fix: margin in rerank switch

# Conflicts:
#	api/poetry.lock

.github/workflows/build-push.yml

@@ -5,7 +5,6 @@ on:
     branches:
       - "main"
       - "deploy/dev"
-      - "release/0.10.0-beta"
   release:
     types: [published]
 
@@ -126,7 +125,7 @@ jobs:
         with:
           images: ${{ env[matrix.image_name_env] }}
           tags: |
-            type=raw,value=latest,enable=${{ startsWith(github.ref, 'refs/tags/') && !contains(github.ref, '-') }}
+            type=raw,value=latest,enable=${{ startsWith(github.ref, 'refs/tags/') }}
             type=ref,event=branch
             type=sha,enable=true,priority=100,prefix=,suffix=,format=long
             type=raw,value=${{ github.ref_name }},enable=${{ startsWith(github.ref, 'refs/tags/') }}

api/.env.example

@@ -201,8 +201,6 @@ OPENSEARCH_SECURE=true
 UPLOAD_FILE_SIZE_LIMIT=15
 UPLOAD_FILE_BATCH_LIMIT=5
 UPLOAD_IMAGE_FILE_SIZE_LIMIT=10
-UPLOAD_VIDEO_FILE_SIZE_LIMIT=100
-UPLOAD_AUDIO_FILE_SIZE_LIMIT=50
 
 # Model Configuration
 MULTIMODAL_SEND_IMAGE_FORMAT=base64
@@ -277,7 +275,6 @@ INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH=1000
 WORKFLOW_MAX_EXECUTION_STEPS=500
 WORKFLOW_MAX_EXECUTION_TIME=1200
 WORKFLOW_CALL_MAX_DEPTH=5
-MAX_VARIABLE_SIZE=204800
 
 # App configuration
 APP_MAX_EXECUTION_TIME=1200

api/.vscode/launch.json.example

@@ -1,15 +1,8 @@
 {
     "version": "0.2.0",
-    "compounds": [
-        {
-            "name": "Launch Flask and Celery",
-            "configurations": ["Python: Flask", "Python: Celery"]
-        }
-    ],
     "configurations": [
         {
             "name": "Python: Flask",
-            "consoleName": "Flask",
             "type": "debugpy",
             "request": "launch",
             "python": "${workspaceFolder}/.venv/bin/python",
@@ -24,12 +17,12 @@
             },
             "args": [
                 "run",
+                "--host=0.0.0.0",
                 "--port=5001"
             ]
         },
         {
             "name": "Python: Celery",
-            "consoleName": "Celery",
             "type": "debugpy",
             "request": "launch",
             "python": "${workspaceFolder}/.venv/bin/python",
@@ -52,10 +45,10 @@
                 "-c",
                 "1",
                 "--loglevel",
-                "DEBUG",
+                "info",
                 "-Q",
                 "dataset,generation,mail,ops_trace,app_deletion"
             ]
-        }
+        },
     ]
-}
+}

api/app.py

@@ -117,7 +117,7 @@ def create_app() -> Flask:
 
     logging.basicConfig(
         level=app.config.get("LOG_LEVEL"),
-        format=app.config["LOG_FORMAT"],
+        format=app.config.get("LOG_FORMAT"),
         datefmt=app.config.get("LOG_DATEFORMAT"),
         handlers=log_handlers,
         force=True,
@@ -134,7 +134,6 @@ def create_app() -> Flask:
             return datetime.utcfromtimestamp(seconds).astimezone(timezone).timetuple()
 
         for handler in logging.root.handlers:
-            assert handler.formatter
             handler.formatter.converter = time_converter
     initialize_extensions(app)
     register_blueprints(app)

api/commands.py

@@ -19,7 +19,7 @@ from extensions.ext_redis import redis_client
 from libs.helper import email as email_validate
 from libs.password import hash_password, password_pattern, valid_password
 from libs.rsa import generate_key_pair
-from models import Tenant
+from models.account import Tenant
 from models.dataset import Dataset, DatasetCollectionBinding, DocumentSegment
 from models.dataset import Document as DatasetDocument
 from models.model import Account, App, AppAnnotationSetting, AppMode, Conversation, MessageAnnotation
@@ -449,14 +449,14 @@ def convert_to_agent_apps():
         # fetch first 1000 apps
         sql_query = """SELECT a.id AS id FROM apps a
             INNER JOIN app_model_configs am ON a.app_model_config_id=am.id
-            WHERE a.mode = 'chat'
-            AND am.agent_mode is not null
+            WHERE a.mode = 'chat' 
+            AND am.agent_mode is not null 
             AND (
-				am.agent_mode like '%"strategy": "function_call"%'
+				am.agent_mode like '%"strategy": "function_call"%' 
                 OR am.agent_mode  like '%"strategy": "react"%'
-			)
+			) 
             AND (
-				am.agent_mode like '{"enabled": true%'
+				am.agent_mode like '{"enabled": true%' 
                 OR am.agent_mode like '{"max_iteration": %'
 			) ORDER BY a.created_at DESC LIMIT 1000
         """

api/configs/feature/__init__.py

@@ -1,4 +1,4 @@
-from typing import Annotated, Literal, Optional
+from typing import Annotated, Optional
 
 from pydantic import AliasChoices, Field, HttpUrl, NegativeInt, NonNegativeInt, PositiveInt, computed_field
 from pydantic_settings import BaseSettings
@@ -11,11 +11,11 @@ class SecurityConfig(BaseSettings):
     Security-related configurations for the application
     """
 
-    SECRET_KEY: str = Field(
+    SECRET_KEY: Optional[str] = Field(
         description="Secret key for secure session cookie signing."
         "Make sure you are changing this key for your deployment with a strong key."
         "Generate a strong key using `openssl rand -base64 42` or set via the `SECRET_KEY` environment variable.",
-        default="",
+        default=None,
     )
 
     RESET_PASSWORD_TOKEN_EXPIRY_HOURS: PositiveInt = Field(
@@ -177,16 +177,6 @@ class FileUploadConfig(BaseSettings):
         default=10,
     )
 
-    UPLOAD_VIDEO_FILE_SIZE_LIMIT: NonNegativeInt = Field(
-        description="video file size limit in Megabytes for uploading files",
-        default=100,
-    )
-
-    UPLOAD_AUDIO_FILE_SIZE_LIMIT: NonNegativeInt = Field(
-        description="audio file size limit in Megabytes for uploading files",
-        default=50,
-    )
-
     BATCH_UPLOAD_LIMIT: NonNegativeInt = Field(
         description="Maximum number of files allowed in a batch upload operation",
         default=20,
@@ -359,8 +349,8 @@ class WorkflowConfig(BaseSettings):
     )
 
     MAX_VARIABLE_SIZE: PositiveInt = Field(
-        description="Maximum size in bytes for a single variable in workflows. Default to 200 KB.",
-        default=200 * 1024,
+        description="Maximum size in bytes for a single variable in workflows. Default to 5KB.",
+        default=5 * 1024,
     )
 
 
@@ -478,7 +468,6 @@ class RagEtlConfig(BaseSettings):
     Configuration for RAG ETL processes
     """
 
-    # TODO: This config is not only for rag etl, it is also for file upload, we should move it to file upload config
     ETL_TYPE: str = Field(
         description="RAG ETL type ('dify' or 'Unstructured'), default to 'dify'",
         default="dify",
@@ -540,7 +529,7 @@ class IndexingConfig(BaseSettings):
 
 
 class ImageFormatConfig(BaseSettings):
-    MULTIMODAL_SEND_IMAGE_FORMAT: Literal["base64", "url"] = Field(
+    MULTIMODAL_SEND_IMAGE_FORMAT: str = Field(
         description="Format for sending images in multimodal contexts ('base64' or 'url'), default is base64",
         default="base64",
     )

api/configs/middleware/__init__.py

@@ -5,6 +5,7 @@ from pydantic import Field, NonNegativeInt, PositiveFloat, PositiveInt, computed
 from pydantic_settings import BaseSettings
 
 from configs.middleware.cache.redis_config import RedisConfig
+from configs.middleware.external.bedrock_config import BedrockConfig
 from configs.middleware.storage.aliyun_oss_storage_config import AliyunOSSStorageConfig
 from configs.middleware.storage.amazon_s3_storage_config import S3StorageConfig
 from configs.middleware.storage.azure_blob_storage_config import AzureBlobStorageConfig
@@ -222,5 +223,6 @@ class MiddlewareConfig(
     TiDBVectorConfig,
     WeaviateConfig,
     ElasticsearchConfig,
+    BedrockConfig,
 ):
     pass

api/configs/middleware/external/bedrock_config.py

@@ -0,0 +1,20 @@
+from typing import Optional
+
+from pydantic import Field
+from pydantic_settings import BaseSettings
+
+
+class BedrockConfig(BaseSettings):
+    """
+    bedrock configs
+    """
+
+    AWS_SECRET_ACCESS_KEY: Optional[str] = Field(
+        description="AWS secret access key",
+        default=None,
+    )
+
+    AWS_ACCESS_KEY_ID: Optional[str] = Field(
+        description="AWS secret access id",
+        default=None,
+    )

api/configs/packaging/__init__.py

@@ -9,7 +9,7 @@ class PackagingInfo(BaseSettings):
 
     CURRENT_VERSION: str = Field(
         description="Dify version",
-        default="0.10.0-beta2",
+        default="0.8.3",
     )
 
     COMMIT_SHA: str = Field(

api/constants/__init__.py

@@ -1,21 +1,2 @@
-from configs import dify_config
-
 HIDDEN_VALUE = "[__HIDDEN__]"
 UUID_NIL = "00000000-0000-0000-0000-000000000000"
-
-IMAGE_EXTENSIONS = ["jpg", "jpeg", "png", "webp", "gif", "svg"]
-IMAGE_EXTENSIONS.extend([ext.upper() for ext in IMAGE_EXTENSIONS])
-
-VIDEO_EXTENSIONS = ["mp4", "mov", "mpeg", "mpga"]
-VIDEO_EXTENSIONS.extend([ext.upper() for ext in VIDEO_EXTENSIONS])
-
-AUDIO_EXTENSIONS = ["mp3", "m4a", "wav", "webm", "amr"]
-AUDIO_EXTENSIONS.extend([ext.upper() for ext in AUDIO_EXTENSIONS])
-
-DOCUMENT_EXTENSIONS = ["txt", "markdown", "md", "pdf", "html", "htm", "xlsx", "xls", "docx", "csv"]
-DOCUMENT_EXTENSIONS.extend([ext.upper() for ext in DOCUMENT_EXTENSIONS])
-
-if dify_config.ETL_TYPE == "Unstructured":
-    DOCUMENT_EXTENSIONS = ["txt", "markdown", "md", "pdf", "html", "htm", "xlsx", "xls"]
-    DOCUMENT_EXTENSIONS.extend(("docx", "csv", "eml", "msg", "pptx", "ppt", "xml", "epub"))
-    DOCUMENT_EXTENSIONS.extend([ext.upper() for ext in DOCUMENT_EXTENSIONS])

api/contexts/__init__.py

@@ -1,9 +1,7 @@
 from contextvars import ContextVar
-from typing import TYPE_CHECKING
 
-if TYPE_CHECKING:
-    from core.workflow.entities.variable_pool import VariablePool
+from core.workflow.entities.variable_pool import VariablePool
 
 tenant_id: ContextVar[str] = ContextVar("tenant_id")
 
-workflow_variable_pool: ContextVar["VariablePool"] = ContextVar("workflow_variable_pool")
+workflow_variable_pool: ContextVar[VariablePool] = ContextVar("workflow_variable_pool")

api/controllers/console/__init__.py

@@ -45,6 +45,7 @@ from .datasets import (
     external,
     file,
     hit_testing,
+    test_external,
     website,
 )
 

api/controllers/console/app/conversation.py

@@ -22,8 +22,7 @@ from fields.conversation_fields import (
 )
 from libs.helper import DatetimeString
 from libs.login import login_required
-from models import Conversation, EndUser, Message, MessageAnnotation
-from models.model import AppMode
+from models.model import AppMode, Conversation, EndUser, Message, MessageAnnotation
 
 
 class CompletionConversationApi(Resource):

api/controllers/console/app/site.py

@@ -12,7 +12,7 @@ from controllers.console.wraps import account_initialization_required
 from extensions.ext_database import db
 from fields.app_fields import app_site_fields
 from libs.login import login_required
-from models import Site
+from models.model import Site
 
 
 def parse_app_site_args():

api/controllers/console/app/workflow.py

@@ -13,14 +13,14 @@ from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
 from core.app.apps.base_app_queue_manager import AppQueueManager
 from core.app.entities.app_invoke_entities import InvokeFrom
-from factories import variable_factory
+from core.app.segments import factory
+from core.errors.error import AppInvokeQuotaExceededError
 from fields.workflow_fields import workflow_fields
 from fields.workflow_run_fields import workflow_run_node_execution_fields
 from libs import helper
 from libs.helper import TimestampField, uuid_value
 from libs.login import current_user, login_required
-from models import App
-from models.model import AppMode
+from models.model import App, AppMode
 from services.app_dsl_service import AppDslService
 from services.app_generate_service import AppGenerateService
 from services.errors.app import WorkflowHashNotEqualError
@@ -101,13 +101,9 @@ class DraftWorkflowApi(Resource):
 
         try:
             environment_variables_list = args.get("environment_variables") or []
-            environment_variables = [
-                variable_factory.build_variable_from_mapping(obj) for obj in environment_variables_list
-            ]
+            environment_variables = [factory.build_variable_from_mapping(obj) for obj in environment_variables_list]
             conversation_variables_list = args.get("conversation_variables") or []
-            conversation_variables = [
-                variable_factory.build_variable_from_mapping(obj) for obj in conversation_variables_list
-            ]
+            conversation_variables = [factory.build_variable_from_mapping(obj) for obj in conversation_variables_list]
             workflow = workflow_service.sync_draft_workflow(
                 app_model=app_model,
                 graph=args["graph"],
@@ -277,15 +273,17 @@ class DraftWorkflowRunApi(Resource):
         parser.add_argument("files", type=list, required=False, location="json")
         args = parser.parse_args()
 
-        response = AppGenerateService.generate(
-            app_model=app_model,
-            user=current_user,
-            args=args,
-            invoke_from=InvokeFrom.DEBUGGER,
-            streaming=True,
-        )
+        try:
+            response = AppGenerateService.generate(
+                app_model=app_model, user=current_user, args=args, invoke_from=InvokeFrom.DEBUGGER, streaming=True
+            )
 
-        return helper.compact_generate_response(response)
+            return helper.compact_generate_response(response)
+        except (ValueError, AppInvokeQuotaExceededError) as e:
+            raise e
+        except Exception as e:
+            logging.exception("internal server error.")
+            raise InternalServerError()
 
 
 class WorkflowTaskStopApi(Resource):

api/controllers/console/app/workflow_app_log.py

@@ -7,8 +7,7 @@ from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
 from fields.workflow_app_log_fields import workflow_app_log_pagination_fields
 from libs.login import login_required
-from models import App
-from models.model import AppMode
+from models.model import App, AppMode
 from services.workflow_app_service import WorkflowAppService
 
 

api/controllers/console/app/workflow_run.py

@@ -13,8 +13,7 @@ from fields.workflow_run_fields import (
 )
 from libs.helper import uuid_value
 from libs.login import login_required
-from models import App
-from models.model import AppMode
+from models.model import App, AppMode
 from services.workflow_run_service import WorkflowRunService
 
 

api/controllers/console/app/workflow_statistic.py

@@ -10,11 +10,11 @@ from controllers.console import api
 from controllers.console.app.wraps import get_app_model
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
-from enums import WorkflowRunTriggeredFrom
 from extensions.ext_database import db
 from libs.helper import DatetimeString
 from libs.login import login_required
 from models.model import AppMode
+from models.workflow import WorkflowRunTriggeredFrom
 
 
 class WorkflowDailyRunsStatistic(Resource):

api/controllers/console/app/wraps.py

@@ -5,8 +5,7 @@ from typing import Optional, Union
 from controllers.console.app.error import AppNotFoundError
 from extensions.ext_database import db
 from libs.login import current_user
-from models import App
-from models.model import AppMode
+from models.model import App, AppMode
 
 
 def get_app_model(view: Optional[Callable] = None, *, mode: Union[AppMode, list[AppMode]] = None):

api/controllers/console/auth/forgot_password.py

@@ -15,7 +15,7 @@ from controllers.console.setup import setup_required
 from extensions.ext_database import db
 from libs.helper import email as email_validate
 from libs.password import hash_password, valid_password
-from models import Account
+from models.account import Account
 from services.account_service import AccountService
 from services.errors.account import RateLimitExceededError
 

api/controllers/console/auth/login.py

@@ -9,7 +9,7 @@ from controllers.console import api
 from controllers.console.setup import setup_required
 from libs.helper import email, get_remote_ip
 from libs.password import valid_password
-from models import Account
+from models.account import Account
 from services.account_service import AccountService, TenantService
 
 

api/controllers/console/auth/oauth.py

@@ -11,8 +11,7 @@ from constants.languages import languages
 from extensions.ext_database import db
 from libs.helper import get_remote_ip
 from libs.oauth import GitHubOAuth, GoogleOAuth, OAuthUserInfo
-from models import Account
-from models.account import AccountStatus
+from models.account import Account, AccountStatus
 from services.account_service import AccountService, RegisterService, TenantService
 
 from .. import api

api/controllers/console/datasets/data_source.py

@@ -15,7 +15,8 @@ from core.rag.extractor.notion_extractor import NotionExtractor
 from extensions.ext_database import db
 from fields.data_source_fields import integrate_list_fields, integrate_notion_info_list_fields
 from libs.login import login_required
-from models import DataSourceOauthBinding, Document
+from models.dataset import Document
+from models.source import DataSourceOauthBinding
 from services.dataset_service import DatasetService, DocumentService
 from tasks.document_indexing_sync_task import document_indexing_sync_task
 

api/controllers/console/datasets/datasets.py

@@ -24,8 +24,8 @@ from fields.app_fields import related_app_list
 from fields.dataset_fields import dataset_detail_fields, dataset_query_detail_fields
 from fields.document_fields import document_status_fields
 from libs.login import login_required
-from models import ApiToken, Dataset, Document, DocumentSegment, UploadFile
-from models.dataset import DatasetPermissionEnum
+from models.dataset import Dataset, DatasetPermissionEnum, Document, DocumentSegment
+from models.model import ApiToken, UploadFile
 from services.dataset_service import DatasetPermissionService, DatasetService, DocumentService
 
 
@@ -613,10 +613,10 @@ class DatasetRetrievalSettingApi(Resource):
             case (
                 VectorType.MILVUS
                 | VectorType.RELYT
+                | VectorType.PGVECTOR
                 | VectorType.TIDB_VECTOR
                 | VectorType.CHROMA
                 | VectorType.TENCENT
-                | VectorType.PGVECTO_RS
             ):
                 return {"retrieval_method": [RetrievalMethod.SEMANTIC_SEARCH.value]}
             case (
@@ -627,7 +627,6 @@ class DatasetRetrievalSettingApi(Resource):
                 | VectorType.MYSCALE
                 | VectorType.ORACLE
                 | VectorType.ELASTICSEARCH
-                | VectorType.PGVECTOR
             ):
                 return {
                     "retrieval_method": [

api/controllers/console/datasets/datasets_document.py

@@ -46,7 +46,8 @@ from fields.document_fields import (
     document_with_segments_fields,
 )
 from libs.login import login_required
-from models import Dataset, DatasetProcessRule, Document, DocumentSegment, UploadFile
+from models.dataset import Dataset, DatasetProcessRule, Document, DocumentSegment
+from models.model import UploadFile
 from services.dataset_service import DatasetService, DocumentService
 from tasks.add_document_to_index_task import add_document_to_index_task
 from tasks.remove_document_from_index_task import remove_document_from_index_task

api/controllers/console/datasets/datasets_segments.py

@@ -24,7 +24,7 @@ from extensions.ext_database import db
 from extensions.ext_redis import redis_client
 from fields.segment_fields import segment_fields
 from libs.login import login_required
-from models import DocumentSegment
+from models.dataset import DocumentSegment
 from services.dataset_service import DatasetService, DocumentService, SegmentService
 from tasks.batch_create_segment_to_index_task import batch_create_segment_to_index_task
 from tasks.disable_segment_from_index_task import disable_segment_from_index_task

api/controllers/console/datasets/external.py

@@ -5,6 +5,7 @@ from werkzeug.exceptions import Forbidden, InternalServerError, NotFound
 
 import services
 from controllers.console import api
+from controllers.console.app.error import ProviderNotInitializeError
 from controllers.console.datasets.error import DatasetNameDuplicateError
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
@@ -157,6 +158,48 @@ class ExternalApiUseCheckApi(Resource):
         return {"is_using": external_knowledge_api_is_using, "count": count}, 200
 
 
+class ExternalDatasetInitApi(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def post(self):
+        # The role of the current user in the ta table must be admin, owner, or editor
+        if not current_user.is_editor:
+            raise Forbidden()
+
+        parser = reqparse.RequestParser()
+        parser.add_argument("external_knowledge_api_id", type=str, required=True, nullable=True, location="json")
+        # parser.add_argument('name', nullable=False, required=True,
+        #                     help='name is required. Name must be between 1 to 100 characters.',
+        #                     type=_validate_name)
+        # parser.add_argument('description', type=str, required=True, nullable=True, location='json')
+        parser.add_argument("data_source", type=dict, required=True, nullable=True, location="json")
+        parser.add_argument("process_parameter", type=dict, required=True, nullable=True, location="json")
+
+        args = parser.parse_args()
+
+        # The role of the current user in the ta table must be admin, owner, or editor, or dataset_operator
+        if not current_user.is_dataset_editor:
+            raise Forbidden()
+
+        # validate args
+        ExternalDatasetService.document_create_args_validate(
+            current_user.current_tenant_id, args["external_knowledge_api_id"], args["process_parameter"]
+        )
+
+        try:
+            dataset, documents, batch = ExternalDatasetService.init_external_dataset(
+                tenant_id=current_user.current_tenant_id,
+                user_id=current_user.id,
+                args=args,
+            )
+        except Exception as ex:
+            raise ProviderNotInitializeError(ex.description)
+        response = {"dataset": dataset, "documents": documents, "batch": batch}
+
+        return response
+
+
 class ExternalDatasetCreateApi(Resource):
     @setup_required
     @login_required

api/controllers/console/datasets/file.py

@@ -1,12 +1,9 @@
-import urllib.parse
-
 from flask import request
 from flask_login import current_user
 from flask_restful import Resource, marshal_with
 
 import services
 from configs import dify_config
-from constants import DOCUMENT_EXTENSIONS
 from controllers.console import api
 from controllers.console.datasets.error import (
     FileTooLargeError,
@@ -16,10 +13,9 @@ from controllers.console.datasets.error import (
 )
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required, cloud_edition_billing_resource_check
-from core.helper import ssrf_proxy
-from fields.file_fields import file_fields, remote_file_info_fields, upload_config_fields
+from fields.file_fields import file_fields, upload_config_fields
 from libs.login import login_required
-from services.file_service import FileService
+from services.file_service import ALLOWED_EXTENSIONS, UNSTRUCTURED_ALLOWED_EXTENSIONS, FileService
 
 PREVIEW_WORDS_LIMIT = 3000
 
@@ -55,7 +51,7 @@ class FileApi(Resource):
         if len(request.files) > 1:
             raise TooManyFilesError()
         try:
-            upload_file = FileService.upload_file(file=file, user=current_user)
+            upload_file = FileService.upload_file(file, current_user)
         except services.errors.file.FileTooLargeError as file_too_large_error:
             raise FileTooLargeError(file_too_large_error.description)
         except services.errors.file.UnsupportedFileTypeError:
@@ -79,24 +75,11 @@ class FileSupportTypeApi(Resource):
     @login_required
     @account_initialization_required
     def get(self):
-        return {"allowed_extensions": DOCUMENT_EXTENSIONS}
-
-
-class RemoteFileInfoApi(Resource):
-    @marshal_with(remote_file_info_fields)
-    def get(self, url):
-        decoded_url = urllib.parse.unquote(url)
-        try:
-            response = ssrf_proxy.head(decoded_url)
-            return {
-                "file_type": response.headers.get("Content-Type", "application/octet-stream"),
-                "file_length": int(response.headers.get("Content-Length", 0)),
-            }
-        except Exception as e:
-            return {"error": str(e)}, 400
+        etl_type = dify_config.ETL_TYPE
+        allowed_extensions = UNSTRUCTURED_ALLOWED_EXTENSIONS if etl_type == "Unstructured" else ALLOWED_EXTENSIONS
+        return {"allowed_extensions": allowed_extensions}
 
 
 api.add_resource(FileApi, "/files/upload")
 api.add_resource(FilePreviewApi, "/files/<uuid:file_id>/preview")
 api.add_resource(FileSupportTypeApi, "/files/support-type")
-api.add_resource(RemoteFileInfoApi, "/remote-files/<path:url>")

api/controllers/console/datasets/test_external.py

@@ -0,0 +1,33 @@
+from flask_restful import Resource, reqparse
+
+from controllers.console import api
+from controllers.console.setup import setup_required
+from controllers.console.wraps import account_initialization_required
+from libs.login import login_required
+from services.external_knowledge_service import ExternalDatasetService
+
+
+class TestExternalApi(Resource):
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument("retrieval_setting", nullable=False, required=True, type=dict, location="json")
+        parser.add_argument(
+            "query",
+            nullable=False,
+            required=True,
+            type=str,
+        )
+        parser.add_argument(
+            "knowledge_id",
+            nullable=False,
+            required=True,
+            type=str,
+        )
+        args = parser.parse_args()
+        result = ExternalDatasetService.test_external_knowledge_retrieval(
+            args["retrieval_setting"], args["query"], args["knowledge_id"]
+        )
+        return result, 200
+
+
+api.add_resource(TestExternalApi, "/retrieval")

api/controllers/console/datasets/website.py

@@ -14,9 +14,7 @@ class WebsiteCrawlApi(Resource):
     @account_initialization_required
     def post(self):
         parser = reqparse.RequestParser()
-        parser.add_argument(
-            "provider", type=str, choices=["firecrawl", "jinareader"], required=True, nullable=True, location="json"
-        )
+        parser.add_argument("provider", type=str, choices=["firecrawl"], required=True, nullable=True, location="json")
         parser.add_argument("url", type=str, required=True, nullable=True, location="json")
         parser.add_argument("options", type=dict, required=True, nullable=True, location="json")
         args = parser.parse_args()
@@ -35,7 +33,7 @@ class WebsiteCrawlStatusApi(Resource):
     @account_initialization_required
     def get(self, job_id: str):
         parser = reqparse.RequestParser()
-        parser.add_argument("provider", type=str, choices=["firecrawl", "jinareader"], required=True, location="args")
+        parser.add_argument("provider", type=str, choices=["firecrawl"], required=True, location="args")
         args = parser.parse_args()
         # get crawl status
         try:

api/controllers/console/explore/installed_app.py

@@ -11,7 +11,7 @@ from controllers.console.wraps import account_initialization_required, cloud_edi
 from extensions.ext_database import db
 from fields.installed_app_fields import installed_app_list_fields
 from libs.login import login_required
-from models import App, InstalledApp, RecommendedApp
+from models.model import App, InstalledApp, RecommendedApp
 from services.account_service import TenantService
 
 

api/controllers/console/explore/wraps.py

@@ -7,7 +7,7 @@ from werkzeug.exceptions import NotFound
 from controllers.console.wraps import account_initialization_required
 from extensions.ext_database import db
 from libs.login import login_required
-from models import InstalledApp
+from models.model import InstalledApp
 
 
 def installed_app_required(view=None):

api/controllers/console/version.py

@@ -38,52 +38,11 @@ class VersionApi(Resource):
             return result
 
         content = json.loads(response.content)
-        if _has_new_version(latest_version=content["version"], current_version=f"{args.get('current_version')}"):
-            result["version"] = content["version"]
-            result["release_date"] = content["releaseDate"]
-            result["release_notes"] = content["releaseNotes"]
-            result["can_auto_update"] = content["canAutoUpdate"]
+        result["version"] = content["version"]
+        result["release_date"] = content["releaseDate"]
+        result["release_notes"] = content["releaseNotes"]
+        result["can_auto_update"] = content["canAutoUpdate"]
         return result
 
 
-def _has_new_version(*, latest_version: str, current_version: str) -> bool:
-    def parse_version(version: str) -> tuple:
-        # Split version into parts and pre-release suffix if any
-        parts = version.split("-")
-        version_parts = parts[0].split(".")
-        pre_release = parts[1] if len(parts) > 1 else None
-
-        # Validate version format
-        if len(version_parts) != 3:
-            raise ValueError(f"Invalid version format: {version}")
-
-        try:
-            # Convert version parts to integers
-            major, minor, patch = map(int, version_parts)
-            return (major, minor, patch, pre_release)
-        except ValueError:
-            raise ValueError(f"Invalid version format: {version}")
-
-    latest = parse_version(latest_version)
-    current = parse_version(current_version)
-
-    # Compare major, minor, and patch versions
-    for latest_part, current_part in zip(latest[:3], current[:3]):
-        if latest_part > current_part:
-            return True
-        elif latest_part < current_part:
-            return False
-
-    # If versions are equal, check pre-release suffixes
-    if latest[3] is None and current[3] is not None:
-        return True
-    elif latest[3] is not None and current[3] is None:
-        return False
-    elif latest[3] is not None and current[3] is not None:
-        # Simple string comparison for pre-release versions
-        return latest[3] > current[3]
-
-    return False
-
-
 api.add_resource(VersionApi, "/version")

api/controllers/console/workspace/account.py

@@ -20,7 +20,7 @@ from extensions.ext_database import db
 from fields.member_fields import account_fields
 from libs.helper import TimestampField, timezone
 from libs.login import login_required
-from models import AccountIntegrate, InvitationCode
+from models.account import AccountIntegrate, InvitationCode
 from services.account_service import AccountService
 from services.errors.account import CurrentPasswordIncorrectError as ServiceCurrentPasswordIncorrectError
 

api/controllers/console/workspace/models.py

@@ -72,9 +72,8 @@ class DefaultModelApi(Resource):
                     provider=model_setting["provider"],
                     model=model_setting["model"],
                 )
-            except Exception as ex:
-                logging.exception(f"{model_setting['model_type']} save error: {ex}")
-                raise ex
+            except Exception:
+                logging.warning(f"{model_setting['model_type']} save error")
 
         return {"result": "success"}
 

api/controllers/console/workspace/tool_providers.py

@@ -360,15 +360,16 @@ class ToolWorkflowProviderCreateApi(Resource):
         args = reqparser.parse_args()
 
         return WorkflowToolManageService.create_workflow_tool(
-            user_id=user_id,
-            tenant_id=tenant_id,
-            workflow_app_id=args["workflow_app_id"],
-            name=args["name"],
-            label=args["label"],
-            icon=args["icon"],
-            description=args["description"],
-            parameters=args["parameters"],
-            privacy_policy=args["privacy_policy"],
+            user_id,
+            tenant_id,
+            args["workflow_app_id"],
+            args["name"],
+            args["label"],
+            args["icon"],
+            args["description"],
+            args["parameters"],
+            args["privacy_policy"],
+            args.get("labels", []),
         )
 
 

api/controllers/console/workspace/workspace.py

@@ -198,7 +198,7 @@ class WebappLogoWorkspaceApi(Resource):
             raise UnsupportedFileTypeError()
 
         try:
-            upload_file = FileService.upload_file(file=file, user=current_user)
+            upload_file = FileService.upload_file(file, current_user, True)
 
         except services.errors.file.FileTooLargeError as file_too_large_error:
             raise FileTooLargeError(file_too_large_error.description)

api/controllers/files/image_preview.py

@@ -21,36 +21,7 @@ class ImagePreviewApi(Resource):
             return {"content": "Invalid request."}, 400
 
         try:
-            generator, mimetype = FileService.get_image_preview(
-                file_id=file_id,
-                timestamp=timestamp,
-                nonce=nonce,
-                sign=sign,
-            )
-        except services.errors.file.UnsupportedFileTypeError:
-            raise UnsupportedFileTypeError()
-
-        return Response(generator, mimetype=mimetype)
-
-
-class FilePreviewApi(Resource):
-    def get(self, file_id):
-        file_id = str(file_id)
-
-        timestamp = request.args.get("timestamp")
-        nonce = request.args.get("nonce")
-        sign = request.args.get("sign")
-
-        if not timestamp or not nonce or not sign:
-            return {"content": "Invalid request."}, 400
-
-        try:
-            generator, mimetype = FileService.get_signed_file_preview(
-                file_id=file_id,
-                timestamp=timestamp,
-                nonce=nonce,
-                sign=sign,
-            )
+            generator, mimetype = FileService.get_image_preview(file_id, timestamp, nonce, sign)
         except services.errors.file.UnsupportedFileTypeError:
             raise UnsupportedFileTypeError()
 
@@ -78,7 +49,6 @@ class WorkspaceWebappLogoApi(Resource):
 
 
 api.add_resource(ImagePreviewApi, "/files/<uuid:file_id>/image-preview")
-api.add_resource(FilePreviewApi, "/files/<uuid:file_id>/file-preview")
 api.add_resource(WorkspaceWebappLogoApi, "/files/workspaces/<uuid:workspace_id>/webapp-logo")
 
 

api/controllers/web/file.py

@@ -1,5 +1,3 @@
-import urllib.parse
-
 from flask import request
 from flask_restful import marshal_with
 
@@ -7,8 +5,7 @@ import services
 from controllers.web import api
 from controllers.web.error import FileTooLargeError, NoFileUploadedError, TooManyFilesError, UnsupportedFileTypeError
 from controllers.web.wraps import WebApiResource
-from core.helper import ssrf_proxy
-from fields.file_fields import file_fields, remote_file_info_fields
+from fields.file_fields import file_fields
 from services.file_service import FileService
 
 
@@ -34,19 +31,4 @@ class FileApi(WebApiResource):
         return upload_file, 201
 
 
-class RemoteFileInfoApi(WebApiResource):
-    @marshal_with(remote_file_info_fields)
-    def get(self, url):
-        decoded_url = urllib.parse.unquote(url)
-        try:
-            response = ssrf_proxy.head(decoded_url)
-            return {
-                "file_type": response.headers.get("Content-Type", "application/octet-stream"),
-                "file_length": int(response.headers.get("Content-Length", 0)),
-            }
-        except Exception as e:
-            return {"error": str(e)}, 400
-
-
 api.add_resource(FileApi, "/files/upload")
-api.add_resource(RemoteFileInfoApi, "/remote-files/<path:url>")

api/controllers/web/message.py

@@ -22,7 +22,6 @@ from core.errors.error import ModelCurrentlyNotSupportError, ProviderTokenNotIni
 from core.model_runtime.errors.invoke import InvokeError
 from fields.conversation_fields import message_file_fields
 from fields.message_fields import agent_thought_fields
-from fields.raws import FilesContainedField
 from libs import helper
 from libs.helper import TimestampField, uuid_value
 from models.model import AppMode
@@ -59,7 +58,7 @@ class MessageListApi(WebApiResource):
         "id": fields.String,
         "conversation_id": fields.String,
         "parent_message_id": fields.String,
-        "inputs": FilesContainedField,
+        "inputs": fields.Raw,
         "query": fields.String,
         "answer": fields.String(attribute="re_sign_file_url_answer"),
         "message_files": fields.List(fields.Nested(message_file_fields), attribute="files"),

api/core/agent/base_agent_runner.py

@@ -16,14 +16,13 @@ from core.app.entities.app_invoke_entities import (
 )
 from core.callback_handler.agent_tool_callback_handler import DifyAgentCallbackHandler
 from core.callback_handler.index_tool_callback_handler import DatasetIndexToolCallbackHandler
-from core.file import file_manager
+from core.file.message_file_parser import MessageFileParser
 from core.memory.token_buffer_memory import TokenBufferMemory
 from core.model_manager import ModelInstance
-from core.model_runtime.entities import (
+from core.model_runtime.entities.llm_entities import LLMUsage
+from core.model_runtime.entities.message_entities import (
     AssistantPromptMessage,
-    LLMUsage,
     PromptMessage,
-    PromptMessageContent,
     PromptMessageTool,
     SystemPromptMessage,
     TextPromptMessageContent,
@@ -41,8 +40,8 @@ from core.tools.entities.tool_entities import (
 from core.tools.tool.dataset_retriever_tool import DatasetRetrieverTool
 from core.tools.tool.tool import Tool
 from core.tools.tool_manager import ToolManager
+from core.tools.utils.tool_parameter_converter import ToolParameterConverter
 from extensions.ext_database import db
-from factories import file_factory
 from models.model import Conversation, Message, MessageAgentThought
 from models.tools import ToolConversationVariables
 
@@ -67,6 +66,23 @@ class BaseAgentRunner(AppRunner):
         db_variables: Optional[ToolConversationVariables] = None,
         model_instance: ModelInstance = None,
     ) -> None:
+        """
+        Agent runner
+        :param tenant_id: tenant id
+        :param application_generate_entity: application generate entity
+        :param conversation: conversation
+        :param app_config: app generate entity
+        :param model_config: model config
+        :param config: dataset config
+        :param queue_manager: queue manager
+        :param message: message
+        :param user_id: user id
+        :param memory: memory
+        :param prompt_messages: prompt messages
+        :param variables_pool: variables pool
+        :param db_variables: db variables
+        :param model_instance: model instance
+        """
         self.tenant_id = tenant_id
         self.application_generate_entity = application_generate_entity
         self.conversation = conversation
@@ -164,7 +180,7 @@ class BaseAgentRunner(AppRunner):
             if parameter.form != ToolParameter.ToolParameterForm.LLM:
                 continue
 
-            parameter_type = parameter.type.as_normal_type()
+            parameter_type = ToolParameterConverter.get_parameter_type(parameter.type)
             enum = []
             if parameter.type == ToolParameter.ToolParameterType.SELECT:
                 enum = [option.value for option in parameter.options]
@@ -249,7 +265,7 @@ class BaseAgentRunner(AppRunner):
             if parameter.form != ToolParameter.ToolParameterForm.LLM:
                 continue
 
-            parameter_type = parameter.type.as_normal_type()
+            parameter_type = ToolParameterConverter.get_parameter_type(parameter.type)
             enum = []
             if parameter.type == ToolParameter.ToolParameterType.SELECT:
                 enum = [option.value for option in parameter.options]
@@ -495,24 +511,26 @@ class BaseAgentRunner(AppRunner):
         return result
 
     def organize_agent_user_prompt(self, message: Message) -> UserPromptMessage:
+        message_file_parser = MessageFileParser(
+            tenant_id=self.tenant_id,
+            app_id=self.app_config.app_id,
+        )
+
         files = message.message_files
         if files:
             file_extra_config = FileUploadConfigManager.convert(message.app_model_config.to_dict())
 
             if file_extra_config:
-                file_objs = file_factory.build_from_message_files(
-                    message_files=files, tenant_id=self.tenant_id, config=file_extra_config
-                )
+                file_objs = message_file_parser.transform_message_files(files, file_extra_config)
             else:
                 file_objs = []
 
             if not file_objs:
                 return UserPromptMessage(content=message.query)
             else:
-                prompt_message_contents: list[PromptMessageContent] = []
-                prompt_message_contents.append(TextPromptMessageContent(data=message.query))
+                prompt_message_contents = [TextPromptMessageContent(data=message.query)]
                 for file_obj in file_objs:
-                    prompt_message_contents.append(file_manager.to_prompt_message_content(file_obj))
+                    prompt_message_contents.append(file_obj.prompt_message_content)
 
                 return UserPromptMessage(content=prompt_message_contents)
         else:

api/core/agent/cot_chat_agent_runner.py

@@ -1,11 +1,9 @@
 import json
 
 from core.agent.cot_agent_runner import CotAgentRunner
-from core.file import file_manager
-from core.model_runtime.entities import (
+from core.model_runtime.entities.message_entities import (
     AssistantPromptMessage,
     PromptMessage,
-    PromptMessageContent,
     SystemPromptMessage,
     TextPromptMessageContent,
     UserPromptMessage,
@@ -34,10 +32,9 @@ class CotChatAgentRunner(CotAgentRunner):
         Organize user query
         """
         if self.files:
-            prompt_message_contents: list[PromptMessageContent] = []
-            prompt_message_contents.append(TextPromptMessageContent(data=query))
+            prompt_message_contents = [TextPromptMessageContent(data=query)]
             for file_obj in self.files:
-                prompt_message_contents.append(file_manager.to_prompt_message_content(file_obj))
+                prompt_message_contents.append(file_obj.prompt_message_content)
 
             prompt_messages.append(UserPromptMessage(content=prompt_message_contents))
         else:

api/core/agent/fc_agent_runner.py

@@ -7,15 +7,10 @@ from typing import Any, Union
 from core.agent.base_agent_runner import BaseAgentRunner
 from core.app.apps.base_app_queue_manager import PublishFrom
 from core.app.entities.queue_entities import QueueAgentThoughtEvent, QueueMessageEndEvent, QueueMessageFileEvent
-from core.file import file_manager
-from core.model_runtime.entities import (
+from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta, LLMUsage
+from core.model_runtime.entities.message_entities import (
     AssistantPromptMessage,
-    LLMResult,
-    LLMResultChunk,
-    LLMResultChunkDelta,
-    LLMUsage,
     PromptMessage,
-    PromptMessageContent,
     PromptMessageContentType,
     SystemPromptMessage,
     TextPromptMessageContent,
@@ -395,10 +390,9 @@ class FunctionCallAgentRunner(BaseAgentRunner):
         Organize user query
         """
         if self.files:
-            prompt_message_contents: list[PromptMessageContent] = []
-            prompt_message_contents.append(TextPromptMessageContent(data=query))
+            prompt_message_contents = [TextPromptMessageContent(data=query)]
             for file_obj in self.files:
-                prompt_message_contents.append(file_manager.to_prompt_message_content(file_obj))
+                prompt_message_contents.append(file_obj.prompt_message_content)
 
             prompt_messages.append(UserPromptMessage(content=prompt_message_contents))
         else:

api/core/app/app_config/easy_ui_based_app/variables/manager.py

@@ -53,11 +53,12 @@ class BasicVariablesConfigManager:
                     VariableEntity(
                         type=variable_type,
                         variable=variable.get("variable"),
-                        description=variable.get("description", ""),
+                        description=variable.get("description"),
                         label=variable.get("label"),
                         required=variable.get("required", False),
                         max_length=variable.get("max_length"),
-                        options=variable.get("options", []),
+                        options=variable.get("options"),
+                        default=variable.get("default"),
                     )
                 )
 

api/core/app/app_config/entities.py

@@ -1,12 +1,11 @@
-from collections.abc import Sequence
 from enum import Enum
 from typing import Any, Optional
 
-from pydantic import BaseModel, Field
+from pydantic import BaseModel
 
-from core.file import FileExtraConfig, FileTransferMethod, FileType
+from core.file.file_obj import FileExtraConfig
 from core.model_runtime.entities.message_entities import PromptMessageRole
-from models.model import AppMode
+from models import AppMode
 
 
 class ModelConfigEntity(BaseModel):
@@ -70,7 +69,7 @@ class PromptTemplateEntity(BaseModel):
         ADVANCED = "advanced"
 
         @classmethod
-        def value_of(cls, value: str):
+        def value_of(cls, value: str) -> "PromptType":
             """
             Get value of given mode.
 
@@ -94,8 +93,6 @@ class VariableEntityType(str, Enum):
     PARAGRAPH = "paragraph"
     NUMBER = "number"
     EXTERNAL_DATA_TOOL = "external_data_tool"
-    FILE = "file"
-    FILE_LIST = "file-list"
 
 
 class VariableEntity(BaseModel):
@@ -105,14 +102,13 @@ class VariableEntity(BaseModel):
 
     variable: str
     label: str
-    description: str = ""
+    description: Optional[str] = None
     type: VariableEntityType
     required: bool = False
     max_length: Optional[int] = None
-    options: Sequence[str] = Field(default_factory=list)
-    allowed_file_types: Sequence[FileType] = Field(default_factory=list)
-    allowed_file_extensions: Sequence[str] = Field(default_factory=list)
-    allowed_file_upload_methods: Sequence[FileTransferMethod] = Field(default_factory=list)
+    options: Optional[list[str]] = None
+    default: Optional[str] = None
+    hint: Optional[str] = None
 
 
 class ExternalDataVariableEntity(BaseModel):
@@ -140,7 +136,7 @@ class DatasetRetrieveConfigEntity(BaseModel):
         MULTIPLE = "multiple"
 
         @classmethod
-        def value_of(cls, value: str):
+        def value_of(cls, value: str) -> "RetrieveStrategy":
             """
             Get value of given mode.
 

api/core/app/app_config/features/file_upload/manager.py

@@ -1,13 +1,12 @@
 from collections.abc import Mapping
-from typing import Any
+from typing import Any, Optional
 
-from core.file.models import FileExtraConfig
-from models import FileUploadConfig
+from core.file.file_obj import FileExtraConfig
 
 
 class FileUploadConfigManager:
     @classmethod
-    def convert(cls, config: Mapping[str, Any], is_vision: bool = True):
+    def convert(cls, config: Mapping[str, Any], is_vision: bool = True) -> Optional[FileExtraConfig]:
         """
         Convert model config to model config
 
@@ -16,18 +15,19 @@ class FileUploadConfigManager:
         """
         file_upload_dict = config.get("file_upload")
         if file_upload_dict:
-            if file_upload_dict.get("enabled"):
-                data = {
-                    "image_config": {
-                        "number_limits": file_upload_dict["number_limits"],
-                        "transfer_methods": file_upload_dict["allowed_file_upload_methods"],
+            if file_upload_dict.get("image"):
+                if "enabled" in file_upload_dict["image"] and file_upload_dict["image"]["enabled"]:
+                    image_config = {
+                        "number_limits": file_upload_dict["image"]["number_limits"],
+                        "transfer_methods": file_upload_dict["image"]["transfer_methods"],
                     }
-                }
 
-                if is_vision:
-                    data["image_config"]["detail"] = file_upload_dict.get("image", {}).get("detail", "low")
+                    if is_vision:
+                        image_config["detail"] = file_upload_dict["image"]["detail"]
 
-                return FileExtraConfig.model_validate(data)
+                    return FileExtraConfig(image_config=image_config)
+
+        return None
 
     @classmethod
     def validate_and_set_defaults(cls, config: dict, is_vision: bool = True) -> tuple[dict, list[str]]:
@@ -39,7 +39,29 @@ class FileUploadConfigManager:
         """
         if not config.get("file_upload"):
             config["file_upload"] = {}
-        else:
-            FileUploadConfig.model_validate(config["file_upload"])
+
+        if not isinstance(config["file_upload"], dict):
+            raise ValueError("file_upload must be of dict type")
+
+        # check image config
+        if not config["file_upload"].get("image"):
+            config["file_upload"]["image"] = {"enabled": False}
+
+        if config["file_upload"]["image"]["enabled"]:
+            number_limits = config["file_upload"]["image"]["number_limits"]
+            if number_limits < 1 or number_limits > 6:
+                raise ValueError("number_limits must be in [1, 6]")
+
+            if is_vision:
+                detail = config["file_upload"]["image"]["detail"]
+                if detail not in {"high", "low"}:
+                    raise ValueError("detail must be in ['high', 'low']")
+
+            transfer_methods = config["file_upload"]["image"]["transfer_methods"]
+            if not isinstance(transfer_methods, list):
+                raise ValueError("transfer_methods must be of list type")
+            for method in transfer_methods:
+                if method not in {"remote_url", "local_file"}:
+                    raise ValueError("transfer_methods must be in ['remote_url', 'local_file']")
 
         return config, ["file_upload"]

api/core/app/app_config/workflow_ui_based_app/variables/manager.py

@@ -17,6 +17,6 @@ class WorkflowVariablesConfigManager:
 
         # variables
         for variable in user_input_form:
-            variables.append(VariableEntity.model_validate(variable))
+            variables.append(VariableEntity(**variable))
 
         return variables

api/core/app/apps/advanced_chat/app_generator.py

@@ -20,11 +20,10 @@ from core.app.apps.message_based_app_generator import MessageBasedAppGenerator
 from core.app.apps.message_based_app_queue_manager import MessageBasedAppQueueManager
 from core.app.entities.app_invoke_entities import AdvancedChatAppGenerateEntity, InvokeFrom
 from core.app.entities.task_entities import ChatbotAppBlockingResponse, ChatbotAppStreamResponse
+from core.file.message_file_parser import MessageFileParser
 from core.model_runtime.errors.invoke import InvokeAuthorizationError, InvokeError
 from core.ops.ops_trace_manager import TraceQueueManager
-from enums import CreatedByRole
 from extensions.ext_database import db
-from factories import file_factory
 from models.account import Account
 from models.model import App, Conversation, EndUser, Message
 from models.workflow import Workflow
@@ -96,16 +95,10 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
 
         # parse files
         files = args["files"] if args.get("files") else []
+        message_file_parser = MessageFileParser(tenant_id=app_model.tenant_id, app_id=app_model.id)
         file_extra_config = FileUploadConfigManager.convert(workflow.features_dict, is_vision=False)
-        role = CreatedByRole.ACCOUNT if isinstance(user, Account) else CreatedByRole.END_USER
         if file_extra_config:
-            file_objs = file_factory.build_from_mappings(
-                mappings=files,
-                tenant_id=app_model.tenant_id,
-                user_id=user.id,
-                role=role,
-                config=file_extra_config,
-            )
+            file_objs = message_file_parser.validate_and_transform_files_arg(files, file_extra_config, user)
         else:
             file_objs = []
 
@@ -113,9 +106,8 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
         app_config = AdvancedChatAppConfigManager.get_app_config(app_model=app_model, workflow=workflow)
 
         # get tracing instance
-        trace_manager = TraceQueueManager(
-            app_id=app_model.id, user_id=user.id if isinstance(user, Account) else user.session_id
-        )
+        user_id = user.id if isinstance(user, Account) else user.session_id
+        trace_manager = TraceQueueManager(app_model.id, user_id)
 
         if invoke_from == InvokeFrom.DEBUGGER:
             # always enable retriever resource in debugger mode
@@ -126,9 +118,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
             task_id=str(uuid.uuid4()),
             app_config=app_config,
             conversation_id=conversation.id if conversation else None,
-            inputs=conversation.inputs
-            if conversation
-            else self._prepare_user_inputs(user_inputs=inputs, app_config=app_config, user_id=user.id, role=role),
+            inputs=conversation.inputs if conversation else self._get_cleaned_inputs(inputs, app_config),
             query=query,
             files=file_objs,
             parent_message_id=args.get("parent_message_id"),

api/core/app/apps/advanced_chat/app_runner.py

@@ -1,26 +1,30 @@
 import logging
+import os
 from collections.abc import Mapping
 from typing import Any, cast
 
 from sqlalchemy import select
 from sqlalchemy.orm import Session
 
-from configs import dify_config
 from core.app.apps.advanced_chat.app_config_manager import AdvancedChatAppConfig
 from core.app.apps.base_app_queue_manager import AppQueueManager
 from core.app.apps.workflow_app_runner import WorkflowBasedAppRunner
-from core.app.entities.app_invoke_entities import AdvancedChatAppGenerateEntity, InvokeFrom
+from core.app.apps.workflow_logging_callback import WorkflowLoggingCallback
+from core.app.entities.app_invoke_entities import (
+    AdvancedChatAppGenerateEntity,
+    InvokeFrom,
+)
 from core.app.entities.queue_entities import (
     QueueAnnotationReplyEvent,
     QueueStopEvent,
     QueueTextChunkEvent,
 )
 from core.moderation.base import ModerationError
-from core.workflow.callbacks import WorkflowCallback, WorkflowLoggingCallback
+from core.workflow.callbacks.base_workflow_callback import WorkflowCallback
+from core.workflow.entities.node_entities import UserFrom
 from core.workflow.entities.variable_pool import VariablePool
 from core.workflow.enums import SystemVariableKey
 from core.workflow.workflow_entry import WorkflowEntry
-from enums import UserFrom
 from extensions.ext_database import db
 from models.model import App, Conversation, EndUser, Message
 from models.workflow import ConversationVariable, WorkflowType
@@ -40,6 +44,12 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
         conversation: Conversation,
         message: Message,
     ) -> None:
+        """
+        :param application_generate_entity: application generate entity
+        :param queue_manager: application queue manager
+        :param conversation: conversation
+        :param message: message
+        """
         super().__init__(queue_manager)
 
         self.application_generate_entity = application_generate_entity
@@ -47,6 +57,10 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
         self.message = message
 
     def run(self) -> None:
+        """
+        Run application
+        :return:
+        """
         app_config = self.application_generate_entity.app_config
         app_config = cast(AdvancedChatAppConfig, app_config)
 
@@ -67,7 +81,7 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
             user_id = self.application_generate_entity.user_id
 
         workflow_callbacks: list[WorkflowCallback] = []
-        if dify_config.DEBUG:
+        if bool(os.environ.get("DEBUG", "False").lower() == "true"):
             workflow_callbacks.append(WorkflowLoggingCallback())
 
         if self.application_generate_entity.single_iteration_run:
@@ -184,6 +198,15 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
         query: str,
         message_id: str,
     ) -> bool:
+        """
+        Handle input moderation
+        :param app_record: app record
+        :param app_generate_entity: application generate entity
+        :param inputs: inputs
+        :param query: query
+        :param message_id: message id
+        :return:
+        """
         try:
             # process sensitive_word_avoidance
             _, inputs, query = self.moderation_for_inputs(
@@ -203,6 +226,14 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
     def handle_annotation_reply(
         self, app_record: App, message: Message, query: str, app_generate_entity: AdvancedChatAppGenerateEntity
     ) -> bool:
+        """
+        Handle annotation reply
+        :param app_record: app record
+        :param message: message
+        :param query: query
+        :param app_generate_entity: application generate entity
+        """
+        # annotation reply
         annotation_reply = self.query_app_annotations_to_reply(
             app_record=app_record,
             message=message,
@@ -224,6 +255,8 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
     def _complete_with_stream_output(self, text: str, stopped_by: QueueStopEvent.StopBy) -> None:
         """
         Direct output
+        :param text: text
+        :return:
         """
         self._publish_event(QueueTextChunkEvent(text=text))
 

api/core/app/apps/advanced_chat/generate_task_pipeline.py

@@ -1,7 +1,7 @@
 import json
 import logging
 import time
-from collections.abc import Generator, Mapping
+from collections.abc import Generator
 from typing import Any, Optional, Union
 
 from constants.tts_auto_play_timeout import TTS_AUTO_PLAY_TIMEOUT, TTS_AUTO_PLAY_YIELD_CPU_TIME
@@ -49,7 +49,6 @@ from core.model_runtime.utils.encoders import jsonable_encoder
 from core.ops.ops_trace_manager import TraceQueueManager
 from core.workflow.enums import SystemVariableKey
 from core.workflow.graph_engine.entities.graph_runtime_state import GraphRuntimeState
-from enums.workflow_nodes import NodeType
 from events.message_event import message_was_created
 from extensions.ext_database import db
 from models.account import Account
@@ -113,7 +112,6 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
         self._task_state = WorkflowTaskState()
 
         self._conversation_name_generate_thread = None
-        self._recorded_files: list[Mapping[str, Any]] = []
 
     def process(self):
         """
@@ -233,8 +231,7 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
             except Exception as e:
                 logger.error(e)
                 break
-        if tts_publisher:
-            yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
+        yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
 
     def _process_stream_response(
         self,
@@ -292,10 +289,6 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
             elif isinstance(event, QueueNodeSucceededEvent):
                 workflow_node_execution = self._handle_workflow_node_execution_success(event)
 
-                # Record files if it's an answer node or end node
-                if event.node_type in [NodeType.ANSWER, NodeType.END]:
-                    self._recorded_files.extend(self._fetch_files_from_node_outputs(event.outputs or {}))
-
                 response = self._workflow_node_finish_to_stream_response(
                     event=event,
                     task_id=self._application_generate_entity.task_id,
@@ -362,7 +355,7 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
                     start_at=graph_runtime_state.start_at,
                     total_tokens=graph_runtime_state.total_tokens,
                     total_steps=graph_runtime_state.node_run_steps,
-                    outputs=event.outputs,
+                    outputs=json.dumps(event.outputs) if event.outputs else None,
                     conversation_id=self._conversation.id,
                     trace_manager=trace_manager,
                 )
@@ -534,7 +527,7 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
                 del extras["metadata"]["annotation_reply"]
 
         return MessageEndStreamResponse(
-            task_id=self._application_generate_entity.task_id, id=self._message.id, files=self._recorded_files, **extras
+            task_id=self._application_generate_entity.task_id, id=self._message.id, **extras
         )
 
     def _handle_output_moderation_chunk(self, text: str) -> bool:

api/core/app/apps/agent_chat/app_generator.py

@@ -17,12 +17,12 @@ from core.app.apps.base_app_queue_manager import AppQueueManager, GenerateTaskSt
 from core.app.apps.message_based_app_generator import MessageBasedAppGenerator
 from core.app.apps.message_based_app_queue_manager import MessageBasedAppQueueManager
 from core.app.entities.app_invoke_entities import AgentChatAppGenerateEntity, InvokeFrom
+from core.file.message_file_parser import MessageFileParser
 from core.model_runtime.errors.invoke import InvokeAuthorizationError, InvokeError
 from core.ops.ops_trace_manager import TraceQueueManager
-from enums import CreatedByRole
 from extensions.ext_database import db
-from factories import file_factory
-from models import Account, App, EndUser
+from models.account import Account
+from models.model import App, EndUser
 
 logger = logging.getLogger(__name__)
 
@@ -49,12 +49,7 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
     ) -> dict: ...
 
     def generate(
-        self,
-        app_model: App,
-        user: Union[Account, EndUser],
-        args: Any,
-        invoke_from: InvokeFrom,
-        stream: bool = True,
+        self, app_model: App, user: Union[Account, EndUser], args: Any, invoke_from: InvokeFrom, stream: bool = True
     ) -> Union[dict, Generator[dict, None, None]]:
         """
         Generate App response.
@@ -102,19 +97,12 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
             # always enable retriever resource in debugger mode
             override_model_config_dict["retriever_resource"] = {"enabled": True}
 
-        role = CreatedByRole.ACCOUNT if isinstance(user, Account) else CreatedByRole.END_USER
-
         # parse files
-        files = args.get("files") or []
+        files = args["files"] if args.get("files") else []
+        message_file_parser = MessageFileParser(tenant_id=app_model.tenant_id, app_id=app_model.id)
         file_extra_config = FileUploadConfigManager.convert(override_model_config_dict or app_model_config.to_dict())
         if file_extra_config:
-            file_objs = file_factory.build_from_mappings(
-                mappings=files,
-                tenant_id=app_model.tenant_id,
-                user_id=user.id,
-                role=role,
-                config=file_extra_config,
-            )
+            file_objs = message_file_parser.validate_and_transform_files_arg(files, file_extra_config, user)
         else:
             file_objs = []
 
@@ -127,7 +115,8 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
         )
 
         # get tracing instance
-        trace_manager = TraceQueueManager(app_model.id, user.id if isinstance(user, Account) else user.session_id)
+        user_id = user.id if isinstance(user, Account) else user.session_id
+        trace_manager = TraceQueueManager(app_model.id, user_id)
 
         # init application generate entity
         application_generate_entity = AgentChatAppGenerateEntity(
@@ -135,9 +124,7 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
             app_config=app_config,
             model_conf=ModelConfigConverter.convert(app_config),
             conversation_id=conversation.id if conversation else None,
-            inputs=conversation.inputs
-            if conversation
-            else self._prepare_user_inputs(user_inputs=inputs, app_config=app_config, user_id=user.id, role=role),
+            inputs=conversation.inputs if conversation else self._get_cleaned_inputs(inputs, app_config),
             query=query,
             files=file_objs,
             parent_message_id=args.get("parent_message_id"),

api/core/app/apps/base_app_generator.py

@@ -1,92 +1,35 @@
 from collections.abc import Mapping
-from typing import TYPE_CHECKING, Any, Optional
+from typing import Any, Optional
 
-from core.app.app_config.entities import VariableEntityType
-from core.file import File, FileExtraConfig
-from factories import file_factory
-
-if TYPE_CHECKING:
-    from core.app.app_config.entities import AppConfig, VariableEntity
-    from enums import CreatedByRole
+from core.app.app_config.entities import AppConfig, VariableEntity, VariableEntityType
 
 
 class BaseAppGenerator:
-    def _prepare_user_inputs(
-        self,
-        *,
-        user_inputs: Optional[Mapping[str, Any]],
-        app_config: "AppConfig",
-        user_id: str,
-        role: "CreatedByRole",
-    ) -> Mapping[str, Any]:
+    def _get_cleaned_inputs(self, user_inputs: Optional[Mapping[str, Any]], app_config: AppConfig) -> Mapping[str, Any]:
         user_inputs = user_inputs or {}
         # Filter input variables from form configuration, handle required fields, default values, and option values
         variables = app_config.variables
-        user_inputs = {var.variable: self._validate_input(inputs=user_inputs, var=var) for var in variables}
-        user_inputs = {k: self._sanitize_value(v) for k, v in user_inputs.items()}
-        # Convert files in inputs to File
-        entity_dictionary = {item.variable: item for item in app_config.variables}
-        # Convert single file to File
-        files_inputs = {
-            k: file_factory.build_from_mapping(
-                mapping=v,
-                tenant_id=app_config.tenant_id,
-                user_id=user_id,
-                role=role,
-                config=FileExtraConfig(
-                    allowed_file_types=entity_dictionary[k].allowed_file_types,
-                    allowed_extensions=entity_dictionary[k].allowed_file_extensions,
-                    allowed_upload_methods=entity_dictionary[k].allowed_file_upload_methods,
-                ),
-            )
-            for k, v in user_inputs.items()
-            if isinstance(v, dict) and entity_dictionary[k].type == VariableEntityType.FILE
-        }
-        # Convert list of files to File
-        file_list_inputs = {
-            k: file_factory.build_from_mappings(
-                mappings=v,
-                tenant_id=app_config.tenant_id,
-                user_id=user_id,
-                role=role,
-                config=FileExtraConfig(
-                    allowed_file_types=entity_dictionary[k].allowed_file_types,
-                    allowed_extensions=entity_dictionary[k].allowed_file_extensions,
-                    allowed_upload_methods=entity_dictionary[k].allowed_file_upload_methods,
-                ),
-            )
-            for k, v in user_inputs.items()
-            if isinstance(v, list)
-            # Ensure skip List<File>
-            and all(isinstance(item, dict) for item in v)
-            and entity_dictionary[k].type == VariableEntityType.FILE_LIST
-        }
-        # Merge all inputs
-        user_inputs = {**user_inputs, **files_inputs, **file_list_inputs}
+        filtered_inputs = {var.variable: self._validate_input(inputs=user_inputs, var=var) for var in variables}
+        filtered_inputs = {k: self._sanitize_value(v) for k, v in filtered_inputs.items()}
+        return filtered_inputs
 
-        # Check if all files are converted to File
-        if any(filter(lambda v: isinstance(v, dict), user_inputs.values())):
-            raise ValueError("Invalid input type")
-        if any(
-            filter(lambda v: isinstance(v, dict), filter(lambda item: isinstance(item, list), user_inputs.values()))
-        ):
-            raise ValueError("Invalid input type")
-
-        return user_inputs
-
-    def _validate_input(self, *, inputs: Mapping[str, Any], var: "VariableEntity"):
+    def _validate_input(self, *, inputs: Mapping[str, Any], var: VariableEntity):
         user_input_value = inputs.get(var.variable)
-        if not user_input_value:
-            if var.required:
-                raise ValueError(f"{var.variable} is required in input form")
-            else:
-                return None
-
-        if var.type in {
-            VariableEntityType.TEXT_INPUT,
-            VariableEntityType.SELECT,
-            VariableEntityType.PARAGRAPH,
-        } and not isinstance(user_input_value, str):
+        if var.required and not user_input_value:
+            raise ValueError(f"{var.variable} is required in input form")
+        if not var.required and not user_input_value:
+            # TODO: should we return None here if the default value is None?
+            return var.default or ""
+        if (
+            var.type
+            in {
+                VariableEntityType.TEXT_INPUT,
+                VariableEntityType.SELECT,
+                VariableEntityType.PARAGRAPH,
+            }
+            and user_input_value
+            and not isinstance(user_input_value, str)
+        ):
             raise ValueError(f"(type '{var.type}') {var.variable} in input form must be a string")
         if var.type == VariableEntityType.NUMBER and isinstance(user_input_value, str):
             # may raise ValueError if user_input_value is not a valid number
@@ -98,24 +41,12 @@ class BaseAppGenerator:
             except ValueError:
                 raise ValueError(f"{var.variable} in input form must be a valid number")
         if var.type == VariableEntityType.SELECT:
-            options = var.options
+            options = var.options or []
             if user_input_value not in options:
                 raise ValueError(f"{var.variable} in input form must be one of the following: {options}")
         elif var.type in {VariableEntityType.TEXT_INPUT, VariableEntityType.PARAGRAPH}:
-            if var.max_length and len(user_input_value) > var.max_length:
+            if var.max_length and user_input_value and len(user_input_value) > var.max_length:
                 raise ValueError(f"{var.variable} in input form must be less than {var.max_length} characters")
-        elif var.type == VariableEntityType.FILE:
-            if not isinstance(user_input_value, dict) and not isinstance(user_input_value, File):
-                raise ValueError(f"{var.variable} in input form must be a file")
-        elif var.type == VariableEntityType.FILE_LIST:
-            if not (
-                isinstance(user_input_value, list)
-                and (
-                    all(isinstance(item, dict) for item in user_input_value)
-                    or all(isinstance(item, File) for item in user_input_value)
-                )
-            ):
-                raise ValueError(f"{var.variable} in input form must be a list of files")
 
         return user_input_value
 

api/core/app/apps/base_app_runner.py

@@ -27,7 +27,7 @@ from core.prompt.simple_prompt_transform import ModelMode, SimplePromptTransform
 from models.model import App, AppMode, Message, MessageAnnotation
 
 if TYPE_CHECKING:
-    from core.file.models import File
+    from core.file.file_obj import FileVar
 
 
 class AppRunner:
@@ -37,7 +37,7 @@ class AppRunner:
         model_config: ModelConfigWithCredentialsEntity,
         prompt_template_entity: PromptTemplateEntity,
         inputs: dict[str, str],
-        files: list["File"],
+        files: list["FileVar"],
         query: Optional[str] = None,
     ) -> int:
         """
@@ -137,7 +137,7 @@ class AppRunner:
         model_config: ModelConfigWithCredentialsEntity,
         prompt_template_entity: PromptTemplateEntity,
         inputs: dict[str, str],
-        files: list["File"],
+        files: list["FileVar"],
         query: Optional[str] = None,
         context: Optional[str] = None,
         memory: Optional[TokenBufferMemory] = None,

api/core/app/apps/chat/app_generator.py

@@ -17,11 +17,10 @@ from core.app.apps.chat.generate_response_converter import ChatAppGenerateRespon
 from core.app.apps.message_based_app_generator import MessageBasedAppGenerator
 from core.app.apps.message_based_app_queue_manager import MessageBasedAppQueueManager
 from core.app.entities.app_invoke_entities import ChatAppGenerateEntity, InvokeFrom
+from core.file.message_file_parser import MessageFileParser
 from core.model_runtime.errors.invoke import InvokeAuthorizationError, InvokeError
 from core.ops.ops_trace_manager import TraceQueueManager
-from enums import CreatedByRole
 from extensions.ext_database import db
-from factories import file_factory
 from models.account import Account
 from models.model import App, EndUser
 
@@ -100,19 +99,12 @@ class ChatAppGenerator(MessageBasedAppGenerator):
             # always enable retriever resource in debugger mode
             override_model_config_dict["retriever_resource"] = {"enabled": True}
 
-        role = CreatedByRole.ACCOUNT if isinstance(user, Account) else CreatedByRole.END_USER
-
         # parse files
         files = args["files"] if args.get("files") else []
+        message_file_parser = MessageFileParser(tenant_id=app_model.tenant_id, app_id=app_model.id)
         file_extra_config = FileUploadConfigManager.convert(override_model_config_dict or app_model_config.to_dict())
         if file_extra_config:
-            file_objs = file_factory.build_from_mappings(
-                mappings=files,
-                tenant_id=app_model.tenant_id,
-                user_id=user.id,
-                role=role,
-                config=file_extra_config,
-            )
+            file_objs = message_file_parser.validate_and_transform_files_arg(files, file_extra_config, user)
         else:
             file_objs = []
 
@@ -125,7 +117,7 @@ class ChatAppGenerator(MessageBasedAppGenerator):
         )
 
         # get tracing instance
-        trace_manager = TraceQueueManager(app_id=app_model.id)
+        trace_manager = TraceQueueManager(app_model.id)
 
         # init application generate entity
         application_generate_entity = ChatAppGenerateEntity(
@@ -133,17 +125,15 @@ class ChatAppGenerator(MessageBasedAppGenerator):
             app_config=app_config,
             model_conf=ModelConfigConverter.convert(app_config),
             conversation_id=conversation.id if conversation else None,
-            inputs=conversation.inputs
-            if conversation
-            else self._prepare_user_inputs(user_inputs=inputs, app_config=app_config, user_id=user.id, role=role),
+            inputs=conversation.inputs if conversation else self._get_cleaned_inputs(inputs, app_config),
             query=query,
             files=file_objs,
             parent_message_id=args.get("parent_message_id"),
             user_id=user.id,
+            stream=stream,
             invoke_from=invoke_from,
             extras=extras,
             trace_manager=trace_manager,
-            stream=stream,
         )
 
         # init generate records

api/core/app/apps/completion/app_generator.py

@@ -17,12 +17,12 @@ from core.app.apps.completion.generate_response_converter import CompletionAppGe
 from core.app.apps.message_based_app_generator import MessageBasedAppGenerator
 from core.app.apps.message_based_app_queue_manager import MessageBasedAppQueueManager
 from core.app.entities.app_invoke_entities import CompletionAppGenerateEntity, InvokeFrom
+from core.file.message_file_parser import MessageFileParser
 from core.model_runtime.errors.invoke import InvokeAuthorizationError, InvokeError
 from core.ops.ops_trace_manager import TraceQueueManager
-from enums import CreatedByRole
 from extensions.ext_database import db
-from factories import file_factory
-from models import Account, App, EndUser, Message
+from models.account import Account
+from models.model import App, EndUser, Message
 from services.errors.app import MoreLikeThisDisabledError
 from services.errors.message import MessageNotExistsError
 
@@ -88,19 +88,12 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
                 tenant_id=app_model.tenant_id, config=args.get("model_config")
             )
 
-        role = CreatedByRole.ACCOUNT if isinstance(user, Account) else CreatedByRole.END_USER
-
         # parse files
         files = args["files"] if args.get("files") else []
+        message_file_parser = MessageFileParser(tenant_id=app_model.tenant_id, app_id=app_model.id)
         file_extra_config = FileUploadConfigManager.convert(override_model_config_dict or app_model_config.to_dict())
         if file_extra_config:
-            file_objs = file_factory.build_from_mappings(
-                mappings=files,
-                tenant_id=app_model.tenant_id,
-                user_id=user.id,
-                role=role,
-                config=file_extra_config,
-            )
+            file_objs = message_file_parser.validate_and_transform_files_arg(files, file_extra_config, user)
         else:
             file_objs = []
 
@@ -110,7 +103,6 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
         )
 
         # get tracing instance
-        user_id = user.id if isinstance(user, Account) else user.session_id
         trace_manager = TraceQueueManager(app_model.id)
 
         # init application generate entity
@@ -118,7 +110,7 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
             task_id=str(uuid.uuid4()),
             app_config=app_config,
             model_conf=ModelConfigConverter.convert(app_config),
-            inputs=self._prepare_user_inputs(user_inputs=inputs, app_config=app_config, user_id=user.id, role=role),
+            inputs=self._get_cleaned_inputs(inputs, app_config),
             query=query,
             files=file_objs,
             user_id=user.id,
@@ -259,16 +251,10 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
         override_model_config_dict["model"] = model_dict
 
         # parse files
-        role = CreatedByRole.ACCOUNT if isinstance(user, Account) else CreatedByRole.END_USER
-        file_extra_config = FileUploadConfigManager.convert(override_model_config_dict)
+        message_file_parser = MessageFileParser(tenant_id=app_model.tenant_id, app_id=app_model.id)
+        file_extra_config = FileUploadConfigManager.convert(override_model_config_dict or app_model_config.to_dict())
         if file_extra_config:
-            file_objs = file_factory.build_from_mappings(
-                mappings=message.files,
-                tenant_id=app_model.tenant_id,
-                user_id=user.id,
-                role=role,
-                config=file_extra_config,
-            )
+            file_objs = message_file_parser.validate_and_transform_files_arg(message.files, file_extra_config, user)
         else:
             file_objs = []
 

api/core/app/apps/message_based_app_generator.py

@@ -26,7 +26,7 @@ from core.app.entities.task_entities import (
 from core.app.task_pipeline.easy_ui_based_generate_task_pipeline import EasyUIBasedGenerateTaskPipeline
 from core.prompt.utils.prompt_template_parser import PromptTemplateParser
 from extensions.ext_database import db
-from models import Account
+from models.account import Account
 from models.model import App, AppMode, AppModelConfig, Conversation, EndUser, Message, MessageFile
 from services.errors.app_model_config import AppModelConfigBrokenError
 from services.errors.conversation import ConversationCompletedError, ConversationNotExistsError
@@ -238,7 +238,7 @@ class MessageBasedAppGenerator(BaseAppGenerator):
                 type=file.type.value,
                 transfer_method=file.transfer_method.value,
                 belongs_to="user",
-                url=file.remote_url,
+                url=file.url,
                 upload_file_id=file.related_id,
                 created_by_role=("account" if account_id else "end_user"),
                 created_by=account_id or end_user_id,

api/core/app/apps/workflow/app_generator.py

@@ -3,7 +3,7 @@ import logging
 import os
 import threading
 import uuid
-from collections.abc import Generator, Mapping, Sequence
+from collections.abc import Generator
 from typing import Any, Literal, Optional, Union, overload
 
 from flask import Flask, current_app
@@ -20,12 +20,13 @@ from core.app.apps.workflow.generate_response_converter import WorkflowAppGenera
 from core.app.apps.workflow.generate_task_pipeline import WorkflowAppGenerateTaskPipeline
 from core.app.entities.app_invoke_entities import InvokeFrom, WorkflowAppGenerateEntity
 from core.app.entities.task_entities import WorkflowAppBlockingResponse, WorkflowAppStreamResponse
+from core.file.message_file_parser import MessageFileParser
 from core.model_runtime.errors.invoke import InvokeAuthorizationError, InvokeError
 from core.ops.ops_trace_manager import TraceQueueManager
-from enums import CreatedByRole
 from extensions.ext_database import db
-from factories import file_factory
-from models import Account, App, EndUser, Workflow
+from models.account import Account
+from models.model import App, EndUser
+from models.workflow import Workflow
 
 logger = logging.getLogger(__name__)
 
@@ -62,45 +63,48 @@ class WorkflowAppGenerator(BaseAppGenerator):
         app_model: App,
         workflow: Workflow,
         user: Union[Account, EndUser],
-        args: Mapping[str, Any],
+        args: dict,
         invoke_from: InvokeFrom,
         stream: bool = True,
         call_depth: int = 0,
         workflow_thread_pool_id: Optional[str] = None,
     ):
-        files: Sequence[Mapping[str, Any]] = args.get("files") or []
+        """
+        Generate App response.
 
-        role = CreatedByRole.ACCOUNT if isinstance(user, Account) else CreatedByRole.END_USER
+        :param app_model: App
+        :param workflow: Workflow
+        :param user: account or end user
+        :param args: request args
+        :param invoke_from: invoke from source
+        :param stream: is stream
+        :param call_depth: call depth
+        :param workflow_thread_pool_id: workflow thread pool id
+        """
+        inputs = args["inputs"]
 
         # parse files
+        files = args["files"] if args.get("files") else []
+        message_file_parser = MessageFileParser(tenant_id=app_model.tenant_id, app_id=app_model.id)
         file_extra_config = FileUploadConfigManager.convert(workflow.features_dict, is_vision=False)
-        system_files = file_factory.build_from_mappings(
-            mappings=files,
-            tenant_id=app_model.tenant_id,
-            user_id=user.id,
-            role=role,
-            config=file_extra_config,
-        )
+        if file_extra_config:
+            file_objs = message_file_parser.validate_and_transform_files_arg(files, file_extra_config, user)
+        else:
+            file_objs = []
 
         # convert to app config
-        app_config = WorkflowAppConfigManager.get_app_config(
-            app_model=app_model,
-            workflow=workflow,
-        )
+        app_config = WorkflowAppConfigManager.get_app_config(app_model=app_model, workflow=workflow)
 
         # get tracing instance
-        trace_manager = TraceQueueManager(
-            app_id=app_model.id,
-            user_id=user.id if isinstance(user, Account) else user.session_id,
-        )
+        user_id = user.id if isinstance(user, Account) else user.session_id
+        trace_manager = TraceQueueManager(app_model.id, user_id)
 
-        inputs: Mapping[str, Any] = args["inputs"]
         # init application generate entity
         application_generate_entity = WorkflowAppGenerateEntity(
             task_id=str(uuid.uuid4()),
             app_config=app_config,
-            inputs=self._prepare_user_inputs(user_inputs=inputs, app_config=app_config, user_id=user.id, role=role),
-            files=system_files,
+            inputs=self._get_cleaned_inputs(inputs, app_config),
+            files=file_objs,
             user_id=user.id,
             stream=stream,
             invoke_from=invoke_from,

api/core/app/apps/workflow/app_runner.py

@@ -1,19 +1,20 @@
 import logging
+import os
 from typing import Optional, cast
 
-from configs import dify_config
 from core.app.apps.base_app_queue_manager import AppQueueManager
 from core.app.apps.workflow.app_config_manager import WorkflowAppConfig
 from core.app.apps.workflow_app_runner import WorkflowBasedAppRunner
+from core.app.apps.workflow_logging_callback import WorkflowLoggingCallback
 from core.app.entities.app_invoke_entities import (
     InvokeFrom,
     WorkflowAppGenerateEntity,
 )
-from core.workflow.callbacks import WorkflowCallback, WorkflowLoggingCallback
+from core.workflow.callbacks.base_workflow_callback import WorkflowCallback
+from core.workflow.entities.node_entities import UserFrom
 from core.workflow.entities.variable_pool import VariablePool
 from core.workflow.enums import SystemVariableKey
 from core.workflow.workflow_entry import WorkflowEntry
-from enums import UserFrom
 from extensions.ext_database import db
 from models.model import App, EndUser
 from models.workflow import WorkflowType
@@ -70,7 +71,7 @@ class WorkflowAppRunner(WorkflowBasedAppRunner):
         db.session.close()
 
         workflow_callbacks: list[WorkflowCallback] = []
-        if dify_config.DEBUG:
+        if bool(os.environ.get("DEBUG", "False").lower() == "true"):
             workflow_callbacks.append(WorkflowLoggingCallback())
 
         # if only single iteration run is requested

api/core/app/apps/workflow/generate_task_pipeline.py

@@ -1,3 +1,4 @@
+import json
 import logging
 import time
 from collections.abc import Generator
@@ -211,8 +212,7 @@ class WorkflowAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCycleMa
             except Exception as e:
                 logger.error(e)
                 break
-        if tts_publisher:
-            yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
+        yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
 
     def _process_stream_response(
         self,
@@ -327,7 +327,9 @@ class WorkflowAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCycleMa
                     start_at=graph_runtime_state.start_at,
                     total_tokens=graph_runtime_state.total_tokens,
                     total_steps=graph_runtime_state.node_run_steps,
-                    outputs=event.outputs,
+                    outputs=json.dumps(event.outputs)
+                    if isinstance(event, QueueWorkflowSucceededEvent) and event.outputs
+                    else None,
                     conversation_id=None,
                     trace_manager=trace_manager,
                 )

api/core/app/apps/workflow_app_runner.py

@@ -20,6 +20,7 @@ from core.app.entities.queue_entities import (
     QueueWorkflowStartedEvent,
     QueueWorkflowSucceededEvent,
 )
+from core.workflow.entities.node_entities import NodeType
 from core.workflow.entities.variable_pool import VariablePool
 from core.workflow.graph_engine.entities.event import (
     GraphEngineEvent,
@@ -44,7 +45,6 @@ from core.workflow.nodes.base_node import BaseNode
 from core.workflow.nodes.iteration.entities import IterationNodeData
 from core.workflow.nodes.node_mapping import node_classes
 from core.workflow.workflow_entry import WorkflowEntry
-from enums import NodeType
 from extensions.ext_database import db
 from models.model import App
 from models.workflow import Workflow

api/core/app/apps/workflow_logging_callback.py

@@ -1,6 +1,7 @@
 from typing import Optional
 
 from core.model_runtime.utils.encoders import jsonable_encoder
+from core.workflow.callbacks.base_workflow_callback import WorkflowCallback
 from core.workflow.graph_engine.entities.event import (
     GraphEngineEvent,
     GraphRunFailedEvent,
@@ -19,8 +20,6 @@ from core.workflow.graph_engine.entities.event import (
     ParallelBranchRunSucceededEvent,
 )
 
-from .base_workflow_callback import WorkflowCallback
-
 _TEXT_COLOR_MAPPING = {
     "blue": "36;1",
     "yellow": "33;1",

api/core/app/entities/app_invoke_entities.py

@@ -1,4 +1,4 @@
-from collections.abc import Mapping, Sequence
+from collections.abc import Mapping
 from enum import Enum
 from typing import Any, Optional
 
@@ -6,7 +6,7 @@ from pydantic import BaseModel, ConfigDict
 
 from core.app.app_config.entities import AppConfig, EasyUIBasedAppConfig, WorkflowUIBasedAppConfig
 from core.entities.provider_configuration import ProviderModelBundle
-from core.file.models import File
+from core.file.file_obj import FileVar
 from core.model_runtime.entities.model_entities import AIModelEntity
 from core.ops.ops_trace_manager import TraceQueueManager
 
@@ -22,7 +22,7 @@ class InvokeFrom(Enum):
     DEBUGGER = "debugger"
 
     @classmethod
-    def value_of(cls, value: str):
+    def value_of(cls, value: str) -> "InvokeFrom":
         """
         Get value of given mode.
 
@@ -81,7 +81,7 @@ class AppGenerateEntity(BaseModel):
     app_config: AppConfig
 
     inputs: Mapping[str, Any]
-    files: Sequence[File]
+    files: list[FileVar] = []
     user_id: str
 
     # extras

api/core/app/entities/queue_entities.py

@@ -6,9 +6,8 @@ from pydantic import BaseModel, field_validator
 
 from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk
 from core.workflow.entities.base_node_data_entities import BaseNodeData
-from core.workflow.entities.node_entities import NodeRunMetadataKey
+from core.workflow.entities.node_entities import NodeRunMetadataKey, NodeType
 from core.workflow.graph_engine.entities.graph_runtime_state import GraphRuntimeState
-from enums import NodeType
 
 
 class QueueEvent(str, Enum):

api/core/app/entities/task_entities.py

@@ -1,4 +1,3 @@
-from collections.abc import Mapping
 from enum import Enum
 from typing import Any, Optional
 
@@ -120,7 +119,6 @@ class MessageEndStreamResponse(StreamResponse):
     event: StreamEvent = StreamEvent.MESSAGE_END
     id: str
     metadata: dict = {}
-    files: Optional[list[Mapping[str, Any]]] = None
 
 
 class MessageFileStreamResponse(StreamResponse):

api/core/app/segments/__init__.py

@@ -1,12 +1,7 @@
 from .segment_group import SegmentGroup
 from .segments import (
     ArrayAnySegment,
-    ArrayFileSegment,
-    ArrayNumberSegment,
-    ArrayObjectSegment,
     ArraySegment,
-    ArrayStringSegment,
-    FileSegment,
     FloatSegment,
     IntegerSegment,
     NoneSegment,
@@ -20,7 +15,6 @@ from .variables import (
     ArrayNumberVariable,
     ArrayObjectVariable,
     ArrayStringVariable,
-    FileVariable,
     FloatVariable,
     IntegerVariable,
     NoneVariable,
@@ -52,10 +46,4 @@ __all__ = [
     "ArrayNumberVariable",
     "ArrayObjectVariable",
     "ArraySegment",
-    "ArrayFileSegment",
-    "ArrayNumberSegment",
-    "ArrayObjectSegment",
-    "ArrayStringSegment",
-    "FileSegment",
-    "FileVariable",
 ]

api/core/app/segments/exc.py

@@ -0,0 +1,2 @@
+class VariableError(Exception):
+    pass

api/core/app/segments/factory.py

@@ -2,32 +2,29 @@ from collections.abc import Mapping
 from typing import Any
 
 from configs import dify_config
-from core.file import File
-from core.variables import (
+
+from .exc import VariableError
+from .segments import (
     ArrayAnySegment,
-    ArrayFileSegment,
-    ArrayNumberSegment,
-    ArrayNumberVariable,
-    ArrayObjectSegment,
-    ArrayObjectVariable,
-    ArrayStringSegment,
-    ArrayStringVariable,
-    FileSegment,
     FloatSegment,
-    FloatVariable,
     IntegerSegment,
-    IntegerVariable,
     NoneSegment,
     ObjectSegment,
+    Segment,
+    StringSegment,
+)
+from .types import SegmentType
+from .variables import (
+    ArrayNumberVariable,
+    ArrayObjectVariable,
+    ArrayStringVariable,
+    FloatVariable,
+    IntegerVariable,
     ObjectVariable,
     SecretVariable,
-    Segment,
-    SegmentType,
-    StringSegment,
     StringVariable,
     Variable,
 )
-from core.variables.exc import VariableError
 
 
 def build_variable_from_mapping(mapping: Mapping[str, Any], /) -> Variable:
@@ -74,22 +71,6 @@ def build_segment(value: Any, /) -> Segment:
         return FloatSegment(value=value)
     if isinstance(value, dict):
         return ObjectSegment(value=value)
-    if isinstance(value, File):
-        return FileSegment(value=value)
     if isinstance(value, list):
-        items = [build_segment(item) for item in value]
-        types = {item.value_type for item in items}
-        if len(types) != 1:
-            return ArrayAnySegment(value=value)
-        match types.pop():
-            case SegmentType.STRING:
-                return ArrayStringSegment(value=value)
-            case SegmentType.NUMBER:
-                return ArrayNumberSegment(value=value)
-            case SegmentType.OBJECT:
-                return ArrayObjectSegment(value=value)
-            case SegmentType.FILE:
-                return ArrayFileSegment(value=value)
-            case _:
-                raise ValueError(f"not supported value {value}")
+        return ArrayAnySegment(value=value)
     raise ValueError(f"not supported value {value}")

api/core/app/segments/parser.py

@@ -0,0 +1,18 @@
+import re
+
+from core.workflow.entities.variable_pool import VariablePool
+
+from . import SegmentGroup, factory
+
+VARIABLE_PATTERN = re.compile(r"\{\{#([a-zA-Z0-9_]{1,50}(?:\.[a-zA-Z_][a-zA-Z0-9_]{0,29}){1,10})#\}\}")
+
+
+def convert_template(*, template: str, variable_pool: VariablePool):
+    parts = re.split(VARIABLE_PATTERN, template)
+    segments = []
+    for part in filter(lambda x: x, parts):
+        if "." in part and (value := variable_pool.get(part.split("."))):
+            segments.append(value)
+        else:
+            segments.append(factory.build_segment(part))
+    return SegmentGroup(value=segments)

api/core/app/segments/segments.py

@@ -5,8 +5,6 @@ from typing import Any
 
 from pydantic import BaseModel, ConfigDict, field_validator
 
-from core.file import File
-
 from .types import SegmentType
 
 
@@ -41,9 +39,6 @@ class Segment(BaseModel):
 
     @property
     def size(self) -> int:
-        """
-        Return the size of the value in bytes.
-        """
         return sys.getsizeof(self.value)
 
     def to_object(self) -> Any:
@@ -104,27 +99,13 @@ class ArraySegment(Segment):
     def markdown(self) -> str:
         items = []
         for item in self.value:
-            items.append(str(item))
+            if hasattr(item, "to_markdown"):
+                items.append(item.to_markdown())
+            else:
+                items.append(str(item))
         return "\n".join(items)
 
 
-class FileSegment(Segment):
-    value_type: SegmentType = SegmentType.FILE
-    value: File
-
-    @property
-    def markdown(self) -> str:
-        return self.value.markdown
-
-    @property
-    def log(self) -> str:
-        return str(self.value)
-
-    @property
-    def text(self) -> str:
-        return str(self.value)
-
-
 class ArrayAnySegment(ArraySegment):
     value_type: SegmentType = SegmentType.ARRAY_ANY
     value: Sequence[Any]
@@ -143,15 +124,3 @@ class ArrayNumberSegment(ArraySegment):
 class ArrayObjectSegment(ArraySegment):
     value_type: SegmentType = SegmentType.ARRAY_OBJECT
     value: Sequence[Mapping[str, Any]]
-
-
-class ArrayFileSegment(ArraySegment):
-    value_type: SegmentType = SegmentType.ARRAY_FILE
-    value: Sequence[File]
-
-    @property
-    def markdown(self) -> str:
-        items = []
-        for item in self.value:
-            items.append(item.markdown)
-        return "\n".join(items)

api/core/app/segments/types.py

@@ -11,7 +11,5 @@ class SegmentType(str, Enum):
     ARRAY_NUMBER = "array[number]"
     ARRAY_OBJECT = "array[object]"
     OBJECT = "object"
-    FILE = "file"
-    ARRAY_FILE = "array[file]"
 
     GROUP = "group"

api/core/app/segments/variables.py

@@ -7,7 +7,6 @@ from .segments import (
     ArrayNumberSegment,
     ArrayObjectSegment,
     ArrayStringSegment,
-    FileSegment,
     FloatSegment,
     IntegerSegment,
     NoneSegment,
@@ -74,7 +73,3 @@ class SecretVariable(StringVariable):
 class NoneVariable(NoneSegment, Variable):
     value_type: SegmentType = SegmentType.NONE
     value: None = None
-
-
-class FileVariable(FileSegment, Variable):
-    pass

api/core/app/task_pipeline/easy_ui_based_generate_task_pipeline.py

@@ -248,8 +248,7 @@ class EasyUIBasedGenerateTaskPipeline(BasedGenerateTaskPipeline, MessageCycleMan
             else:
                 start_listener_time = time.time()
                 yield MessageAudioStreamResponse(audio=audio.audio, task_id=task_id)
-        if publisher:
-            yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
+        yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
 
     def _process_stream_response(
         self, publisher: AppGeneratorTTSPublisher, trace_manager: Optional[TraceQueueManager] = None

api/core/app/task_pipeline/workflow_cycle_manage.py

@@ -1,6 +1,5 @@
 import json
 import time
-from collections.abc import Mapping
 from datetime import datetime, timezone
 from typing import Any, Optional, Union, cast
 
@@ -28,15 +27,15 @@ from core.app.entities.task_entities import (
     WorkflowStartStreamResponse,
     WorkflowTaskState,
 )
-from core.file import FILE_MODEL_IDENTITY, File
+from core.file.file_obj import FileVar
 from core.model_runtime.utils.encoders import jsonable_encoder
 from core.ops.entities.trace_entity import TraceTaskName
 from core.ops.ops_trace_manager import TraceQueueManager, TraceTask
 from core.tools.tool_manager import ToolManager
+from core.workflow.entities.node_entities import NodeType
 from core.workflow.enums import SystemVariableKey
 from core.workflow.nodes.tool.entities import ToolNodeData
 from core.workflow.workflow_entry import WorkflowEntry
-from enums import NodeType, WorkflowRunTriggeredFrom
 from extensions.ext_database import db
 from models.account import Account
 from models.model import EndUser
@@ -48,6 +47,7 @@ from models.workflow import (
     WorkflowNodeExecutionTriggeredFrom,
     WorkflowRun,
     WorkflowRunStatus,
+    WorkflowRunTriggeredFrom,
 )
 
 
@@ -113,7 +113,7 @@ class WorkflowCycleManage:
         start_at: float,
         total_tokens: int,
         total_steps: int,
-        outputs: Mapping[str, Any] | None = None,
+        outputs: Optional[str] = None,
         conversation_id: Optional[str] = None,
         trace_manager: Optional[TraceQueueManager] = None,
     ) -> WorkflowRun:
@@ -129,10 +129,8 @@ class WorkflowCycleManage:
         """
         workflow_run = self._refetch_workflow_run(workflow_run.id)
 
-        outputs = WorkflowEntry.handle_special_values(outputs)
-
         workflow_run.status = WorkflowRunStatus.SUCCEEDED.value
-        workflow_run.outputs = json.dumps(outputs) if outputs else None
+        workflow_run.outputs = outputs
         workflow_run.elapsed_time = time.perf_counter() - start_at
         workflow_run.total_tokens = total_tokens
         workflow_run.total_steps = total_steps
@@ -262,11 +260,10 @@ class WorkflowCycleManage:
 
         inputs = WorkflowEntry.handle_special_values(event.inputs)
         outputs = WorkflowEntry.handle_special_values(event.outputs)
-        process_data = WorkflowEntry.handle_special_values(event.process_data)
 
         workflow_node_execution.status = WorkflowNodeExecutionStatus.SUCCEEDED.value
         workflow_node_execution.inputs = json.dumps(inputs) if inputs else None
-        workflow_node_execution.process_data = json.dumps(process_data) if process_data else None
+        workflow_node_execution.process_data = json.dumps(event.process_data) if event.process_data else None
         workflow_node_execution.outputs = json.dumps(outputs) if outputs else None
         workflow_node_execution.execution_metadata = (
             json.dumps(jsonable_encoder(event.execution_metadata)) if event.execution_metadata else None
@@ -290,13 +287,12 @@ class WorkflowCycleManage:
 
         inputs = WorkflowEntry.handle_special_values(event.inputs)
         outputs = WorkflowEntry.handle_special_values(event.outputs)
-        process_data = WorkflowEntry.handle_special_values(event.process_data)
 
         workflow_node_execution.status = WorkflowNodeExecutionStatus.FAILED.value
         workflow_node_execution.error = event.error
         workflow_node_execution.finished_at = datetime.now(timezone.utc).replace(tzinfo=None)
         workflow_node_execution.inputs = json.dumps(inputs) if inputs else None
-        workflow_node_execution.process_data = json.dumps(process_data) if process_data else None
+        workflow_node_execution.process_data = json.dumps(event.process_data) if event.process_data else None
         workflow_node_execution.outputs = json.dumps(outputs) if outputs else None
         workflow_node_execution.elapsed_time = (workflow_node_execution.finished_at - event.start_at).total_seconds()
 
@@ -613,11 +609,11 @@ class WorkflowCycleManage:
         if not outputs_dict:
             return []
 
-        files = [self._fetch_files_from_variable_value(output_value) for output_value in outputs_dict.values()]
-        # Remove None
-        files = [file for file in files if file]
-        # Flatten list
-        files = [file for sublist in files for file in sublist]
+        files = []
+        for output_var, output_value in outputs_dict.items():
+            file_vars = self._fetch_files_from_variable_value(output_value)
+            if file_vars:
+                files.extend(file_vars)
 
         return files
 
@@ -643,7 +639,7 @@ class WorkflowCycleManage:
 
         return files
 
-    def _get_file_var_from_value(self, value: Union[dict, list]) -> Mapping[str, str | int | None] | None:
+    def _get_file_var_from_value(self, value: Union[dict, list]) -> Optional[dict]:
         """
         Get file var from value
         :param value: variable value
@@ -652,11 +648,14 @@ class WorkflowCycleManage:
         if not value:
             return None
 
-        if isinstance(value, dict) and value.get("dify_model_identity") == FILE_MODEL_IDENTITY:
-            return value
-        elif isinstance(value, File):
+        if isinstance(value, dict):
+            if "__variant" in value and value["__variant"] == FileVar.__name__:
+                return value
+        elif isinstance(value, FileVar):
             return value.to_dict()
 
+        return None
+
     def _refetch_workflow_run(self, workflow_run_id: str) -> WorkflowRun:
         """
         Refetch workflow run

api/core/entities/message_entities.py

@@ -0,0 +1,29 @@
+import enum
+from typing import Any
+
+from pydantic import BaseModel
+
+
+class PromptMessageFileType(enum.Enum):
+    IMAGE = "image"
+
+    @staticmethod
+    def value_of(value):
+        for member in PromptMessageFileType:
+            if member.value == value:
+                return member
+        raise ValueError(f"No matching enum found for value '{value}'")
+
+
+class PromptMessageFile(BaseModel):
+    type: PromptMessageFileType
+    data: Any = None
+
+
+class ImagePromptMessageFile(PromptMessageFile):
+    class DETAIL(enum.Enum):
+        LOW = "low"
+        HIGH = "high"
+
+    type: PromptMessageFileType = PromptMessageFileType.IMAGE
+    detail: DETAIL = DETAIL.LOW

api/core/entities/provider_configuration.py

@@ -119,7 +119,7 @@ class ProviderConfiguration(BaseModel):
                         credentials = model_configuration.credentials
                         break
 
-            if not credentials and self.custom_configuration.provider:
+            if self.custom_configuration.provider:
                 credentials = self.custom_configuration.provider.credentials
 
             return credentials

api/core/file/__init__.py

@@ -1,19 +0,0 @@
-from .constants import FILE_MODEL_IDENTITY
-from .enums import ArrayFileAttribute, FileAttribute, FileBelongsTo, FileTransferMethod, FileType
-from .models import (
-    File,
-    FileExtraConfig,
-    ImageConfig,
-)
-
-__all__ = [
-    "FileType",
-    "FileExtraConfig",
-    "FileTransferMethod",
-    "FileBelongsTo",
-    "File",
-    "ImageConfig",
-    "FileAttribute",
-    "ArrayFileAttribute",
-    "FILE_MODEL_IDENTITY",
-]

api/core/file/constants.py

@@ -1 +0,0 @@
-FILE_MODEL_IDENTITY = "__dify__file__"

api/core/file/enums.py

@@ -1,55 +0,0 @@
-from enum import Enum
-
-
-class FileType(str, Enum):
-    IMAGE = "image"
-    DOCUMENT = "document"
-    AUDIO = "audio"
-    VIDEO = "video"
-    CUSTOM = "custom"
-
-    @staticmethod
-    def value_of(value):
-        for member in FileType:
-            if member.value == value:
-                return member
-        raise ValueError(f"No matching enum found for value '{value}'")
-
-
-class FileTransferMethod(str, Enum):
-    REMOTE_URL = "remote_url"
-    LOCAL_FILE = "local_file"
-    TOOL_FILE = "tool_file"
-
-    @staticmethod
-    def value_of(value):
-        for member in FileTransferMethod:
-            if member.value == value:
-                return member
-        raise ValueError(f"No matching enum found for value '{value}'")
-
-
-class FileBelongsTo(str, Enum):
-    USER = "user"
-    ASSISTANT = "assistant"
-
-    @staticmethod
-    def value_of(value):
-        for member in FileBelongsTo:
-            if member.value == value:
-                return member
-        raise ValueError(f"No matching enum found for value '{value}'")
-
-
-class FileAttribute(str, Enum):
-    TYPE = "type"
-    SIZE = "size"
-    NAME = "name"
-    MIME_TYPE = "mime_type"
-    TRANSFER_METHOD = "transfer_method"
-    URL = "url"
-    EXTENSION = "extension"
-
-
-class ArrayFileAttribute(str, Enum):
-    LENGTH = "length"

api/core/file/file_manager.py

@@ -1,136 +0,0 @@
-import base64
-
-from configs import dify_config
-from core.model_runtime.entities.message_entities import ImagePromptMessageContent
-from extensions.ext_database import db
-from extensions.ext_storage import storage
-from models import UploadFile
-
-from . import helpers
-from .enums import FileAttribute
-from .models import File, FileTransferMethod, FileType
-from .tool_file_parser import ToolFileParser
-
-
-def get_attr(*, file: "File", attr: "FileAttribute"):
-    match attr:
-        case FileAttribute.TYPE:
-            return file.type.value
-        case FileAttribute.SIZE:
-            return file.size
-        case FileAttribute.NAME:
-            return file.filename
-        case FileAttribute.MIME_TYPE:
-            return file.mime_type
-        case FileAttribute.TRANSFER_METHOD:
-            return file.transfer_method.value
-        case FileAttribute.URL:
-            return file.remote_url
-        case FileAttribute.EXTENSION:
-            return file.extension
-        case _:
-            raise ValueError(f"Invalid file attribute: {attr}")
-
-
-def to_prompt_message_content(file: "File", /):
-    """
-    Convert a File object to an ImagePromptMessageContent object.
-
-    This function takes a File object and converts it to an ImagePromptMessageContent
-    object, which can be used as a prompt for image-based AI models.
-
-    Args:
-        file (File): The File object to convert. Must be of type FileType.IMAGE.
-
-    Returns:
-        ImagePromptMessageContent: An object containing the image data and detail level.
-
-    Raises:
-        ValueError: If the file is not an image or if the file data is missing.
-
-    Note:
-        The detail level of the image prompt is determined by the file's extra_config.
-        If not specified, it defaults to ImagePromptMessageContent.DETAIL.LOW.
-    """
-    if file.type != FileType.IMAGE:
-        raise ValueError("Only image file can convert to prompt message content")
-
-    url_or_b64_data = _get_url_or_b64_data(file=file)
-    if url_or_b64_data is None:
-        raise ValueError("Missing file data")
-
-    # decide the detail of image prompt message content
-    if file._extra_config and file._extra_config.image_config and file._extra_config.image_config.detail:
-        detail = file._extra_config.image_config.detail
-    else:
-        detail = ImagePromptMessageContent.DETAIL.LOW
-
-    return ImagePromptMessageContent(data=url_or_b64_data, detail=detail)
-
-
-def download(*, upload_file_id: str, tenant_id: str):
-    upload_file = (
-        db.session.query(UploadFile).filter(UploadFile.id == upload_file_id, UploadFile.tenant_id == tenant_id).first()
-    )
-
-    if not upload_file:
-        raise ValueError("upload file not found")
-
-    return _download(upload_file.key)
-
-
-def _download(path: str, /):
-    """
-    Download and return the contents of a file as bytes.
-
-    This function loads the file from storage and ensures it's in bytes format.
-
-    Args:
-        path (str): The path to the file in storage.
-
-    Returns:
-        bytes: The contents of the file as a bytes object.
-
-    Raises:
-        ValueError: If the loaded file is not a bytes object.
-    """
-    data = storage.load(path, stream=False)
-    if not isinstance(data, bytes):
-        raise ValueError(f"file {path} is not a bytes object")
-    return data
-
-
-def _get_base64(*, upload_file_id: str, tenant_id: str) -> str | None:
-    upload_file = (
-        db.session.query(UploadFile).filter(UploadFile.id == upload_file_id, UploadFile.tenant_id == tenant_id).first()
-    )
-
-    if not upload_file:
-        return None
-
-    data = _download(upload_file.key)
-    if data is None:
-        return None
-
-    encoded_string = base64.b64encode(data).decode("utf-8")
-    return f"data:{upload_file.mime_type};base64,{encoded_string}"
-
-
-def _get_url_or_b64_data(file: "File"):
-    if file.type == FileType.IMAGE:
-        if file.transfer_method == FileTransferMethod.REMOTE_URL:
-            return file.remote_url
-        elif file.transfer_method == FileTransferMethod.LOCAL_FILE:
-            if file.related_id is None:
-                raise ValueError("Missing file related_id")
-
-            if dify_config.MULTIMODAL_SEND_IMAGE_FORMAT == "url":
-                return helpers.get_signed_image_url(upload_file_id=file.related_id)
-            return _get_base64(upload_file_id=file.related_id, tenant_id=file.tenant_id)
-        elif file.transfer_method == FileTransferMethod.TOOL_FILE:
-            # add sign url
-            if file.related_id is None or file.extension is None:
-                raise ValueError("Missing file related_id or extension")
-            return ToolFileParser.get_tool_file_manager().sign_file(
-                tool_file_id=file.related_id, extension=file.extension
-            )

api/core/file/file_obj.py

@@ -0,0 +1,145 @@
+import enum
+from typing import Any, Optional
+
+from pydantic import BaseModel
+
+from core.file.tool_file_parser import ToolFileParser
+from core.file.upload_file_parser import UploadFileParser
+from core.model_runtime.entities.message_entities import ImagePromptMessageContent
+from extensions.ext_database import db
+
+
+class FileExtraConfig(BaseModel):
+    """
+    File Upload Entity.
+    """
+
+    image_config: Optional[dict[str, Any]] = None
+
+
+class FileType(enum.Enum):
+    IMAGE = "image"
+
+    @staticmethod
+    def value_of(value):
+        for member in FileType:
+            if member.value == value:
+                return member
+        raise ValueError(f"No matching enum found for value '{value}'")
+
+
+class FileTransferMethod(enum.Enum):
+    REMOTE_URL = "remote_url"
+    LOCAL_FILE = "local_file"
+    TOOL_FILE = "tool_file"
+
+    @staticmethod
+    def value_of(value):
+        for member in FileTransferMethod:
+            if member.value == value:
+                return member
+        raise ValueError(f"No matching enum found for value '{value}'")
+
+
+class FileBelongsTo(enum.Enum):
+    USER = "user"
+    ASSISTANT = "assistant"
+
+    @staticmethod
+    def value_of(value):
+        for member in FileBelongsTo:
+            if member.value == value:
+                return member
+        raise ValueError(f"No matching enum found for value '{value}'")
+
+
+class FileVar(BaseModel):
+    id: Optional[str] = None  # message file id
+    tenant_id: str
+    type: FileType
+    transfer_method: FileTransferMethod
+    url: Optional[str] = None  # remote url
+    related_id: Optional[str] = None
+    extra_config: Optional[FileExtraConfig] = None
+    filename: Optional[str] = None
+    extension: Optional[str] = None
+    mime_type: Optional[str] = None
+
+    def to_dict(self) -> dict:
+        return {
+            "__variant": self.__class__.__name__,
+            "tenant_id": self.tenant_id,
+            "type": self.type.value,
+            "transfer_method": self.transfer_method.value,
+            "url": self.preview_url,
+            "remote_url": self.url,
+            "related_id": self.related_id,
+            "filename": self.filename,
+            "extension": self.extension,
+            "mime_type": self.mime_type,
+        }
+
+    def to_markdown(self) -> str:
+        """
+        Convert file to markdown
+        :return:
+        """
+        preview_url = self.preview_url
+        if self.type == FileType.IMAGE:
+            text = f'![{self.filename or ""}]({preview_url})'
+        else:
+            text = f"[{self.filename or preview_url}]({preview_url})"
+
+        return text
+
+    @property
+    def data(self) -> Optional[str]:
+        """
+        Get image data, file signed url or base64 data
+        depending on config MULTIMODAL_SEND_IMAGE_FORMAT
+        :return:
+        """
+        return self._get_data()
+
+    @property
+    def preview_url(self) -> Optional[str]:
+        """
+        Get signed preview url
+        :return:
+        """
+        return self._get_data(force_url=True)
+
+    @property
+    def prompt_message_content(self) -> ImagePromptMessageContent:
+        if self.type == FileType.IMAGE:
+            image_config = self.extra_config.image_config
+
+            return ImagePromptMessageContent(
+                data=self.data,
+                detail=ImagePromptMessageContent.DETAIL.HIGH
+                if image_config.get("detail") == "high"
+                else ImagePromptMessageContent.DETAIL.LOW,
+            )
+
+    def _get_data(self, force_url: bool = False) -> Optional[str]:
+        from models.model import UploadFile
+
+        if self.type == FileType.IMAGE:
+            if self.transfer_method == FileTransferMethod.REMOTE_URL:
+                return self.url
+            elif self.transfer_method == FileTransferMethod.LOCAL_FILE:
+                upload_file = (
+                    db.session.query(UploadFile)
+                    .filter(UploadFile.id == self.related_id, UploadFile.tenant_id == self.tenant_id)
+                    .first()
+                )
+
+                return UploadFileParser.get_image_data(upload_file=upload_file, force_url=force_url)
+            elif self.transfer_method == FileTransferMethod.TOOL_FILE:
+                extension = self.extension
+                # add sign url
+                return ToolFileParser.get_tool_file_manager().sign_file(
+                    tool_file_id=self.related_id, extension=extension
+                )
+
+        return None

api/core/file/helpers.py

@@ -1,61 +0,0 @@
-import base64
-import hashlib
-import hmac
-import os
-import time
-
-from configs import dify_config
-
-
-def get_signed_image_url(upload_file_id: str) -> str:
-    url = f"{dify_config.FILES_URL}/files/{upload_file_id}/image-preview"
-
-    timestamp = str(int(time.time()))
-    nonce = os.urandom(16).hex()
-    key = dify_config.SECRET_KEY.encode()
-    msg = f"image-preview|{upload_file_id}|{timestamp}|{nonce}"
-    sign = hmac.new(key, msg.encode(), hashlib.sha256).digest()
-    encoded_sign = base64.urlsafe_b64encode(sign).decode()
-
-    return f"{url}?timestamp={timestamp}&nonce={nonce}&sign={encoded_sign}"
-
-
-def get_signed_file_url(upload_file_id: str) -> str:
-    url = f"{dify_config.FILES_URL}/files/{upload_file_id}/file-preview"
-
-    timestamp = str(int(time.time()))
-    nonce = os.urandom(16).hex()
-    key = dify_config.SECRET_KEY.encode()
-    msg = f"file-preview|{upload_file_id}|{timestamp}|{nonce}"
-    sign = hmac.new(key, msg.encode(), hashlib.sha256).digest()
-    encoded_sign = base64.urlsafe_b64encode(sign).decode()
-
-    return f"{url}?timestamp={timestamp}&nonce={nonce}&sign={encoded_sign}"
-
-
-def verify_image_signature(*, upload_file_id: str, timestamp: str, nonce: str, sign: str) -> bool:
-    data_to_sign = f"image-preview|{upload_file_id}|{timestamp}|{nonce}"
-    secret_key = dify_config.SECRET_KEY.encode()
-    recalculated_sign = hmac.new(secret_key, data_to_sign.encode(), hashlib.sha256).digest()
-    recalculated_encoded_sign = base64.urlsafe_b64encode(recalculated_sign).decode()
-
-    # verify signature
-    if sign != recalculated_encoded_sign:
-        return False
-
-    current_time = int(time.time())
-    return current_time - int(timestamp) <= dify_config.FILES_ACCESS_TIMEOUT
-
-
-def verify_file_signature(*, upload_file_id: str, timestamp: str, nonce: str, sign: str) -> bool:
-    data_to_sign = f"file-preview|{upload_file_id}|{timestamp}|{nonce}"
-    secret_key = dify_config.SECRET_KEY.encode()
-    recalculated_sign = hmac.new(secret_key, data_to_sign.encode(), hashlib.sha256).digest()
-    recalculated_encoded_sign = base64.urlsafe_b64encode(recalculated_sign).decode()
-
-    # verify signature
-    if sign != recalculated_encoded_sign:
-        return False
-
-    current_time = int(time.time())
-    return current_time - int(timestamp) <= dify_config.FILES_ACCESS_TIMEOUT

api/core/file/message_file_parser.py

@@ -0,0 +1,225 @@
+import re
+from collections.abc import Mapping, Sequence
+from typing import Any, Union
+from urllib.parse import parse_qs, urlparse
+
+import requests
+
+from core.file.file_obj import FileBelongsTo, FileExtraConfig, FileTransferMethod, FileType, FileVar
+from extensions.ext_database import db
+from models.account import Account
+from models.model import EndUser, MessageFile, UploadFile
+from services.file_service import IMAGE_EXTENSIONS
+
+
+class MessageFileParser:
+    def __init__(self, tenant_id: str, app_id: str) -> None:
+        self.tenant_id = tenant_id
+        self.app_id = app_id
+
+    def validate_and_transform_files_arg(
+        self, files: Sequence[Mapping[str, Any]], file_extra_config: FileExtraConfig, user: Union[Account, EndUser]
+    ) -> list[FileVar]:
+        """
+        validate and transform files arg
+
+        :param files:
+        :param file_extra_config:
+        :param user:
+        :return:
+        """
+        for file in files:
+            if not isinstance(file, dict):
+                raise ValueError("Invalid file format, must be dict")
+            if not file.get("type"):
+                raise ValueError("Missing file type")
+            FileType.value_of(file.get("type"))
+            if not file.get("transfer_method"):
+                raise ValueError("Missing file transfer method")
+            FileTransferMethod.value_of(file.get("transfer_method"))
+            if file.get("transfer_method") == FileTransferMethod.REMOTE_URL.value:
+                if not file.get("url"):
+                    raise ValueError("Missing file url")
+                if not file.get("url").startswith("http"):
+                    raise ValueError("Invalid file url")
+            if file.get("transfer_method") == FileTransferMethod.LOCAL_FILE.value and not file.get("upload_file_id"):
+                raise ValueError("Missing file upload_file_id")
+            if file.get("transform_method") == FileTransferMethod.TOOL_FILE.value and not file.get("tool_file_id"):
+                raise ValueError("Missing file tool_file_id")
+
+        # transform files to file objs
+        type_file_objs = self._to_file_objs(files, file_extra_config)
+
+        # validate files
+        new_files = []
+        for file_type, file_objs in type_file_objs.items():
+            if file_type == FileType.IMAGE:
+                # parse and validate files
+                image_config = file_extra_config.image_config
+
+                # check if image file feature is enabled
+                if not image_config:
+                    continue
+
+                # Validate number of files
+                if len(files) > image_config["number_limits"]:
+                    raise ValueError(f"Number of image files exceeds the maximum limit {image_config['number_limits']}")
+
+                for file_obj in file_objs:
+                    # Validate transfer method
+                    if file_obj.transfer_method.value not in image_config["transfer_methods"]:
+                        raise ValueError(f"Invalid transfer method: {file_obj.transfer_method.value}")
+
+                    # Validate file type
+                    if file_obj.type != FileType.IMAGE:
+                        raise ValueError(f"Invalid file type: {file_obj.type}")
+
+                    if file_obj.transfer_method == FileTransferMethod.REMOTE_URL:
+                        # check remote url valid and is image
+                        result, error = self._check_image_remote_url(file_obj.url)
+                        if result is False:
+                            raise ValueError(error)
+                    elif file_obj.transfer_method == FileTransferMethod.LOCAL_FILE:
+                        # get upload file from upload_file_id
+                        upload_file = (
+                            db.session.query(UploadFile)
+                            .filter(
+                                UploadFile.id == file_obj.related_id,
+                                UploadFile.tenant_id == self.tenant_id,
+                                UploadFile.created_by == user.id,
+                                UploadFile.created_by_role == ("account" if isinstance(user, Account) else "end_user"),
+                                UploadFile.extension.in_(IMAGE_EXTENSIONS),
+                            )
+                            .first()
+                        )
+
+                        # check upload file is belong to tenant and user
+                        if not upload_file:
+                            raise ValueError("Invalid upload file")
+
+                    new_files.append(file_obj)
+
+        # return all file objs
+        return new_files
+
+    def transform_message_files(self, files: list[MessageFile], file_extra_config: FileExtraConfig):
+        """
+        transform message files
+
+        :param files:
+        :param file_extra_config:
+        :return:
+        """
+        # transform files to file objs
+        type_file_objs = self._to_file_objs(files, file_extra_config)
+
+        # return all file objs
+        return [file_obj for file_objs in type_file_objs.values() for file_obj in file_objs]
+
+    def _to_file_objs(
+        self, files: list[Union[dict, MessageFile]], file_extra_config: FileExtraConfig
+    ) -> dict[FileType, list[FileVar]]:
+        """
+        transform files to file objs
+
+        :param files:
+        :param file_extra_config:
+        :return:
+        """
+        type_file_objs: dict[FileType, list[FileVar]] = {
+            # Currently only support image
+            FileType.IMAGE: []
+        }
+
+        if not files:
+            return type_file_objs
+
+        # group by file type and convert file args or message files to FileObj
+        for file in files:
+            if isinstance(file, MessageFile):
+                if file.belongs_to == FileBelongsTo.ASSISTANT.value:
+                    continue
+
+            file_obj = self._to_file_obj(file, file_extra_config)
+            if file_obj.type not in type_file_objs:
+                continue
+
+            type_file_objs[file_obj.type].append(file_obj)
+
+        return type_file_objs
+
+    def _to_file_obj(self, file: Union[dict, MessageFile], file_extra_config: FileExtraConfig):
+        """
+        transform file to file obj
+
+        :param file:
+        :return:
+        """
+        if isinstance(file, dict):
+            transfer_method = FileTransferMethod.value_of(file.get("transfer_method"))
+            if transfer_method != FileTransferMethod.TOOL_FILE:
+                return FileVar(
+                    tenant_id=self.tenant_id,
+                    type=FileType.value_of(file.get("type")),
+                    transfer_method=transfer_method,
+                    url=file.get("url") if transfer_method == FileTransferMethod.REMOTE_URL else None,
+                    related_id=file.get("upload_file_id") if transfer_method == FileTransferMethod.LOCAL_FILE else None,
+                    extra_config=file_extra_config,
+                )
+            return FileVar(
+                tenant_id=self.tenant_id,
+                type=FileType.value_of(file.get("type")),
+                transfer_method=transfer_method,
+                url=None,
+                related_id=file.get("tool_file_id"),
+                extra_config=file_extra_config,
+            )
+        else:
+            return FileVar(
+                id=file.id,
+                tenant_id=self.tenant_id,
+                type=FileType.value_of(file.type),
+                transfer_method=FileTransferMethod.value_of(file.transfer_method),
+                url=file.url,
+                related_id=file.upload_file_id or None,
+                extra_config=file_extra_config,
+            )
+
+    def _check_image_remote_url(self, url):
+        try:
+            headers = {
+                "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko)"
+                " Chrome/91.0.4472.124 Safari/537.36"
+            }
+
+            def is_s3_presigned_url(url):
+                try:
+                    parsed_url = urlparse(url)
+                    if "amazonaws.com" not in parsed_url.netloc:
+                        return False
+                    query_params = parse_qs(parsed_url.query)
+                    required_params = ["Signature", "Expires"]
+                    for param in required_params:
+                        if param not in query_params:
+                            return False
+                    if not query_params["Expires"][0].isdigit():
+                        return False
+                    signature = query_params["Signature"][0]
+                    if not re.match(r"^[A-Za-z0-9+/]+={0,2}$", signature):
+                        return False
+                    return True
+                except Exception:
+                    return False
+
+            if is_s3_presigned_url(url):
+                response = requests.get(url, headers=headers, allow_redirects=True)
+                if response.status_code in {200, 304}:
+                    return True, ""
+
+            response = requests.head(url, headers=headers, allow_redirects=True)
+            if response.status_code in {200, 304}:
+                return True, ""
+            else:
+                return False, "URL does not exist."
+        except requests.RequestException as e:
+            return False, f"Error checking URL: {e}"

api/core/file/models.py

@@ -1,140 +0,0 @@
-from collections.abc import Mapping, Sequence
-from typing import Optional
-
-from pydantic import BaseModel, Field, model_validator
-
-from core.model_runtime.entities.message_entities import ImagePromptMessageContent
-
-from . import helpers
-from .constants import FILE_MODEL_IDENTITY
-from .enums import FileTransferMethod, FileType
-from .tool_file_parser import ToolFileParser
-
-
-class ImageConfig(BaseModel):
-    """
-    NOTE: This part of validation is deprecated, but still used in app features "Image Upload".
-    """
-
-    number_limits: int = 0
-    transfer_methods: Sequence[FileTransferMethod] = Field(default_factory=list)
-    detail: ImagePromptMessageContent.DETAIL | None = None
-
-
-class FileExtraConfig(BaseModel):
-    """
-    File Upload Entity.
-    """
-
-    image_config: Optional[ImageConfig] = None
-    allowed_file_types: Sequence[FileType] = Field(default_factory=list)
-    allowed_extensions: Sequence[str] = Field(default_factory=list)
-    allowed_upload_methods: Sequence[FileTransferMethod] = Field(default_factory=list)
-    number_limits: int = 0
-
-
-class File(BaseModel):
-    dify_model_identity: str = FILE_MODEL_IDENTITY
-
-    id: Optional[str] = None  # message file id
-    tenant_id: str
-    type: FileType
-    transfer_method: FileTransferMethod
-    remote_url: Optional[str] = None  # remote url
-    related_id: Optional[str] = None
-    filename: Optional[str] = None
-    extension: Optional[str] = None
-    mime_type: Optional[str] = None
-    size: int = 0
-    _extra_config: FileExtraConfig | None = None
-
-    def to_dict(self) -> Mapping[str, str | int | None]:
-        data = self.model_dump()
-        return {
-            **data,
-            "url": self.generate_url(),
-        }
-
-    @property
-    def markdown(self) -> str:
-        url = self.generate_url()
-        if self.type == FileType.IMAGE:
-            text = f'![{self.filename or ""}]({url})'
-        else:
-            text = f"[{self.filename or url}]({url})"
-
-        return text
-
-    def generate_url(self) -> Optional[str]:
-        if self.type == FileType.IMAGE:
-            if self.transfer_method == FileTransferMethod.REMOTE_URL:
-                return self.remote_url
-            elif self.transfer_method == FileTransferMethod.LOCAL_FILE:
-                if self.related_id is None:
-                    raise ValueError("Missing file related_id")
-                return helpers.get_signed_image_url(upload_file_id=self.related_id)
-            elif self.transfer_method == FileTransferMethod.TOOL_FILE:
-                assert self.related_id is not None
-                assert self.extension is not None
-                return ToolFileParser.get_tool_file_manager().sign_file(
-                    tool_file_id=self.related_id, extension=self.extension
-                )
-        else:
-            if self.transfer_method == FileTransferMethod.REMOTE_URL:
-                return self.remote_url
-            elif self.transfer_method == FileTransferMethod.LOCAL_FILE:
-                if self.related_id is None:
-                    raise ValueError("Missing file related_id")
-                return helpers.get_signed_file_url(upload_file_id=self.related_id)
-            elif self.transfer_method == FileTransferMethod.TOOL_FILE:
-                assert self.related_id is not None
-                assert self.extension is not None
-                return ToolFileParser.get_tool_file_manager().sign_file(
-                    tool_file_id=self.related_id, extension=self.extension
-                )
-
-    @model_validator(mode="after")
-    def validate_after(self):
-        match self.transfer_method:
-            case FileTransferMethod.REMOTE_URL:
-                if not self.remote_url:
-                    raise ValueError("Missing file url")
-                if not isinstance(self.remote_url, str) or not self.remote_url.startswith("http"):
-                    raise ValueError("Invalid file url")
-            case FileTransferMethod.LOCAL_FILE:
-                if not self.related_id:
-                    raise ValueError("Missing file related_id")
-            case FileTransferMethod.TOOL_FILE:
-                if not self.related_id:
-                    raise ValueError("Missing file related_id")
-
-        # Validate the extra config.
-        if not self._extra_config:
-            return self
-
-        if self._extra_config.allowed_file_types:
-            if self.type not in self._extra_config.allowed_file_types and self.type != FileType.CUSTOM:
-                raise ValueError(f"Invalid file type: {self.type}")
-
-        if self._extra_config.allowed_extensions and self.extension not in self._extra_config.allowed_extensions:
-            raise ValueError(f"Invalid file extension: {self.extension}")
-
-        if (
-            self._extra_config.allowed_upload_methods
-            and self.transfer_method not in self._extra_config.allowed_upload_methods
-        ):
-            raise ValueError(f"Invalid transfer method: {self.transfer_method}")
-
-        match self.type:
-            case FileType.IMAGE:
-                # NOTE: This part of validation is deprecated, but still used in app features "Image Upload".
-                if not self._extra_config.image_config:
-                    return self
-                # TODO: skip check if transfer_methods is empty, because many test cases are not setting this field
-                if (
-                    self._extra_config.image_config.transfer_methods
-                    and self.transfer_method not in self._extra_config.image_config.transfer_methods
-                ):
-                    raise ValueError(f"Invalid transfer method: {self.transfer_method}")
-
-        return self

api/core/file/tool_file_parser.py

@@ -1,9 +1,4 @@
-from typing import TYPE_CHECKING, Any
-
-if TYPE_CHECKING:
-    from core.tools.tool_file_manager import ToolFileManager
-
-tool_file_manager: dict[str, Any] = {"manager": None}
+tool_file_manager = {"manager": None}
 
 
 class ToolFileParser:

api/core/file/upload_file_parser.py

@@ -0,0 +1,79 @@
+import base64
+import hashlib
+import hmac
+import logging
+import os
+import time
+from typing import Optional
+
+from configs import dify_config
+from extensions.ext_storage import storage
+
+IMAGE_EXTENSIONS = ["jpg", "jpeg", "png", "webp", "gif", "svg"]
+IMAGE_EXTENSIONS.extend([ext.upper() for ext in IMAGE_EXTENSIONS])
+
+
+class UploadFileParser:
+    @classmethod
+    def get_image_data(cls, upload_file, force_url: bool = False) -> Optional[str]:
+        if not upload_file:
+            return None
+
+        if upload_file.extension not in IMAGE_EXTENSIONS:
+            return None
+
+        if dify_config.MULTIMODAL_SEND_IMAGE_FORMAT == "url" or force_url:
+            return cls.get_signed_temp_image_url(upload_file.id)
+        else:
+            # get image file base64
+            try:
+                data = storage.load(upload_file.key)
+            except FileNotFoundError:
+                logging.error(f"File not found: {upload_file.key}")
+                return None
+
+            encoded_string = base64.b64encode(data).decode("utf-8")
+            return f"data:{upload_file.mime_type};base64,{encoded_string}"
+
+    @classmethod
+    def get_signed_temp_image_url(cls, upload_file_id) -> str:
+        """
+        get signed url from upload file
+
+        :param upload_file: UploadFile object
+        :return:
+        """
+        base_url = dify_config.FILES_URL
+        image_preview_url = f"{base_url}/files/{upload_file_id}/image-preview"
+
+        timestamp = str(int(time.time()))
+        nonce = os.urandom(16).hex()
+        data_to_sign = f"image-preview|{upload_file_id}|{timestamp}|{nonce}"
+        secret_key = dify_config.SECRET_KEY.encode()
+        sign = hmac.new(secret_key, data_to_sign.encode(), hashlib.sha256).digest()
+        encoded_sign = base64.urlsafe_b64encode(sign).decode()
+
+        return f"{image_preview_url}?timestamp={timestamp}&nonce={nonce}&sign={encoded_sign}"
+
+    @classmethod
+    def verify_image_file_signature(cls, upload_file_id: str, timestamp: str, nonce: str, sign: str) -> bool:
+        """
+        verify signature
+
+        :param upload_file_id: file id
+        :param timestamp: timestamp
+        :param nonce: nonce
+        :param sign: signature
+        :return:
+        """
+        data_to_sign = f"image-preview|{upload_file_id}|{timestamp}|{nonce}"
+        secret_key = dify_config.SECRET_KEY.encode()
+        recalculated_sign = hmac.new(secret_key, data_to_sign.encode(), hashlib.sha256).digest()
+        recalculated_encoded_sign = base64.urlsafe_b64encode(recalculated_sign).decode()
+
+        # verify signature
+        if sign != recalculated_encoded_sign:
+            return False
+
+        current_time = int(time.time())
+        return current_time - int(timestamp) <= dify_config.FILES_ACCESS_TIMEOUT

api/core/helper/ssrf_proxy.py

@@ -13,11 +13,8 @@ SSRF_PROXY_HTTP_URL = os.getenv("SSRF_PROXY_HTTP_URL", "")
 SSRF_PROXY_HTTPS_URL = os.getenv("SSRF_PROXY_HTTPS_URL", "")
 SSRF_DEFAULT_MAX_RETRIES = int(os.getenv("SSRF_DEFAULT_MAX_RETRIES", "3"))
 
-proxy_mounts = (
-    {
-        "http://": httpx.HTTPTransport(proxy=SSRF_PROXY_HTTP_URL),
-        "https://": httpx.HTTPTransport(proxy=SSRF_PROXY_HTTPS_URL),
-    }
+proxies = (
+    {"http://": SSRF_PROXY_HTTP_URL, "https://": SSRF_PROXY_HTTPS_URL}
     if SSRF_PROXY_HTTP_URL and SSRF_PROXY_HTTPS_URL
     else None
 )
@@ -36,14 +33,11 @@ def make_request(method, url, max_retries=SSRF_DEFAULT_MAX_RETRIES, **kwargs):
     while retries <= max_retries:
         try:
             if SSRF_PROXY_ALL_URL:
-                with httpx.Client(proxy=SSRF_PROXY_ALL_URL) as client:
-                    response = client.request(method=method, url=url, **kwargs)
-            elif proxy_mounts:
-                with httpx.Client(mounts=proxy_mounts) as client:
-                    response = client.request(method=method, url=url, **kwargs)
+                response = httpx.request(method=method, url=url, proxy=SSRF_PROXY_ALL_URL, **kwargs)
+            elif proxies:
+                response = httpx.request(method=method, url=url, proxies=proxies, **kwargs)
             else:
-                with httpx.Client() as client:
-                    response = client.request(method=method, url=url, **kwargs)
+                response = httpx.request(method=method, url=url, **kwargs)
 
             if response.status_code not in STATUS_FORCELIST:
                 return response

api/core/memory/token_buffer_memory.py

@@ -1,20 +1,18 @@
 from typing import Optional
 
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
-from core.file import file_manager
+from core.file.message_file_parser import MessageFileParser
 from core.model_manager import ModelInstance
-from core.model_runtime.entities import (
+from core.model_runtime.entities.message_entities import (
     AssistantPromptMessage,
     ImagePromptMessageContent,
     PromptMessage,
-    PromptMessageContent,
     PromptMessageRole,
     TextPromptMessageContent,
     UserPromptMessage,
 )
 from core.prompt.utils.extract_thread_messages import extract_thread_messages
 from extensions.ext_database import db
-from factories import file_factory
 from models.model import AppMode, Conversation, Message, MessageFile
 from models.workflow import WorkflowRun
 
@@ -63,6 +61,7 @@ class TokenBufferMemory:
         thread_messages.pop(0)
         messages = list(reversed(thread_messages))
 
+        message_file_parser = MessageFileParser(tenant_id=app_record.tenant_id, app_id=app_record.id)
         prompt_messages = []
         for message in messages:
             files = db.session.query(MessageFile).filter(MessageFile.message_id == message.id).all()
@@ -81,20 +80,17 @@ class TokenBufferMemory:
                                 workflow_run.workflow.features_dict, is_vision=False
                             )
 
-                if file_extra_config and app_record:
-                    file_objs = file_factory.build_from_message_files(
-                        message_files=files, tenant_id=app_record.tenant_id, config=file_extra_config
-                    )
+                if file_extra_config:
+                    file_objs = message_file_parser.transform_message_files(files, file_extra_config)
                 else:
                     file_objs = []
 
                 if not file_objs:
                     prompt_messages.append(UserPromptMessage(content=message.query))
                 else:
-                    prompt_message_contents: list[PromptMessageContent] = []
-                    prompt_message_contents.append(TextPromptMessageContent(data=message.query))
+                    prompt_message_contents = [TextPromptMessageContent(data=message.query)]
                     for file_obj in file_objs:
-                        prompt_message_contents.append(file_manager.to_prompt_message_content(file_obj))
+                        prompt_message_contents.append(file_obj.prompt_message_content)
 
                     prompt_messages.append(UserPromptMessage(content=prompt_message_contents))
             else:

api/core/model_runtime/callbacks/base_callback.py

@@ -1,4 +1,3 @@
-from abc import ABC, abstractmethod
 from typing import Optional
 
 from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk
@@ -14,7 +13,7 @@ _TEXT_COLOR_MAPPING = {
 }
 
 
-class Callback(ABC):
+class Callback:
     """
     Base class for callbacks.
     Only for LLM.
@@ -22,7 +21,6 @@ class Callback(ABC):
 
     raise_error: bool = False
 
-    @abstractmethod
     def on_before_invoke(
         self,
         llm_instance: AIModel,
@@ -50,7 +48,6 @@ class Callback(ABC):
         """
         raise NotImplementedError()
 
-    @abstractmethod
     def on_new_chunk(
         self,
         llm_instance: AIModel,
@@ -80,7 +77,6 @@ class Callback(ABC):
         """
         raise NotImplementedError()
 
-    @abstractmethod
     def on_after_invoke(
         self,
         llm_instance: AIModel,
@@ -110,7 +106,6 @@ class Callback(ABC):
         """
         raise NotImplementedError()
 
-    @abstractmethod
     def on_invoke_error(
         self,
         llm_instance: AIModel,

api/core/model_runtime/docs/en_US/customizable_model_scale_out.md

@@ -1,310 +0,0 @@
-## Custom Integration of Pre-defined Models
-
-### Introduction
-
-After completing the vendors integration, the next step is to connect the vendor's models. To illustrate the entire connection process, we will use Xinference as an example to demonstrate a complete vendor integration.
-
-It is important to note that for custom models, each model connection requires a complete vendor credential.
-
-Unlike pre-defined models, a custom vendor integration always includes the following two parameters, which do not need to be defined in the vendor YAML file.
-
-![](images/index/image-3.png)
-
-As mentioned earlier, vendors do not need to implement validate_provider_credential. The runtime will automatically call the corresponding model layer's validate_credentials to validate the credentials based on the model type and name selected by the user.
-
-### Writing the Vendor YAML
-
-First, we need to identify the types of models supported by the vendor we are integrating.
-
-Currently supported model types are as follows:
-
-- `llm` Text Generation Models
-
-- `text_embedding` Text Embedding Models
-
-- `rerank` Rerank Models
-
-- `speech2text` Speech-to-Text
-
-- `tts` Text-to-Speech
-
-- `moderation` Moderation
-
-Xinference supports LLM, Text Embedding, and Rerank. So we will start by writing xinference.yaml.
-
-```yaml
-provider: xinference #Define the vendor identifier
-label: # Vendor display name, supports both en_US (English) and zh_Hans (Simplified Chinese). If zh_Hans is not set, it will use en_US by default.
-  en_US: Xorbits Inference
-icon_small: # Small icon, refer to other vendors' icons stored in the _assets directory within the vendor implementation directory; follows the same language policy as the label
-  en_US: icon_s_en.svg
-icon_large: # Large icon
-  en_US: icon_l_en.svg
-help: # Help information
-  title:
-    en_US: How to deploy Xinference
-    zh_Hans: 如何部署 Xinference
-  url:
-    en_US: https://github.com/xorbitsai/inference
-supported_model_types: # Supported model types. Xinference supports LLM, Text Embedding, and Rerank
-- llm
-- text-embedding
-- rerank
-configurate_methods: # Since Xinference is a locally deployed vendor with no predefined models, users need to deploy whatever models they need according to Xinference documentation. Thus, it only supports custom models.
-- customizable-model
-provider_credential_schema:
-  credential_form_schemas:
-```
-
-
-Then, we need to determine what credentials are required to define a model in Xinference.
-
-- Since it supports three different types of models, we need to specify the model_type to denote the model type. Here is how we can define it:
-
-```yaml
-provider_credential_schema:
-  credential_form_schemas:
-  - variable: model_type
-    type: select
-    label:
-      en_US: Model type
-      zh_Hans: 模型类型
-    required: true
-    options:
-    - value: text-generation
-      label:
-        en_US: Language Model
-        zh_Hans: 语言模型
-    - value: embeddings
-      label:
-        en_US: Text Embedding
-    - value: reranking
-      label:
-        en_US: Rerank
-```
-
-- Next, each model has its own model_name, so we need to define that here:
-
-```yaml
-  - variable: model_name
-    type: text-input
-    label:
-      en_US: Model name
-      zh_Hans: 模型名称
-    required: true
-    placeholder:
-      zh_Hans: 填写模型名称
-      en_US: Input model name
-```
-
-- Specify the Xinference local deployment address:
-
-```yaml
-  - variable: server_url
-    label:
-      zh_Hans: 服务器URL
-      en_US: Server url
-    type: text-input
-    required: true
-    placeholder:
-      zh_Hans: 在此输入Xinference的服务器地址，如 https://example.com/xxx
-      en_US: Enter the url of your Xinference, for example https://example.com/xxx
-```
-
-- Each model has a unique model_uid, so we also need to define that here:
-
-```yaml
-  - variable: model_uid
-    label:
-      zh_Hans: 模型UID
-      en_US: Model uid
-    type: text-input
-    required: true
-    placeholder:
-      zh_Hans: 在此输入您的Model UID
-      en_US: Enter the model uid
-```
-
-Now, we have completed the basic definition of the vendor.
-
-### Writing the Model Code
-
-Next, let's take the `llm` type as an example and write `xinference.llm.llm.py`.
-
-In `llm.py`, create a Xinference LLM class, we name it `XinferenceAILargeLanguageModel` (this can be arbitrary), inheriting from the `__base.large_language_model.LargeLanguageModel` base class, and implement the following methods:
-
-- LLM Invocation
-
-Implement the core method for LLM invocation, supporting both stream and synchronous responses.
-
-```python
-def _invoke(self, model: str, credentials: dict,
-            prompt_messages: list[PromptMessage], model_parameters: dict,
-            tools: Optional[list[PromptMessageTool]] = None, stop: Optional[list[str]] = None,
-            stream: bool = True, user: Optional[str] = None) \
-        -> Union[LLMResult, Generator]:
-    """
-    Invoke large language model
-    
-    :param model: model name
-	:param credentials: model credentials
-	:param prompt_messages: prompt messages
-	:param model_parameters: model parameters
-	:param tools: tools for tool usage
-	:param stop: stop words
-	:param stream: is the response a stream
-	:param user: unique user id
-	:return: full response or stream response chunk generator result
-	"""
-```
-
-When implementing, ensure to use two functions to return data separately for synchronous and stream responses. This is important because Python treats functions containing the `yield` keyword as generator functions, mandating them to return `Generator` types. Here’s an example (note that the example uses simplified parameters; in real implementation, use the parameter list as defined above):
-
-```python
-def _invoke(self, stream: bool, **kwargs) \
-        -> Union[LLMResult, Generator]:
-    if stream:
-          return self._handle_stream_response(**kwargs)
-    return self._handle_sync_response(**kwargs)
-
-def _handle_stream_response(self, **kwargs) -> Generator:
-    for chunk in response:
-          yield chunk
-def _handle_sync_response(self, **kwargs) -> LLMResult:
-    return LLMResult(**response)
-```
-
-- Pre-compute Input Tokens
-
-If the model does not provide an interface for pre-computing tokens, you can return 0 directly.
-
-```python
-def get_num_tokens(self, model: str, credentials: dict, prompt_messages: list[PromptMessage],tools: Optional[list[PromptMessageTool]] = None) -> int:
-  """
-  Get number of tokens for given prompt messages
-
-  :param model: model name
-  :param credentials: model credentials
-  :param prompt_messages: prompt messages
-  :param tools: tools for tool usage
-  :return: token count
-  """
-```
-
-
-Sometimes, you might not want to return 0 directly. In such cases, you can use `self._get_num_tokens_by_gpt2(text: str)` to get pre-computed tokens. This method is provided by the `AIModel` base class, and it uses GPT2's Tokenizer for calculation. However, it should be noted that this is only a substitute and may not be fully accurate.
-
-- Model Credentials Validation
-
-Similar to vendor credentials validation, this method validates individual model credentials.
-
-```python
-def validate_credentials(self, model: str, credentials: dict) -> None:
-    """
-    Validate model credentials
-    
-    :param model: model name
-	:param credentials: model credentials
-	:return: None
-	"""
-```
-
-- Model Parameter Schema
-
-Unlike custom types, since the YAML file does not define which parameters a model supports, we need to dynamically generate the model parameter schema.
-
-For instance, Xinference supports `max_tokens`, `temperature`, and `top_p` parameters.
-
-However, some vendors may support different parameters for different models. For example, the `OpenLLM` vendor supports `top_k`, but not all models provided by this vendor support `top_k`. Let's say model A supports `top_k` but model B does not. In such cases, we need to dynamically generate the model parameter schema, as illustrated below:
-
-```python
-    def get_customizable_model_schema(self, model: str, credentials: dict) -> AIModelEntity | None:
-        """
-            used to define customizable model schema
-        """
-        rules = [
-            ParameterRule(
-                name='temperature', type=ParameterType.FLOAT,
-                use_template='temperature',
-                label=I18nObject(
-                    zh_Hans='温度', en_US='Temperature'
-                )
-            ),
-            ParameterRule(
-                name='top_p', type=ParameterType.FLOAT,
-                use_template='top_p',
-                label=I18nObject(
-                    zh_Hans='Top P', en_US='Top P'
-                )
-            ),
-            ParameterRule(
-                name='max_tokens', type=ParameterType.INT,
-                use_template='max_tokens',
-                min=1,
-                default=512,
-                label=I18nObject(
-                    zh_Hans='最大生成长度', en_US='Max Tokens'
-                )
-            )
-        ]
-
-        # if model is A, add top_k to rules
-        if model == 'A':
-            rules.append(
-                ParameterRule(
-                    name='top_k', type=ParameterType.INT,
-                    use_template='top_k',
-                    min=1,
-                    default=50,
-                    label=I18nObject(
-                        zh_Hans='Top K', en_US='Top K'
-                    )
-                )
-            )
-
-        """
-            some NOT IMPORTANT code here
-        """
-
-        entity = AIModelEntity(
-            model=model,
-            label=I18nObject(
-                en_US=model
-            ),
-            fetch_from=FetchFrom.CUSTOMIZABLE_MODEL,
-            model_type=model_type,
-            model_properties={ 
-                ModelPropertyKey.MODE:  ModelType.LLM,
-            },
-            parameter_rules=rules
-        )
-
-        return entity
-```
-
-- Exception Error Mapping
-
-When a model invocation error occurs, it should be mapped to the runtime's specified `InvokeError` type, enabling Dify to handle different errors appropriately.
-
-Runtime Errors:
-
-- `InvokeConnectionError` Connection error during invocation
-- `InvokeServerUnavailableError` Service provider unavailable
-- `InvokeRateLimitError` Rate limit reached
-- `InvokeAuthorizationError` Authorization failure
-- `InvokeBadRequestError` Invalid request parameters
-
-```python
-  @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
-      """
-```
-
-For interface method details, see: [Interfaces](./interfaces.md). For specific implementations, refer to: [llm.py](https://github.com/langgenius/dify-runtime/blob/main/lib/model_providers/anthropic/llm/llm.py).

api/core/model_runtime/docs/en_US/predefined_model_scale_out.md

@@ -1,173 +0,0 @@
-## Predefined Model Integration
-
-After completing the vendor integration, the next step is to integrate the models from the vendor.
-
-First, we need to determine the type of model to be integrated and create the corresponding model type `module` under the respective vendor's directory.
-
-Currently supported model types are:
-
-- `llm` Text Generation Model
-- `text_embedding` Text Embedding Model
-- `rerank` Rerank Model
-- `speech2text` Speech-to-Text
-- `tts` Text-to-Speech
-- `moderation` Moderation
-
-Continuing with `Anthropic` as an example, `Anthropic` only supports LLM, so create a `module` named `llm` under `model_providers.anthropic`.
-
-For predefined models, we first need to create a YAML file named after the model under the `llm` `module`, such as `claude-2.1.yaml`.
-
-### Prepare Model YAML
-
-```yaml
-model: claude-2.1  # Model identifier
-# Display name of the model, which can be set to en_US English or zh_Hans Chinese. If zh_Hans is not set, it will default to en_US.
-# This can also be omitted, in which case the model identifier will be used as the label
-label:
-  en_US: claude-2.1
-model_type: llm  # Model type, claude-2.1 is an LLM
-features:  # Supported features, agent-thought supports Agent reasoning, vision supports image understanding
-- agent-thought
-model_properties:  # Model properties
-  mode: chat  # LLM mode, complete for text completion models, chat for conversation models
-  context_size: 200000  # Maximum context size
-parameter_rules:  # Parameter rules for the model call; only LLM requires this
-- name: temperature  # Parameter variable name
-  # Five default configuration templates are provided: temperature/top_p/max_tokens/presence_penalty/frequency_penalty
-  # The template variable name can be set directly in use_template, which will use the default configuration in entities.defaults.PARAMETER_RULE_TEMPLATE
-  # Additional configuration parameters will override the default configuration if set
-  use_template: temperature
-- name: top_p
-  use_template: top_p
-- name: top_k
-  label:  # Display name of the parameter
-    zh_Hans: 取样数量
-    en_US: Top k
-  type: int  # Parameter type, supports float/int/string/boolean
-  help:  # Help information, describing the parameter's function
-    zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
-    en_US: Only sample from the top K options for each subsequent token.
-  required: false  # Whether the parameter is mandatory; can be omitted
-- name: max_tokens_to_sample
-  use_template: max_tokens
-  default: 4096  # Default value of the parameter
-  min: 1  # Minimum value of the parameter, applicable to float/int only
-  max: 4096  # Maximum value of the parameter, applicable to float/int only
-pricing:  # Pricing information
-  input: '8.00'  # Input unit price, i.e., prompt price
-  output: '24.00'  # Output unit price, i.e., response content price
-  unit: '0.000001'  # Price unit, meaning the above prices are per 100K
-  currency: USD  # Price currency
-```
-
-It is recommended to prepare all model configurations before starting the implementation of the model code.
-
-You can also refer to the YAML configuration information under the corresponding model type directories of other vendors in the `model_providers` directory. For the complete YAML rules, refer to: [Schema](schema.md#aimodelentity).
-
-### Implement the Model Call Code
-
-Next, create a Python file named `llm.py` under the `llm` `module` to write the implementation code.
-
-Create an Anthropic LLM class named `AnthropicLargeLanguageModel` (or any other name), inheriting from the `__base.large_language_model.LargeLanguageModel` base class, and implement the following methods:
-
-- LLM Call
-
-Implement the core method for calling the LLM, supporting both streaming and synchronous responses.
-
-```python
-  def _invoke(self, model: str, credentials: dict,
-              prompt_messages: list[PromptMessage], model_parameters: dict,
-              tools: Optional[list[PromptMessageTool]] = None, stop: Optional[list[str]] = None,
-              stream: bool = True, user: Optional[str] = None) \
-          -> Union[LLMResult, Generator]:
-      """
-      Invoke large language model
-  
-      :param model: model name
-      :param credentials: model credentials
-      :param prompt_messages: prompt messages
-      :param model_parameters: model parameters
-      :param tools: tools for tool calling
-      :param stop: stop words
-      :param stream: is stream response
-      :param user: unique user id
-      :return: full response or stream response chunk generator result
-      """
-```
-
-Ensure to use two functions for returning data, one for synchronous returns and the other for streaming returns, because Python identifies functions containing the `yield` keyword as generator functions, fixing the return type to `Generator`. Thus, synchronous and streaming returns need to be implemented separately, as shown below (note that the example uses simplified parameters, for actual implementation follow the above parameter list):
-
-```python
-  def _invoke(self, stream: bool, **kwargs) \
-          -> Union[LLMResult, Generator]:
-      if stream:
-            return self._handle_stream_response(**kwargs)
-      return self._handle_sync_response(**kwargs)
-
-  def _handle_stream_response(self, **kwargs) -> Generator:
-      for chunk in response:
-            yield chunk
-  def _handle_sync_response(self, **kwargs) -> LLMResult:
-      return LLMResult(**response)
-```
-
-- Pre-compute Input Tokens
-
-If the model does not provide an interface to precompute tokens, return 0 directly.
-
-```python
-  def get_num_tokens(self, model: str, credentials: dict, prompt_messages: list[PromptMessage],
-                     tools: Optional[list[PromptMessageTool]] = None) -> int:
-      """
-      Get number of tokens for given prompt messages
-
-      :param model: model name
-      :param credentials: model credentials
-      :param prompt_messages: prompt messages
-      :param tools: tools for tool calling
-      :return:
-      """
-```
-
-- Validate Model Credentials
-
-Similar to vendor credential validation, but specific to a single model.
-
-```python
-  def validate_credentials(self, model: str, credentials: dict) -> None:
-      """
-      Validate model credentials
-  
-      :param model: model name
-      :param credentials: model credentials
-      :return:
-      """
-```
-
-- Map Invoke Errors
-
-When a model call fails, map it to a specific `InvokeError` type as required by Runtime, allowing Dify to handle different errors accordingly.
-
-Runtime Errors:
-
-- `InvokeConnectionError` Connection error
-
-- `InvokeServerUnavailableError` Service provider unavailable
-- `InvokeRateLimitError` Rate limit reached
-- `InvokeAuthorizationError` Authorization failed
-- `InvokeBadRequestError` Parameter error
-
-```python
-  @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
-      """
-```
-
-For interface method explanations, see: [Interfaces](./interfaces.md). For detailed implementation, refer to: [llm.py](https://github.com/langgenius/dify-runtime/blob/main/lib/model_providers/anthropic/llm/llm.py).

api/core/model_runtime/docs/en_US/provider_scale_out.md

@@ -58,7 +58,7 @@ provider_credential_schema:  # Provider credential rules, as Anthropic only supp
       en_US: Enter your API URL
 ```
 
-You can also refer to the YAML configuration information under other provider directories in `model_providers`. The complete YAML rules are available at: [Schema](schema.md#provider).
+You can also refer to the YAML configuration information under other provider directories in `model_providers`. The complete YAML rules are available at: [Schema](schema.md#Provider).
 
 ### Implementing Provider Code
 

api/core/model_runtime/docs/zh_Hans/provider_scale_out.md

@@ -117,7 +117,7 @@ model_credential_schema:
       en_US: Enter your API Base
 ```
 
-也可以参考  `model_providers` 目录下其他供应商目录下的 YAML 配置信息，完整的 YAML 规则见：[Schema](schema.md#provider)。
+也可以参考  `model_providers` 目录下其他供应商目录下的 YAML 配置信息，完整的 YAML 规则见：[Schema](schema.md#Provider)。
 
 #### 实现供应商代码
 

api/core/model_runtime/entities/__init__.py

@@ -1,36 +0,0 @@
-from .llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta, LLMUsage
-from .message_entities import (
-    AssistantPromptMessage,
-    ImagePromptMessageContent,
-    PromptMessage,
-    PromptMessageContent,
-    PromptMessageContentType,
-    PromptMessageRole,
-    PromptMessageTool,
-    SystemPromptMessage,
-    TextPromptMessageContent,
-    ToolPromptMessage,
-    UserPromptMessage,
-)
-from .model_entities import ModelPropertyKey
-
-__all__ = [
-    "ImagePromptMessageContent",
-    "PromptMessage",
-    "PromptMessageRole",
-    "LLMUsage",
-    "ModelPropertyKey",
-    "AssistantPromptMessage",
-    "PromptMessage",
-    "PromptMessageContent",
-    "PromptMessageRole",
-    "SystemPromptMessage",
-    "TextPromptMessageContent",
-    "UserPromptMessage",
-    "PromptMessageTool",
-    "ToolPromptMessage",
-    "PromptMessageContentType",
-    "LLMResult",
-    "LLMResultChunk",
-    "LLMResultChunkDelta",
-]

api/core/model_runtime/entities/message_entities.py

@@ -79,7 +79,7 @@ class ImagePromptMessageContent(PromptMessageContent):
     Model class for image prompt message content.
     """
 
-    class DETAIL(str, Enum):
+    class DETAIL(Enum):
         LOW = "low"
         HIGH = "high"
 

api/core/model_runtime/model_providers/__base/large_language_model.py

@@ -1,4 +1,5 @@
 import logging
+import os
 import re
 import time
 from abc import abstractmethod
@@ -7,7 +8,6 @@ from typing import Optional, Union
 
 from pydantic import ConfigDict
 
-from configs import dify_config
 from core.model_runtime.callbacks.base_callback import Callback
 from core.model_runtime.callbacks.logging_callback import LoggingCallback
 from core.model_runtime.entities.llm_entities import LLMMode, LLMResult, LLMResultChunk, LLMResultChunkDelta, LLMUsage
@@ -77,7 +77,7 @@ class LargeLanguageModel(AIModel):
 
         callbacks = callbacks or []
 
-        if dify_config.DEBUG:
+        if bool(os.environ.get("DEBUG", "False").lower() == "true"):
             callbacks.append(LoggingCallback())
 
         # trigger before invoke callbacks