP_WEB_URL`: Frontend application URL.
-   - `FILES_URL`: Base URL for file downloads and previews.
+    - `CONSOLE_API_URL`, `SERVICE_API_URL`: URLs for different API services.
+    - `APP_WEB_URL`: Frontend application URL.
+    - `FILES_URL`: Base URL for file downloads and previews.
 
 2. **Server Configuration**:
-   - `LOG_LEVEL`, `DEBUG`, `FLASK_DEBUG`: Logging and debug settings.
-   - `SECRET_KEY`: A key for encrypting session cookies and other sensitive data.
+    - `LOG_LEVEL`, `DEBUG`, `FLASK_DEBUG`: Logging and debug settings.
+    - `SECRET_KEY`: A key for encrypting session cookies and other sensitive data.
 
 3. **Database Configuration**:
-   - `DB_USERNAME`, `DB_PASSWORD`, `DB_HOST`, `DB_PORT`, `DB_DATABASE`: PostgreSQL database credentials and connection details.
+    - `DB_USERNAME`, `DB_PASSWORD`, `DB_HOST`, `DB_PORT`, `DB_DATABASE`: PostgreSQL database credentials and connection details.
 
 4. **Redis Configuration**:
-   - `REDIS_HOST`, `REDIS_PORT`, `REDIS_PASSWORD`: Redis server connection settings.
+    - `REDIS_HOST`, `REDIS_PORT`, `REDIS_PASSWORD`: Redis server connection settings.
 
 5. **Celery Configuration**:
-   - `CELERY_BROKER_URL`: Configuration for Celery message broker.
+    - `CELERY_BROKER_URL`: Configuration for Celery message broker.
 
 6. **Storage Configuration**:
-   - `STORAGE_TYPE`, `S3_BUCKET_NAME`, `AZURE_BLOB_ACCOUNT_NAME`: Settings for file storage options like local, S3, Azure Blob, etc.
+    - `STORAGE_TYPE`, `S3_BUCKET_NAME`, `AZURE_BLOB_ACCOUNT_NAME`: Settings for file storage options like local, S3, Azure Blob, etc.
 
 7. **Vector Database Configuration**:
-   - `VECTOR_STORE`: Type of vector database (e.g., `weaviate`, `milvus`).
-   - Specific settings for each vector store like `WEAVIATE_ENDPOINT`, `MILVUS_HOST`.
+    - `VECTOR_STORE`: Type of vector database (e.g., `weaviate`, `milvus`).
+    - Specific settings for each vector store like `WEAVIATE_ENDPOINT`, `MILVUS_HOST`.
 
 8. **CORS Configuration**:
-   - `WEB_API_CORS_ALLOW_ORIGINS`, `CONSOLE_CORS_ALLOW_ORIGINS`: Settings for cross-origin resource sharing.
+    - `WEB_API_CORS_ALLOW_ORIGINS`, `CONSOLE_CORS_ALLOW_ORIGINS`: Settings for cross-origin resource sharing.
 
 9. **Other Service-Specific Environment Variables**:
-   - Each service like `nginx`, `redis`, `db`, and vector databases have specific environment variables that are directly referenced in the `docker-compose.yaml`.
-
+    - Each service like `nginx`, `redis`, `db`, and vector databases have specific environment variables that are directly referenced in the `docker-compose.yaml`.
 
 ### Additional Information
-- **Continuous Improvement Phase**: We are actively seeking feedback from the community to refine and enhance the deployment process. As more users adopt this new method, we will continue to make improvements based on your experiences and suggestions.
-- **Support**: For detailed configuration options and environment variable settings, refer to the `.env.example` file and the Docker Compose configuration files in the `docker` directory.
+
+- **Continuous Improvement Phase
+  **: We are actively seeking feedback from the community to refine and enhance the deployment process. As more users adopt this new method, we will continue to make improvements based on your experiences and suggestions.
+- **Support
+  **: For detailed configuration options and environment variable settings, refer to the `.env.example` file and the Docker Compose configuration files in the `docker` directory.
 
 This README aims to guide you through the deployment process using the new Docker Compose setup. For any issues or further assistance, please refer to the official documentation or contact support.

docker/certbot/README.md

@@ -0,0 +1,76 @@
+# Launching new servers with SSL certificates
+
+## Short description
+
+Docker-compose certbot configurations with Backward compatibility (without certbot container).  
+Use `docker-compose --profile certbot up` to use this features.
+
+## The simplest way for launching new servers with SSL certificates
+
+1. Get letsencrypt certs  
+   set `.env` values
+   ```properties
+   NGINX_SSL_CERT_FILENAME=fullchain.pem
+   NGINX_SSL_CERT_KEY_FILENAME=privkey.pem
+   NGINX_ENABLE_CERTBOT_CHALLENGE=true
+   CERTBOT_DOMAIN=your_domain.com
+   CERTBOT_EMAIL=example@your_domain.com
+   ```
+   excecute command:
+   ```shell
+   sudo docker network prune
+   sudo docker-compose --profile certbot up --force-recreate -d
+   ```
+   then after the containers launched:
+   ```shell
+   sudo docker-compose exec -it certbot /bin/sh /update-cert.sh
+   ```
+2. Edit `.env` file and `sudo docker-compose --profile certbot up` again.  
+   set `.env` value additionally
+   ```properties
+   NGINX_HTTPS_ENABLED=true
+   ```
+   excecute command:
+   ```shell
+   sudo docker-compose --profile certbot up -d --no-deps --force-recreate nginx
+   ```
+   Then you can access your serve with HTTPS.  
+   [https://your_domain.com](https://your_domain.com)
+
+## SSL certificates renewal
+
+For SSL certificates renewal, execute commands below:
+
+```shell
+sudo docker-compose exec -it certbot /bin/sh /update-cert.sh
+sudo docker-compose exec nginx nginx -s reload
+```
+
+## Options for certbot
+
+`CERTBOT_OPTIONS` key might be helpful for testing. i.e.,
+
+```properties
+CERTBOT_OPTIONS=--dry-run
+```
+
+To apply changes to `CERTBOT_OPTIONS`, regenerate the certbot container before updating the certificates.
+
+```shell
+sudo docker-compose --profile certbot up -d --no-deps --force-recreate certbot
+sudo docker-compose exec -it certbot /bin/sh /update-cert.sh
+```
+
+Then, reload the nginx container if necessary.
+
+```shell
+sudo docker-compose exec nginx nginx -s reload
+```
+
+## For legacy servers
+
+To use cert files dir `nginx/ssl` as before, simply launch containers WITHOUT `--profile certbot` option.
+
+```shell
+sudo docker-compose up -d
+```

docker/certbot/docker-entrypoint.sh

@@ -0,0 +1,30 @@
+#!/bin/sh
+set -e
+
+printf '%s\n' "Docker entrypoint script is running"
+
+printf '%s\n' "\nChecking specific environment variables:"
+printf '%s\n' "CERTBOT_EMAIL: ${CERTBOT_EMAIL:-Not set}"
+printf '%s\n' "CERTBOT_DOMAIN: ${CERTBOT_DOMAIN:-Not set}"
+printf '%s\n' "CERTBOT_OPTIONS: ${CERTBOT_OPTIONS:-Not set}"
+
+printf '%s\n' "\nChecking mounted directories:"
+for dir in "/etc/letsencrypt" "/var/www/html" "/var/log/letsencrypt"; do
+    if [ -d "$dir" ]; then
+        printf '%s\n' "$dir exists. Contents:"
+        ls -la "$dir"
+    else
+        printf '%s\n' "$dir does not exist."
+    fi
+done
+
+printf '%s\n' "\nGenerating update-cert.sh from template"
+sed -e "s|\${CERTBOT_EMAIL}|$CERTBOT_EMAIL|g" \
+    -e "s|\${CERTBOT_DOMAIN}|$CERTBOT_DOMAIN|g" \
+    -e "s|\${CERTBOT_OPTIONS}|$CERTBOT_OPTIONS|g" \
+    /update-cert.template.txt > /update-cert.sh
+
+chmod +x /update-cert.sh
+
+printf '%s\n' "\nExecuting command:" "$@"
+exec "$@"

docker/certbot/update-cert.template.txt

@@ -0,0 +1,19 @@
+#!/bin/bash
+set -e
+
+DOMAIN="${CERTBOT_DOMAIN}"
+EMAIL="${CERTBOT_EMAIL}"
+OPTIONS="${CERTBOT_OPTIONS}"
+CERT_NAME="${DOMAIN}" # 証明書名をドメイン名と同じにする
+
+# Check if the certificate already exists
+if [ -f "/etc/letsencrypt/renewal/${CERT_NAME}.conf" ]; then
+  echo "Certificate exists. Attempting to renew..."
+  certbot renew --noninteractive --cert-name ${CERT_NAME} --webroot --webroot-path=/var/www/html --email ${EMAIL} --agree-tos --no-eff-email ${OPTIONS}
+else
+  echo "Certificate does not exist. Obtaining a new certificate..."
+  certbot certonly --noninteractive --webroot --webroot-path=/var/www/html --email ${EMAIL} --agree-tos --no-eff-email -d ${DOMAIN} ${OPTIONS}
+fi
+echo "Certificate operation successful"
+# Note: Nginx reload should be handled outside this container
+echo "Please ensure to reload Nginx to apply any certificate changes."

docker/docker-compose.yaml

@@ -22,6 +22,9 @@ x-shared-env: &shared-api-worker-env
   CELERY_WORKER_CLASS: ${CELERY_WORKER_CLASS:-}
   GUNICORN_TIMEOUT: ${GUNICORN_TIMEOUT:-360}
   CELERY_WORKER_AMOUNT: ${CELERY_WORKER_AMOUNT:-}
+  CELERY_AUTO_SCALE: ${CELERY_AUTO_SCALE:-false}
+  CELERY_MAX_WORKERS: ${CELERY_MAX_WORKERS:-}
+  CELERY_MIN_WORKERS: ${CELERY_MIN_WORKERS:-}
   API_TOOL_DEFAULT_CONNECT_TIMEOUT: ${API_TOOL_DEFAULT_CONNECT_TIMEOUT:-10}
   API_TOOL_DEFAULT_READ_TIMEOUT: ${API_TOOL_DEFAULT_READ_TIMEOUT:-60}
   DB_USERNAME: ${DB_USERNAME:-postgres}
@@ -295,6 +298,26 @@ services:
       - ssrf_proxy_network
       - default
 
+  # Certbot service
+  # use `docker-compose --profile certbot up` to start the certbot service.
+  certbot:
+    image: certbot/certbot
+    profiles:
+      - certbot
+    volumes:
+      - ./volumes/certbot/conf:/etc/letsencrypt
+      - ./volumes/certbot/www:/var/www/html
+      - ./volumes/certbot/logs:/var/log/letsencrypt
+      - ./volumes/certbot/conf/live:/etc/letsencrypt/live
+      - ./certbot/update-cert.template.txt:/update-cert.template.txt
+      - ./certbot/docker-entrypoint.sh:/docker-entrypoint.sh
+    environment:
+      - CERTBOT_EMAIL=${CERTBOT_EMAIL}
+      - CERTBOT_DOMAIN=${CERTBOT_DOMAIN}
+      - CERTBOT_OPTIONS=${CERTBOT_OPTIONS:-}
+    entrypoint: [ "/docker-entrypoint.sh" ]
+    command: ["tail", "-f", "/dev/null"]
+
   # The nginx reverse proxy.
   # used for reverse proxying the API service and Web service.
   nginx:
@@ -306,7 +329,10 @@ services:
       - ./nginx/https.conf.template:/etc/nginx/https.conf.template
       - ./nginx/conf.d:/etc/nginx/conf.d
       - ./nginx/docker-entrypoint.sh:/docker-entrypoint-mount.sh
-      - ./nginx/ssl:/etc/ssl
+      - ./nginx/ssl:/etc/ssl # cert dir (legacy)
+      - ./volumes/certbot/conf/live:/etc/letsencrypt/live # cert dir (with certbot container)
+      - ./volumes/certbot/conf:/etc/letsencrypt
+      - ./volumes/certbot/www:/var/www/html
     entrypoint: [ "sh", "-c", "cp /docker-entrypoint-mount.sh /docker-entrypoint.sh && sed -i 's/\r$$//' /docker-entrypoint.sh && chmod +x /docker-entrypoint.sh && /docker-entrypoint.sh" ]
     environment:
       NGINX_SERVER_NAME: ${NGINX_SERVER_NAME:-_}
@@ -323,6 +349,8 @@ services:
       NGINX_KEEPALIVE_TIMEOUT: ${NGINX_KEEPALIVE_TIMEOUT:-65}
       NGINX_PROXY_READ_TIMEOUT: ${NGINX_PROXY_READ_TIMEOUT:-3600s}
       NGINX_PROXY_SEND_TIMEOUT: ${NGINX_PROXY_SEND_TIMEOUT:-3600s}
+      NGINX_ENABLE_CERTBOT_CHALLENGE: ${NGINX_ENABLE_CERTBOT_CHALLENGE:-false}
+      CERTBOT_DOMAIN: ${CERTBOT_DOMAIN:-}
     depends_on:
       - api
       - web
@@ -453,7 +481,7 @@ services:
       - ./volumes/milvus/etcd:/etcd
     command: etcd -advertise-client-urls=http://127.0.0.1:2379 -listen-client-urls http://0.0.0.0:2379 --data-dir /etcd
     healthcheck:
-      test: ["CMD", "etcdctl", "endpoint", "health"]
+      test: [ "CMD", "etcdctl", "endpoint", "health" ]
       interval: 30s
       timeout: 20s
       retries: 3
@@ -472,7 +500,7 @@ services:
       - ./volumes/milvus/minio:/minio_data
     command: minio server /minio_data --console-address ":9001"
     healthcheck:
-      test: ["CMD", "curl", "-f", "http://localhost:9000/minio/health/live"]
+      test: [ "CMD", "curl", "-f", "http://localhost:9000/minio/health/live" ]
       interval: 30s
       timeout: 20s
       retries: 3
@@ -484,7 +512,7 @@ services:
     image: milvusdb/milvus:v2.3.1
     profiles:
       - milvus
-    command: ["milvus", "run", "standalone"]
+    command: [ "milvus", "run", "standalone" ]
     environment:
       ETCD_ENDPOINTS: ${ETCD_ENDPOINTS:-etcd:2379}
       MINIO_ADDRESS: ${MINIO_ADDRESS:-minio:9000}
@@ -492,7 +520,7 @@ services:
     volumes:
       - ./volumes/milvus/milvus:/var/lib/milvus
     healthcheck:
-      test: ["CMD", "curl", "-f", "http://localhost:9091/healthz"]
+      test: [ "CMD", "curl", "-f", "http://localhost:9091/healthz" ]
       interval: 30s
       start_period: 90s
       timeout: 20s

docker/nginx/conf.d/default.conf.template

@@ -29,6 +29,9 @@ server {
       include proxy.conf;
     }
 
+    # placeholder for acme challenge location
+    ${ACME_CHALLENGE_LOCATION}
+
     # placeholder for https config defined in https.conf.template
     ${HTTPS_CONFIG}
 }

docker/nginx/docker-entrypoint.sh

@@ -1,6 +1,19 @@
 #!/bin/bash
 
 if [ "${NGINX_HTTPS_ENABLED}" = "true" ]; then
+    # Check if the certificate and key files for the specified domain exist
+    if [ -n "${CERTBOT_DOMAIN}" ] && \
+       [ -f "/etc/letsencrypt/live/${CERTBOT_DOMAIN}/${NGINX_SSL_CERT_FILENAME}" ] && \
+       [ -f "/etc/letsencrypt/live/${CERTBOT_DOMAIN}/${NGINX_SSL_CERT_KEY_FILENAME}" ]; then
+        SSL_CERTIFICATE_PATH="/etc/letsencrypt/live/${CERTBOT_DOMAIN}/${NGINX_SSL_CERT_FILENAME}"
+        SSL_CERTIFICATE_KEY_PATH="/etc/letsencrypt/live/${CERTBOT_DOMAIN}/${NGINX_SSL_CERT_KEY_FILENAME}"
+    else
+        SSL_CERTIFICATE_PATH="/etc/ssl/${NGINX_SSL_CERT_FILENAME}"
+        SSL_CERTIFICATE_KEY_PATH="/etc/ssl/${NGINX_SSL_CERT_KEY_FILENAME}"
+    fi
+    export SSL_CERTIFICATE_PATH
+    export SSL_CERTIFICATE_KEY_PATH
+
     # set the HTTPS_CONFIG environment variable to the content of the https.conf.template
     HTTPS_CONFIG=$(envsubst < /etc/nginx/https.conf.template)
     export HTTPS_CONFIG
@@ -8,6 +21,13 @@ if [ "${NGINX_HTTPS_ENABLED}" = "true" ]; then
     envsubst '${HTTPS_CONFIG}' < /etc/nginx/conf.d/default.conf.template > /etc/nginx/conf.d/default.conf
 fi
 
+if [ "${NGINX_ENABLE_CERTBOT_CHALLENGE}" = "true" ]; then
+    ACME_CHALLENGE_LOCATION='location /.well-known/acme-challenge/ { root /var/www/html; }'
+else
+    ACME_CHALLENGE_LOCATION=''
+fi
+export ACME_CHALLENGE_LOCATION
+
 env_vars=$(printenv | cut -d= -f1 | sed 's/^/$/g' | paste -sd, -)
 
 envsubst "$env_vars" < /etc/nginx/nginx.conf.template > /etc/nginx/nginx.conf

docker/nginx/https.conf.template

@@ -1,8 +1,8 @@
 # Please do not directly edit this file. Instead, modify the .env variables related to NGINX configuration.
 
 listen ${NGINX_SSL_PORT} ssl;
-ssl_certificate ./../ssl/${NGINX_SSL_CERT_FILENAME};
-ssl_certificate_key ./../ssl/${NGINX_SSL_CERT_KEY_FILENAME};
+ssl_certificate ${SSL_CERTIFICATE_PATH};
+ssl_certificate_key ${SSL_CERTIFICATE_KEY_PATH};
 ssl_protocols ${NGINX_SSL_PROTOCOLS};
 ssl_prefer_server_ciphers on;
 ssl_session_cache shared:SSL:10m;

web/app/components/app/configuration/dataset-config/params-config/config-content.tsx

@@ -32,6 +32,7 @@ import { RerankingModeEnum } from '@/models/datasets'
 import cn from '@/utils/classnames'
 import { useSelectedDatasetsMode } from '@/app/components/workflow/nodes/knowledge-retrieval/hooks'
 import Switch from '@/app/components/base/switch'
+import { useGetLanguage } from '@/context/i18n'
 
 type Props = {
   datasetConfigs: DatasetConfigs
@@ -43,6 +44,11 @@ type Props = {
   selectedDatasets?: DataSet[]
 }
 
+const LEGACY_LINK_MAP = {
+  en_US: 'https://docs.dify.ai/guides/knowledge-base/integrate-knowledge-within-application',
+  zh_Hans: 'https://docs.dify.ai/v/zh-hans/guides/knowledge-base/integrate_knowledge_within_application',
+} as Record<string, string>
+
 const ConfigContent: FC<Props> = ({
   datasetConfigs,
   onChange,
@@ -53,6 +59,7 @@ const ConfigContent: FC<Props> = ({
   selectedDatasets = [],
 }) => {
   const { t } = useTranslation()
+  const language = useGetLanguage()
   const selectedDatasetsMode = useSelectedDatasetsMode(selectedDatasets)
   const type = datasetConfigs.retrieval_model
   const setType = (value: RETRIEVE_TYPE) => {
@@ -167,7 +174,21 @@ const ConfigContent: FC<Props> = ({
           title={(
             <div className='flex items-center'>
               {t('appDebug.datasetConfig.retrieveOneWay.title')}
-              <TooltipPlus popupContent={<div className='w-[320px]'>{t('dataset.nTo1RetrievalLegacy')}</div>}>
+              <TooltipPlus
+                popupContent={(
+                  <div className='w-[320px]'>
+                    {t('dataset.nTo1RetrievalLegacy')}
+                    <a
+                      className='underline'
+                      href={LEGACY_LINK_MAP[language]}
+                      target='_blank'
+                      rel='noopener noreferrer'
+                    >
+                      ({t('dataset.nTo1RetrievalLegacyLink')})
+                    </a>
+                  </div>
+                )}
+              >
                 <div className='ml-1 flex items-center px-[5px] h-[18px] rounded-[5px] border border-text-accent-secondary system-2xs-medium-uppercase text-text-accent-secondary'>legacy</div>
               </TooltipPlus>
             </div>

web/app/components/base/tag-management/filter.tsx

@@ -109,11 +109,11 @@ const TagFilter: FC<TagFilterProps> = ({
           </div>
         </PortalToFollowElemTrigger>
         <PortalToFollowElemContent className='z-[1002]'>
-          <div className='relative w-[240px] bg-white rounded-lg border-[0.5px] border-gray-200  shadow-lg'>
+          <div className='relative w-[240px] bg-white rounded-lg border-[0.5px] border-gray-200 shadow-lg'>
             <div className='p-2 border-b-[0.5px] border-black/5'>
               <SearchInput white value={keywords} onChange={handleKeywordsChange} />
             </div>
-            <div className='p-1'>
+            <div className='p-1 max-h-72 overflow-auto'>
               {filteredTagList.map(tag => (
                 <div
                   key={tag.id}

web/app/components/workflow/header/undo-redo.tsx

@@ -30,7 +30,7 @@ const UndoRedo: FC<UndoRedoProps> = ({ handleUndo, handleRedo }) => {
 
   return (
     <div className='flex items-center p-0.5 rounded-lg border-[0.5px] border-gray-100 bg-white shadow-lg text-gray-500'>
-      <TipPopup title={t('workflow.common.undo')!} >
+      <TipPopup title={t('workflow.common.undo')!} shortcuts={['ctrl', 'z']}>
         <div
           data-tooltip-id='workflow.undo'
           className={`
@@ -43,7 +43,7 @@ const UndoRedo: FC<UndoRedoProps> = ({ handleUndo, handleRedo }) => {
           <RiArrowGoBackLine className='h-4 w-4' />
         </div>
       </TipPopup>
-      <TipPopup title={t('workflow.common.redo')!} >
+      <TipPopup title={t('workflow.common.redo')!} shortcuts={['ctrl', 'y']}>
         <div
           data-tooltip-id='workflow.redo'
           className={`

web/app/components/workflow/nodes/_base/components/option-card.tsx

@@ -0,0 +1,62 @@
+'use client'
+import type { FC } from 'react'
+import React, { useCallback } from 'react'
+import type { VariantProps } from 'class-variance-authority'
+import { cva } from 'class-variance-authority'
+import cn from '@/utils/classnames'
+
+const variants = cva([], {
+  variants: {
+    align: {
+      left: 'justify-start',
+      center: 'justify-center',
+      right: 'justify-end',
+    },
+  },
+  defaultVariants: {
+    align: 'center',
+  },
+},
+)
+
+type Props = {
+  className?: string
+  title: string
+  onSelect: () => void
+  selected: boolean
+  disabled?: boolean
+  align?: 'left' | 'center' | 'right'
+} & VariantProps<typeof variants>
+
+const OptionCard: FC<Props> = ({
+  className,
+  title,
+  onSelect,
+  selected,
+  disabled,
+  align = 'center',
+}) => {
+  const handleSelect = useCallback(() => {
+    if (selected || disabled)
+      return
+    onSelect()
+  }, [onSelect, selected, disabled])
+
+  return (
+    <div
+      className={cn(
+        'flex items-center px-2 h-8 rounded-md system-sm-regular bg-components-option-card-option-bg border border-components-option-card-option-bg text-text-secondary cursor-default',
+        (!selected && !disabled) && 'hover:bg-components-option-card-option-bg-hover hover:border-components-option-card-option-border-hover hover:shadow-xs cursor-pointer',
+        selected && 'bg-components-option-card-option-selected-bg border-[1.5px] border-components-option-card-option-selected-border system-sm-medium shadow-xs',
+        disabled && 'text-text-disabled',
+        variants({ align }),
+        className,
+      )}
+      onClick={handleSelect}
+    >
+      {title}
+    </div>
+  )
+}
+
+export default React.memo(OptionCard)

web/app/components/workflow/nodes/llm/components/resolution-picker.tsx

@@ -2,35 +2,11 @@
 import type { FC } from 'react'
 import React, { useCallback } from 'react'
 import { useTranslation } from 'react-i18next'
-import cn from '@/utils/classnames'
+import OptionCard from '@/app/components/workflow/nodes/_base/components/option-card'
 import { Resolution } from '@/types/app'
 
 const i18nPrefix = 'workflow.nodes.llm'
 
-type ItemProps = {
-  title: string
-  value: Resolution
-  onSelect: (value: Resolution) => void
-  isSelected: boolean
-}
-
-const Item: FC<ItemProps> = ({ title, value, onSelect, isSelected }) => {
-  const handleSelect = useCallback(() => {
-    if (isSelected)
-      return
-    onSelect(value)
-  }, [value, onSelect, isSelected])
-
-  return (
-    <div
-      className={cn(isSelected ? 'bg-white border-[2px] border-primary-400  shadow-xs' : 'bg-gray-25 border border-gray-100', 'flex items-center h-8 px-3 rounded-lg text-[13px] font-normal text-gray-900 cursor-pointer')}
-      onClick={handleSelect}
-    >
-      {title}
-    </div>
-  )
-}
-
 type Props = {
   value: Resolution
   onChange: (value: Resolution) => void
@@ -42,21 +18,24 @@ const ResolutionPicker: FC<Props> = ({
 }) => {
   const { t } = useTranslation()
 
+  const handleOnChange = useCallback((value: Resolution) => {
+    return () => {
+      onChange(value)
+    }
+  }, [onChange])
   return (
     <div className='flex items-center justify-between'>
       <div className='mr-2 text-xs font-medium text-gray-500 uppercase'>{t(`${i18nPrefix}.resolution.name`)}</div>
       <div className='flex items-center space-x-1'>
-        <Item
+        <OptionCard
           title={t(`${i18nPrefix}.resolution.high`)}
-          value={Resolution.high}
-          onSelect={onChange}
-          isSelected={value === Resolution.high}
+          onSelect={handleOnChange(Resolution.high)}
+          selected={value === Resolution.high}
         />
-        <Item
+        <OptionCard
           title={t(`${i18nPrefix}.resolution.low`)}
-          value={Resolution.low}
-          onSelect={onChange}
-          isSelected={value === Resolution.low}
+          onSelect={handleOnChange(Resolution.low)}
+          selected={value === Resolution.low}
         />
       </div>
     </div>

web/app/components/workflow/operator/control.tsx

@@ -16,7 +16,7 @@ import {
   useSelectionInteractions,
   useWorkflow,
 } from '../hooks'
-import { isEventTargetInputArea } from '../utils'
+import { getKeyboardKeyCodeBySystem, isEventTargetInputArea } from '../utils'
 import { useStore } from '../store'
 import AddBlock from './add-block'
 import TipPopup from './tip-popup'
@@ -78,6 +78,11 @@ const Control = () => {
     handleLayout()
   }
 
+  useKeyPress(`${getKeyboardKeyCodeBySystem('ctrl')}.o`, (e) => {
+    e.preventDefault()
+    goLayout()
+  }, { exactMatch: true, useCapture: true })
+
   const addNote = (e: MouseEvent<HTMLDivElement>) => {
     if (getNodesReadOnly())
       return
@@ -101,7 +106,7 @@ const Control = () => {
         </div>
       </TipPopup>
       <div className='mx-[3px] w-[1px] h-3.5 bg-gray-200'></div>
-      <TipPopup title={t('workflow.common.pointerMode')}>
+      <TipPopup title={t('workflow.common.pointerMode')} shortcuts={['v']}>
         <div
           className={cn(
             'flex items-center justify-center mr-[1px] w-8 h-8 rounded-lg cursor-pointer',
@@ -113,7 +118,7 @@ const Control = () => {
           <RiCursorLine className='w-4 h-4' />
         </div>
       </TipPopup>
-      <TipPopup title={t('workflow.common.handMode')}>
+      <TipPopup title={t('workflow.common.handMode')} shortcuts={['h']}>
         <div
           className={cn(
             'flex items-center justify-center w-8 h-8 rounded-lg cursor-pointer',
@@ -126,7 +131,7 @@ const Control = () => {
         </div>
       </TipPopup>
       <div className='mx-[3px] w-[1px] h-3.5 bg-gray-200'></div>
-      <TipPopup title={t('workflow.panel.organizeBlocks')}>
+      <TipPopup title={t('workflow.panel.organizeBlocks')} shortcuts={['ctrl', 'o']}>
         <div
           className={cn(
             'flex items-center justify-center w-8 h-8 rounded-lg hover:bg-black/5 hover:text-gray-700 cursor-pointer',

web/hooks/use-knowledge.ts

@@ -8,7 +8,10 @@ export const useKnowledge = () => {
     return t(`dataset.indexingTechnique.${indexingTechnique}`)
   }, [t])
 
-  const formatIndexingMethod = useCallback((indexingMethod: string) => {
+  const formatIndexingMethod = useCallback((indexingMethod: string, isEco?: boolean) => {
+    if (isEco)
+      return t('dataset.indexingMethod.invertedIndex')
+
     return t(`dataset.indexingMethod.${indexingMethod}`)
   }, [t])
 
@@ -16,7 +19,7 @@ export const useKnowledge = () => {
     let result = formatIndexingTechnique(indexingTechnique)
 
     if (indexingMethod)
-      result += ` · ${formatIndexingMethod(indexingMethod)}`
+      result += ` · ${formatIndexingMethod(indexingMethod, indexingTechnique === 'economy')}`
 
     return result
   }, [formatIndexingTechnique, formatIndexingMethod])

web/i18n/en-US/app-debug.ts

@@ -409,7 +409,7 @@ const translation = {
     },
     retrieveMultiWay: {
       title: 'Multi-path retrieval',
-      description: 'Based on user intent, queries across all Knowledge, retrieves relevant text from multi-sources, and selects the best results matching the user query after reranking. Configuration of the Rerank model API is required.',
+      description: 'Based on user intent, queries across all Knowledge, retrieves relevant text from multi-sources, and selects the best results matching the user query after reranking. ',
     },
     rerankModelRequired: 'Rerank model is required',
     params: 'Params',

web/i18n/en-US/dataset.ts

@@ -33,7 +33,7 @@ const translation = {
     },
     hybrid_search: {
       title: 'Hybrid Search',
-      description: 'Execute full-text search and vector searches simultaneously, re-rank to select the best match for the user\'s query. Configuration of the Rerank model APIs necessary.',
+      description: 'Execute full-text search and vector searches simultaneously, re-rank to select the best match for the user\'s query. Users can choose to set weights or configure to a Rerank model.',
       recommend: 'Recommend',
     },
     invertedIndex: {
@@ -53,6 +53,7 @@ const translation = {
     semantic_search: 'VECTOR',
     full_text_search: 'FULL TEXT',
     hybrid_search: 'HYBRID',
+    invertedIndex: 'INVERTED',
   },
   mixtureHighQualityAndEconomicTip: 'The Rerank model is required for mixture of high quality and economical knowledge bases.',
   inconsistentEmbeddingModelTip: 'The Rerank model is required if the Embedding models of the selected knowledge bases are inconsistent.',
@@ -67,7 +68,8 @@ const translation = {
     semantic: 'Semantic',
     keyword: 'Keyword',
   },
-  nTo1RetrievalLegacy: 'According to product planning, N-to-1 retrieval will be officially deprecated in September. Until then you can still use it normally.',
+  nTo1RetrievalLegacy: 'N-to-1 retrieval will be officially deprecated from September. It is recommended to use the latest Multi-path retrieval to obtain better results. ',
+  nTo1RetrievalLegacyLink: 'Learn more',
 }
 
 export default translation

web/i18n/zh-Hans/app-debug.ts

@@ -404,7 +404,7 @@ const translation = {
     },
     retrieveMultiWay: {
       title: '多路召回',
-      description: '根据用户意图同时匹配所有知识库，从多路知识库查询相关文本片段，经过重排序步骤，从多路查询结果中选择匹配用户问题的最佳结果，需配置 Rerank 模型 API。',
+      description: '根据用户意图同时匹配所有知识库，从多路知识库查询相关文本片段，经过重排序步骤，从多路查询结果中选择匹配用户问题的最佳结果。',
     },
     rerankModelRequired: '请选择 Rerank 模型',
     params: '参数设置',

web/i18n/zh-Hans/dataset.ts

@@ -33,7 +33,7 @@ const translation = {
     },
     hybrid_search: {
       title: '混合检索',
-      description: '同时执行全文检索和向量检索，并应用重排序步骤，从两类查询结果中选择匹配用户问题的最佳结果，需配置 Rerank 模型 API',
+      description: '同时执行全文检索和向量检索，并应用重排序步骤，从两类查询结果中选择匹配用户问题的最佳结果，用户可以选择设置权重或配置重新排序模型。',
       recommend: '推荐',
     },
     invertedIndex: {
@@ -53,6 +53,7 @@ const translation = {
     semantic_search: '向量检索',
     full_text_search: '全文检索',
     hybrid_search: '混合检索',
+    invertedIndex: '倒排索引',
   },
   mixtureHighQualityAndEconomicTip: '混合使用高质量和经济型知识库需要配置 Rerank 模型。',
   inconsistentEmbeddingModelTip: '当所选知识库配置的 Embedding 模型不一致时，需要配置 Rerank 模型。',
@@ -67,7 +68,8 @@ const translation = {
     semantic: '语义',
     keyword: '关键词',
   },
-  nTo1RetrievalLegacy: '根据产品规划，N 选 1 召回将于 9 月正式弃用。在那之前，您仍然可以正常使用它。',
+  nTo1RetrievalLegacy: '9 月 1 日起我们将不再提供此能力，推荐使用最新的多路召回获得更好的检索效果。',
+  nTo1RetrievalLegacyLink: '了解更多',
 }
 
 export default translation