bump to 0.6.13 (#6078)

Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>

.devcontainer/post_create_command.sh

@@ -1,6 +1,7 @@
 #!/bin/bash
 
 cd web && npm install
+pipx install poetry
 
 echo 'alias start-api="cd /workspaces/dify/api && flask run --host 0.0.0.0 --port=5001 --debug"' >> ~/.bashrc
 echo 'alias start-worker="cd /workspaces/dify/api && celery -A app.celery worker -P gevent -c 1 --loglevel INFO -Q dataset,generation,mail,ops_trace,app_deletion"' >> ~/.bashrc

.github/pull_request_template.md

@@ -1,13 +1,21 @@
+# Checklist:
+
+> [!IMPORTANT]  
+> Please review the checklist below before submitting your pull request.
+
+- [ ] Please open an issue before creating a PR or link to an existing issue
+- [ ] I have performed a self-review of my own code
+- [ ] I have commented my code, particularly in hard-to-understand areas
+- [ ] I ran `dev/reformat`(backend) and `cd web && npx lint-staged`(frontend) to appease the lint gods
+
 # Description
 
-Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.
+Describe the big picture of your changes here to communicate to the maintainers why we should accept this pull request. If it fixes a bug or resolves a feature request, be sure to link to that issue. Close issue syntax: `Fixes #<issue number>`, see [documentation](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword) for more details.
 
-Fixes # (issue)
+Fixes 
 
 ## Type of Change
 
-Please delete options that are not relevant.
-
 - [ ] Bug fix (non-breaking change which fixes an issue)
 - [ ] New feature (non-breaking change which adds functionality)
 - [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)
@@ -15,18 +23,12 @@ Please delete options that are not relevant.
 - [ ] Improvement, including but not limited to code refactoring, performance optimization, and UI/UX improvement
 - [ ] Dependency upgrade
 
-# How Has This Been Tested?
+# Testing Instructions
 
 Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration
 
-- [ ] TODO
+- [ ] Test A
+- [ ] Test B
+
 
-# Suggested Checklist:
 
-- [ ] I have performed a self-review of my own code
-- [ ] I have commented my code, particularly in hard-to-understand areas
-- [ ] My changes generate no new warnings
-- [ ] I ran `dev/reformat`(backend) and `cd web && npx lint-staged`(frontend) to appease the lint gods
-- [ ] `optional` I have made corresponding changes to the documentation 
-- [ ] `optional` I have added tests that prove my fix is effective or that my feature works
-- [ ] `optional` New and existing unit tests pass locally with my changes

.github/workflows/build-push.yml

@@ -48,18 +48,18 @@ jobs:
           platform=${{ matrix.platform }}
           echo "PLATFORM_PAIR=${platform//\//-}" >> $GITHUB_ENV
 
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
       - name: Login to Docker Hub
         uses: docker/login-action@v2
         with:
           username: ${{ env.DOCKERHUB_USER }}
           password: ${{ env.DOCKERHUB_TOKEN }}
 
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
       - name: Extract metadata for Docker
         id: meta
         uses: docker/metadata-action@v5

.gitignore

@@ -174,3 +174,5 @@ sdks/python-client/dify_client.egg-info
 .vscode/*
 !.vscode/launch.json
 pyrightconfig.json
+
+.idea/

.vscode/launch.json

@@ -13,7 +13,6 @@
             "jinja": true,
             "env": {
                 "FLASK_APP": "app.py",
-                "FLASK_DEBUG": "1",
                 "GEVENT_SUPPORT": "True"
             },
             "args": [

api/Dockerfile

@@ -5,8 +5,7 @@ WORKDIR /app/api
 
 # Install Poetry
 ENV POETRY_VERSION=1.8.3
-RUN pip install --no-cache-dir --upgrade pip && \
-    pip install --no-cache-dir --upgrade poetry==${POETRY_VERSION}
+RUN pip install --no-cache-dir poetry==${POETRY_VERSION}
 
 # Configure Poetry
 ENV POETRY_CACHE_DIR=/tmp/poetry_cache

api/README.md

@@ -11,6 +11,7 @@
 
    ```bash
    cd ../docker
+   cp middleware.env.example middleware.env
    docker compose -f docker-compose.middleware.yaml -p dify up -d
    cd ../api
    ```

api/app.py

@@ -1,8 +1,8 @@
 import os
 
-from configs.app_config import DifyConfig
+from configs import dify_config
 
-if not os.environ.get("DEBUG") or os.environ.get("DEBUG", "false").lower() != 'true':
+if os.environ.get("DEBUG", "false").lower() != 'true':
     from gevent import monkey
 
     monkey.patch_all()
@@ -43,6 +43,8 @@ from extensions import (
 from extensions.ext_database import db
 from extensions.ext_login import login_manager
 from libs.passport import PassportService
+
+# TODO: Find a way to avoid importing models here
 from models import account, dataset, model, source, task, tool, tools, web
 from services.account_service import AccountService
 
@@ -81,7 +83,7 @@ def create_flask_app_with_configs() -> Flask:
     with configs loaded from .env file
     """
     dify_app = DifyApp(__name__)
-    dify_app.config.from_mapping(DifyConfig().model_dump())
+    dify_app.config.from_mapping(dify_config.model_dump())
 
     # populate configs into system environment variables
     for key, value in dify_app.config.items():

api/commands.py

@@ -8,6 +8,7 @@ import click
 from flask import current_app
 from werkzeug.exceptions import NotFound
 
+from configs import dify_config
 from constants.languages import languages
 from core.rag.datasource.vdb.vector_factory import Vector
 from core.rag.datasource.vdb.vector_type import VectorType
@@ -112,7 +113,7 @@ def reset_encrypt_key_pair():
     After the reset, all LLM credentials will become invalid, requiring re-entry.
     Only support SELF_HOSTED mode.
     """
-    if current_app.config['EDITION'] != 'SELF_HOSTED':
+    if dify_config.EDITION != 'SELF_HOSTED':
         click.echo(click.style('Sorry, only support SELF_HOSTED mode.', fg='red'))
         return
 

api/configs/__init__.py

@@ -0,0 +1,3 @@
+from .app_config import DifyConfig
+
+dify_config = DifyConfig()

api/configs/app_config.py

@@ -1,4 +1,5 @@
-from pydantic_settings import BaseSettings, SettingsConfigDict
+from pydantic import Field, computed_field
+from pydantic_settings import SettingsConfigDict
 
 from configs.deploy import DeploymentConfig
 from configs.enterprise import EnterpriseFeatureConfig
@@ -8,13 +9,7 @@ from configs.middleware import MiddlewareConfig
 from configs.packaging import PackagingInfo
 
 
-# TODO: Both `BaseModel` and `BaseSettings` has `model_config` attribute but they are in different types.
-# This inheritance is depends on the order of the classes.
-# It is better to use `BaseSettings` as the base class.
 class DifyConfig(
-    # based on pydantic-settings
-    BaseSettings,
-
     # Packaging info
     PackagingInfo,
 
@@ -34,12 +29,39 @@ class DifyConfig(
     # **Before using, please contact business@dify.ai by email to inquire about licensing matters.**
     EnterpriseFeatureConfig,
 ):
+    DEBUG: bool = Field(default=False, description='whether to enable debug mode.')
 
     model_config = SettingsConfigDict(
         # read from dotenv format config file
         env_file='.env',
         env_file_encoding='utf-8',
+        frozen=True,
 
         # ignore extra attributes
         extra='ignore',
     )
+
+    CODE_MAX_NUMBER: int = 9223372036854775807
+    CODE_MIN_NUMBER: int = -9223372036854775808
+    CODE_MAX_STRING_LENGTH: int = 80000
+    CODE_MAX_STRING_ARRAY_LENGTH: int = 30
+    CODE_MAX_OBJECT_ARRAY_LENGTH: int = 30
+    CODE_MAX_NUMBER_ARRAY_LENGTH: int = 1000
+
+    HTTP_REQUEST_MAX_CONNECT_TIMEOUT: int = 300
+    HTTP_REQUEST_MAX_READ_TIMEOUT: int = 600
+    HTTP_REQUEST_MAX_WRITE_TIMEOUT: int = 600
+    HTTP_REQUEST_NODE_MAX_BINARY_SIZE: int = 1024 * 1024 * 10
+
+    @computed_field
+    def HTTP_REQUEST_NODE_READABLE_MAX_BINARY_SIZE(self) -> str:
+        return f'{self.HTTP_REQUEST_NODE_MAX_BINARY_SIZE / 1024 / 1024:.2f}MB'
+
+    HTTP_REQUEST_NODE_MAX_TEXT_SIZE: int = 1024 * 1024
+
+    @computed_field
+    def HTTP_REQUEST_NODE_READABLE_MAX_TEXT_SIZE(self) -> str:
+        return f'{self.HTTP_REQUEST_NODE_MAX_TEXT_SIZE / 1024 / 1024:.2f}MB'
+
+    SSRF_PROXY_HTTP_URL: str | None = None
+    SSRF_PROXY_HTTPS_URL: str | None = None

api/configs/deploy/__init__.py

@@ -1,7 +1,8 @@
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class DeploymentConfig(BaseModel):
+class DeploymentConfig(BaseSettings):
     """
     Deployment configs
     """

api/configs/enterprise/__init__.py

@@ -1,7 +1,8 @@
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class EnterpriseFeatureConfig(BaseModel):
+class EnterpriseFeatureConfig(BaseSettings):
     """
     Enterprise feature configs.
     **Before using, please contact business@dify.ai by email to inquire about licensing matters.**

api/configs/extra/__init__.py

@@ -1,5 +1,3 @@
-from pydantic import BaseModel
-
 from configs.extra.notion_config import NotionConfig
 from configs.extra.sentry_config import SentryConfig
 

api/configs/extra/notion_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class NotionConfig(BaseModel):
+class NotionConfig(BaseSettings):
     """
     Notion integration configs
     """

api/configs/extra/sentry_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, NonNegativeFloat
+from pydantic import Field, NonNegativeFloat
+from pydantic_settings import BaseSettings
 
 
-class SentryConfig(BaseModel):
+class SentryConfig(BaseSettings):
     """
     Sentry configs
     """

api/configs/feature/__init__.py

@@ -1,11 +1,12 @@
 from typing import Optional
 
-from pydantic import AliasChoices, BaseModel, Field, NonNegativeInt, PositiveInt, computed_field
+from pydantic import AliasChoices, Field, NonNegativeInt, PositiveInt, computed_field
+from pydantic_settings import BaseSettings
 
 from configs.feature.hosted_service import HostedServiceConfig
 
 
-class SecurityConfig(BaseModel):
+class SecurityConfig(BaseSettings):
     """
     Secret Key configs
     """
@@ -17,8 +18,12 @@ class SecurityConfig(BaseModel):
         default=None,
     )
 
+    RESET_PASSWORD_TOKEN_EXPIRY_HOURS: PositiveInt = Field(
+        description='Expiry time in hours for reset token',
+        default=24,
+    )
 
-class AppExecutionConfig(BaseModel):
+class AppExecutionConfig(BaseSettings):
     """
     App Execution configs
     """
@@ -28,7 +33,7 @@ class AppExecutionConfig(BaseModel):
     )
 
 
-class CodeExecutionSandboxConfig(BaseModel):
+class CodeExecutionSandboxConfig(BaseSettings):
     """
     Code Execution Sandbox configs
     """
@@ -43,36 +48,36 @@ class CodeExecutionSandboxConfig(BaseModel):
     )
 
 
-class EndpointConfig(BaseModel):
+class EndpointConfig(BaseSettings):
     """
     Module URL configs
     """
     CONSOLE_API_URL: str = Field(
         description='The backend URL prefix of the console API.'
                     'used to concatenate the login authorization callback or notion integration callback.',
-        default='https://cloud.dify.ai',
+        default='',
     )
 
     CONSOLE_WEB_URL: str = Field(
         description='The front-end URL prefix of the console web.'
                     'used to concatenate some front-end addresses and for CORS configuration use.',
-        default='https://cloud.dify.ai',
+        default='',
     )
 
     SERVICE_API_URL: str = Field(
         description='Service API Url prefix.'
                     'used to display Service API Base Url to the front-end.',
-        default='https://api.dify.ai',
+        default='',
     )
 
     APP_WEB_URL: str = Field(
         description='WebApp Url prefix.'
                     'used to display WebAPP API Base Url to the front-end.',
-        default='https://udify.app',
+        default='',
     )
 
 
-class FileAccessConfig(BaseModel):
+class FileAccessConfig(BaseSettings):
     """
     File Access configs
     """
@@ -82,7 +87,7 @@ class FileAccessConfig(BaseModel):
                     'Url is signed and has expiration time.',
         validation_alias=AliasChoices('FILES_URL', 'CONSOLE_API_URL'),
         alias_priority=1,
-        default='https://cloud.dify.ai',
+        default='',
     )
 
     FILES_ACCESS_TIMEOUT: int = Field(
@@ -91,7 +96,7 @@ class FileAccessConfig(BaseModel):
     )
 
 
-class FileUploadConfig(BaseModel):
+class FileUploadConfig(BaseSettings):
     """
     File Uploading configs
     """
@@ -116,7 +121,7 @@ class FileUploadConfig(BaseModel):
     )
 
 
-class HttpConfig(BaseModel):
+class HttpConfig(BaseSettings):
     """
     HTTP configs
     """
@@ -148,7 +153,7 @@ class HttpConfig(BaseModel):
         return self.inner_WEB_API_CORS_ALLOW_ORIGINS.split(',')
 
 
-class InnerAPIConfig(BaseModel):
+class InnerAPIConfig(BaseSettings):
     """
     Inner API configs
     """
@@ -163,7 +168,7 @@ class InnerAPIConfig(BaseModel):
     )
 
 
-class LoggingConfig(BaseModel):
+class LoggingConfig(BaseSettings):
     """
     Logging configs
     """
@@ -195,7 +200,7 @@ class LoggingConfig(BaseModel):
     )
 
 
-class ModelLoadBalanceConfig(BaseModel):
+class ModelLoadBalanceConfig(BaseSettings):
     """
     Model load balance configs
     """
@@ -205,7 +210,7 @@ class ModelLoadBalanceConfig(BaseModel):
     )
 
 
-class BillingConfig(BaseModel):
+class BillingConfig(BaseSettings):
     """
     Platform Billing Configurations
     """
@@ -215,7 +220,7 @@ class BillingConfig(BaseModel):
     )
 
 
-class UpdateConfig(BaseModel):
+class UpdateConfig(BaseSettings):
     """
     Update configs
     """
@@ -225,7 +230,7 @@ class UpdateConfig(BaseModel):
     )
 
 
-class WorkflowConfig(BaseModel):
+class WorkflowConfig(BaseSettings):
     """
     Workflow feature configs
     """
@@ -246,7 +251,7 @@ class WorkflowConfig(BaseModel):
     )
 
 
-class OAuthConfig(BaseModel):
+class OAuthConfig(BaseSettings):
     """
     oauth configs
     """
@@ -276,7 +281,7 @@ class OAuthConfig(BaseModel):
     )
 
 
-class ModerationConfig(BaseModel):
+class ModerationConfig(BaseSettings):
     """
     Moderation in app configs.
     """
@@ -288,7 +293,7 @@ class ModerationConfig(BaseModel):
     )
 
 
-class ToolConfig(BaseModel):
+class ToolConfig(BaseSettings):
     """
     Tool configs
     """
@@ -299,7 +304,7 @@ class ToolConfig(BaseModel):
     )
 
 
-class MailConfig(BaseModel):
+class MailConfig(BaseSettings):
     """
     Mail Configurations
     """
@@ -355,7 +360,7 @@ class MailConfig(BaseModel):
     )
 
 
-class RagEtlConfig(BaseModel):
+class RagEtlConfig(BaseSettings):
     """
     RAG ETL Configurations.
     """
@@ -381,7 +386,7 @@ class RagEtlConfig(BaseModel):
     )
 
 
-class DataSetConfig(BaseModel):
+class DataSetConfig(BaseSettings):
     """
     Dataset configs
     """
@@ -392,7 +397,7 @@ class DataSetConfig(BaseModel):
     )
 
 
-class WorkspaceConfig(BaseModel):
+class WorkspaceConfig(BaseSettings):
     """
     Workspace configs
     """
@@ -403,7 +408,7 @@ class WorkspaceConfig(BaseModel):
     )
 
 
-class IndexingConfig(BaseModel):
+class IndexingConfig(BaseSettings):
     """
     Indexing configs.
     """
@@ -414,7 +419,7 @@ class IndexingConfig(BaseModel):
     )
 
 
-class ImageFormatConfig(BaseModel):
+class ImageFormatConfig(BaseSettings):
     MULTIMODAL_SEND_IMAGE_FORMAT: str = Field(
         description='multi model send image format, support base64, url, default is base64',
         default='base64',

api/configs/feature/hosted_service/__init__.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, NonNegativeInt
+from pydantic import Field, NonNegativeInt
+from pydantic_settings import BaseSettings
 
 
-class HostedOpenAiConfig(BaseModel):
+class HostedOpenAiConfig(BaseSettings):
     """
     Hosted OpenAI service config
     """
@@ -68,7 +69,7 @@ class HostedOpenAiConfig(BaseModel):
     )
 
 
-class HostedAzureOpenAiConfig(BaseModel):
+class HostedAzureOpenAiConfig(BaseSettings):
     """
     Hosted OpenAI service config
     """
@@ -94,7 +95,7 @@ class HostedAzureOpenAiConfig(BaseModel):
     )
 
 
-class HostedAnthropicConfig(BaseModel):
+class HostedAnthropicConfig(BaseSettings):
     """
     Hosted Azure OpenAI service config
     """
@@ -125,7 +126,7 @@ class HostedAnthropicConfig(BaseModel):
     )
 
 
-class HostedMinmaxConfig(BaseModel):
+class HostedMinmaxConfig(BaseSettings):
     """
     Hosted Minmax service config
     """
@@ -136,7 +137,7 @@ class HostedMinmaxConfig(BaseModel):
     )
 
 
-class HostedSparkConfig(BaseModel):
+class HostedSparkConfig(BaseSettings):
     """
     Hosted Spark service config
     """
@@ -147,7 +148,7 @@ class HostedSparkConfig(BaseModel):
     )
 
 
-class HostedZhipuAIConfig(BaseModel):
+class HostedZhipuAIConfig(BaseSettings):
     """
     Hosted Minmax service config
     """
@@ -158,7 +159,7 @@ class HostedZhipuAIConfig(BaseModel):
     )
 
 
-class HostedModerationConfig(BaseModel):
+class HostedModerationConfig(BaseSettings):
     """
     Hosted Moderation service config
     """
@@ -174,7 +175,7 @@ class HostedModerationConfig(BaseModel):
     )
 
 
-class HostedFetchAppTemplateConfig(BaseModel):
+class HostedFetchAppTemplateConfig(BaseSettings):
     """
     Hosted Moderation service config
     """

api/configs/middleware/__init__.py

@@ -1,6 +1,7 @@
 from typing import Any, Optional
 
-from pydantic import BaseModel, Field, NonNegativeInt, PositiveInt, computed_field
+from pydantic import Field, NonNegativeInt, PositiveInt, computed_field
+from pydantic_settings import BaseSettings
 
 from configs.middleware.cache.redis_config import RedisConfig
 from configs.middleware.storage.aliyun_oss_storage_config import AliyunOSSStorageConfig
@@ -22,7 +23,7 @@ from configs.middleware.vdb.tidb_vector_config import TiDBVectorConfig
 from configs.middleware.vdb.weaviate_config import WeaviateConfig
 
 
-class StorageConfig(BaseModel):
+class StorageConfig(BaseSettings):
     STORAGE_TYPE: str = Field(
         description='storage type,'
                     ' default to `local`,'
@@ -36,14 +37,14 @@ class StorageConfig(BaseModel):
     )
 
 
-class VectorStoreConfig(BaseModel):
+class VectorStoreConfig(BaseSettings):
     VECTOR_STORE: Optional[str] = Field(
         description='vector store type',
         default=None,
     )
 
 
-class KeywordStoreConfig(BaseModel):
+class KeywordStoreConfig(BaseSettings):
     KEYWORD_STORE: str = Field(
         description='keyword store type',
         default='jieba',
@@ -81,6 +82,11 @@ class DatabaseConfig:
         default='',
     )
 
+    DB_EXTRAS: str = Field(
+        description='db extras options. Example: keepalives_idle=60&keepalives=1',
+        default='',
+    )
+
     SQLALCHEMY_DATABASE_URI_SCHEME: str = Field(
         description='db uri scheme',
         default='postgresql',
@@ -89,7 +95,12 @@ class DatabaseConfig:
     @computed_field
     @property
     def SQLALCHEMY_DATABASE_URI(self) -> str:
-        db_extras = f"?client_encoding={self.DB_CHARSET}" if self.DB_CHARSET else ""
+        db_extras = (
+            f"{self.DB_EXTRAS}&client_encoding={self.DB_CHARSET}"
+            if self.DB_CHARSET
+            else self.DB_EXTRAS
+        ).strip("&")
+        db_extras = f"?{db_extras}" if db_extras else ""
         return (f"{self.SQLALCHEMY_DATABASE_URI_SCHEME}://"
                 f"{self.DB_USERNAME}:{self.DB_PASSWORD}@{self.DB_HOST}:{self.DB_PORT}/{self.DB_DATABASE}"
                 f"{db_extras}")
@@ -114,7 +125,7 @@ class DatabaseConfig:
         default=False,
     )
 
-    SQLALCHEMY_ECHO: bool = Field(
+    SQLALCHEMY_ECHO: bool | str = Field(
         description='whether to enable SqlAlchemy echo',
         default=False,
     )

api/configs/middleware/cache/redis_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, NonNegativeInt, PositiveInt
+from pydantic import Field, NonNegativeInt, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class RedisConfig(BaseModel):
+class RedisConfig(BaseSettings):
     """
     Redis configs
     """

api/configs/middleware/storage/aliyun_oss_storage_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class AliyunOSSStorageConfig(BaseModel):
+class AliyunOSSStorageConfig(BaseSettings):
     """
     Aliyun storage configs
     """

api/configs/middleware/storage/amazon_s3_storage_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class S3StorageConfig(BaseModel):
+class S3StorageConfig(BaseSettings):
     """
     S3 storage configs
     """

api/configs/middleware/storage/azure_blob_storage_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class AzureBlobStorageConfig(BaseModel):
+class AzureBlobStorageConfig(BaseSettings):
     """
     Azure Blob storage configs
     """

api/configs/middleware/storage/google_cloud_storage_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class GoogleCloudStorageConfig(BaseModel):
+class GoogleCloudStorageConfig(BaseSettings):
     """
     Google Cloud storage configs
     """

api/configs/middleware/storage/oci_storage_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class OCIStorageConfig(BaseModel):
+class OCIStorageConfig(BaseSettings):
     """
     OCI storage configs
     """

api/configs/middleware/storage/tencent_cos_storage_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class TencentCloudCOSStorageConfig(BaseModel):
+class TencentCloudCOSStorageConfig(BaseSettings):
     """
     Tencent Cloud COS storage configs
     """

api/configs/middleware/vdb/chroma_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class ChromaConfig(BaseModel):
+class ChromaConfig(BaseSettings):
     """
     Chroma configs
     """

api/configs/middleware/vdb/milvus_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class MilvusConfig(BaseModel):
+class MilvusConfig(BaseSettings):
     """
     Milvus configs
     """

api/configs/middleware/vdb/opensearch_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class OpenSearchConfig(BaseModel):
+class OpenSearchConfig(BaseSettings):
     """
     OpenSearch configs
     """

api/configs/middleware/vdb/oracle_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class OracleConfig(BaseModel):
+class OracleConfig(BaseSettings):
     """
     ORACLE configs
     """

api/configs/middleware/vdb/pgvector_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class PGVectorConfig(BaseModel):
+class PGVectorConfig(BaseSettings):
     """
     PGVector configs
     """

api/configs/middleware/vdb/pgvectors_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class PGVectoRSConfig(BaseModel):
+class PGVectoRSConfig(BaseSettings):
     """
     PGVectoRS configs
     """

api/configs/middleware/vdb/qdrant_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, NonNegativeInt, PositiveInt
+from pydantic import Field, NonNegativeInt, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class QdrantConfig(BaseModel):
+class QdrantConfig(BaseSettings):
     """
     Qdrant configs
     """

api/configs/middleware/vdb/relyt_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class RelytConfig(BaseModel):
+class RelytConfig(BaseSettings):
     """
     Relyt configs
     """

api/configs/middleware/vdb/tencent_vector_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, NonNegativeInt, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class TencentVectorDBConfig(BaseModel):
+class TencentVectorDBConfig(BaseSettings):
     """
     Tencent Vector configs
     """
@@ -24,7 +25,7 @@ class TencentVectorDBConfig(BaseModel):
     )
 
     TENCENT_VECTOR_DB_USERNAME: Optional[str] = Field(
-        description='Tencent Vector password',
+        description='Tencent Vector username',
         default=None,
     )
 
@@ -38,7 +39,12 @@ class TencentVectorDBConfig(BaseModel):
         default=1,
     )
 
-    TENCENT_VECTOR_DB_REPLICAS: PositiveInt = Field(
+    TENCENT_VECTOR_DB_REPLICAS: NonNegativeInt = Field(
         description='Tencent Vector replicas',
         default=2,
     )
+
+    TENCENT_VECTOR_DB_DATABASE: Optional[str] = Field(
+        description='Tencent Vector Database',
+        default=None,
+    )

api/configs/middleware/vdb/tidb_vector_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class TiDBVectorConfig(BaseModel):
+class TiDBVectorConfig(BaseSettings):
     """
     TiDB Vector configs
     """

api/configs/middleware/vdb/weaviate_config.py

@@ -1,9 +1,10 @@
 from typing import Optional
 
-from pydantic import BaseModel, Field, PositiveInt
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
 
 
-class WeaviateConfig(BaseModel):
+class WeaviateConfig(BaseSettings):
     """
     Weaviate configs
     """

api/configs/packaging/__init__.py

@@ -1,14 +1,15 @@
-from pydantic import BaseModel, Field
+from pydantic import Field
+from pydantic_settings import BaseSettings
 
 
-class PackagingInfo(BaseModel):
+class PackagingInfo(BaseSettings):
     """
     Packaging build information
     """
 
     CURRENT_VERSION: str = Field(
         description='Dify version',
-        default='0.6.12-fix1',
+        default='0.6.13',
     )
 
     COMMIT_SHA: str = Field(

api/constants/languages.py

@@ -14,7 +14,7 @@ language_timezone_mapping = {
     'vi-VN': 'Asia/Ho_Chi_Minh',
     'ro-RO': 'Europe/Bucharest',
     'pl-PL': 'Europe/Warsaw',
-    'hi-IN': 'Asia/Kolkata'
+    'hi-IN': 'Asia/Kolkata',
 }
 
 languages = list(language_timezone_mapping.keys())

api/controllers/console/__init__.py

@@ -30,7 +30,7 @@ from .app import (
 )
 
 # Import auth controllers
-from .auth import activate, data_source_bearer_auth, data_source_oauth, login, oauth
+from .auth import activate, data_source_bearer_auth, data_source_oauth, forgot_password, login, oauth
 
 # Import billing controllers
 from .billing import billing

api/controllers/console/auth/data_source_oauth.py

@@ -6,6 +6,7 @@ from flask_login import current_user
 from flask_restful import Resource
 from werkzeug.exceptions import Forbidden
 
+from configs import dify_config
 from controllers.console import api
 from libs.login import login_required
 from libs.oauth_data_source import NotionOAuth
@@ -16,11 +17,11 @@ from ..wraps import account_initialization_required
 
 def get_oauth_providers():
     with current_app.app_context():
-        notion_oauth = NotionOAuth(client_id=current_app.config.get('NOTION_CLIENT_ID'),
-                                   client_secret=current_app.config.get(
-                                       'NOTION_CLIENT_SECRET'),
-                                   redirect_uri=current_app.config.get(
-                                       'CONSOLE_API_URL') + '/console/api/oauth/data-source/callback/notion')
+        if not dify_config.NOTION_CLIENT_ID or not dify_config.NOTION_CLIENT_SECRET:
+            return {}
+        notion_oauth = NotionOAuth(client_id=dify_config.NOTION_CLIENT_ID,
+                                   client_secret=dify_config.NOTION_CLIENT_SECRET,
+                                   redirect_uri=dify_config.CONSOLE_API_URL + '/console/api/oauth/data-source/callback/notion')
 
         OAUTH_PROVIDERS = {
             'notion': notion_oauth
@@ -39,8 +40,10 @@ class OAuthDataSource(Resource):
             print(vars(oauth_provider))
         if not oauth_provider:
             return {'error': 'Invalid provider'}, 400
-        if current_app.config.get('NOTION_INTEGRATION_TYPE') == 'internal':
-            internal_secret = current_app.config.get('NOTION_INTERNAL_SECRET')
+        if dify_config.NOTION_INTEGRATION_TYPE == 'internal':
+            internal_secret = dify_config.NOTION_INTERNAL_SECRET
+            if not internal_secret:
+                return {'error': 'Internal secret is not set'},
             oauth_provider.save_internal_access_token(internal_secret)
             return { 'data': '' }
         else:
@@ -60,13 +63,13 @@ class OAuthDataSourceCallback(Resource):
         if 'code' in request.args:
             code = request.args.get('code')
 
-            return redirect(f'{current_app.config.get("CONSOLE_WEB_URL")}?type=notion&code={code}')
+            return redirect(f'{dify_config.CONSOLE_WEB_URL}?type=notion&code={code}')
         elif 'error' in request.args:
             error = request.args.get('error')
 
-            return redirect(f'{current_app.config.get("CONSOLE_WEB_URL")}?type=notion&error={error}')
+            return redirect(f'{dify_config.CONSOLE_WEB_URL}?type=notion&error={error}')
         else:
-            return redirect(f'{current_app.config.get("CONSOLE_WEB_URL")}?type=notion&error=Access denied')
+            return redirect(f'{dify_config.CONSOLE_WEB_URL}?type=notion&error=Access denied')
         
 
 class OAuthDataSourceBinding(Resource):

api/controllers/console/auth/error.py

@@ -5,3 +5,28 @@ class ApiKeyAuthFailedError(BaseHTTPException):
     error_code = 'auth_failed'
     description = "{message}"
     code = 500
+
+
+class InvalidEmailError(BaseHTTPException):
+    error_code = 'invalid_email'
+    description = "The email address is not valid."
+    code = 400
+
+
+class PasswordMismatchError(BaseHTTPException):
+    error_code = 'password_mismatch'
+    description = "The passwords do not match."
+    code = 400
+
+
+class InvalidTokenError(BaseHTTPException):
+    error_code = 'invalid_or_expired_token'
+    description = "The token is invalid or has expired."
+    code = 400
+
+
+class PasswordResetRateLimitExceededError(BaseHTTPException):
+    error_code = 'password_reset_rate_limit_exceeded'
+    description = "Password reset rate limit exceeded. Try again later."
+    code = 429
+

api/controllers/console/auth/forgot_password.py

@@ -0,0 +1,107 @@
+import base64
+import logging
+import secrets
+
+from flask_restful import Resource, reqparse
+
+from controllers.console import api
+from controllers.console.auth.error import (
+    InvalidEmailError,
+    InvalidTokenError,
+    PasswordMismatchError,
+    PasswordResetRateLimitExceededError,
+)
+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.account import Account
+from services.account_service import AccountService
+from services.errors.account import RateLimitExceededError
+
+
+class ForgotPasswordSendEmailApi(Resource):
+
+    @setup_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('email', type=str, required=True, location='json')
+        args = parser.parse_args()
+
+        email = args['email']
+
+        if not email_validate(email):
+            raise InvalidEmailError()
+
+        account = Account.query.filter_by(email=email).first()
+
+        if account:
+            try:
+                AccountService.send_reset_password_email(account=account)
+            except RateLimitExceededError:
+                logging.warning(f"Rate limit exceeded for email: {account.email}")
+                raise PasswordResetRateLimitExceededError()
+        else:
+            # Return success to avoid revealing email registration status
+            logging.warning(f"Attempt to reset password for unregistered email: {email}")
+
+        return {"result": "success"}
+
+
+class ForgotPasswordCheckApi(Resource):
+
+    @setup_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('token', type=str, required=True, nullable=False, location='json')
+        args = parser.parse_args()
+        token = args['token']
+
+        reset_data = AccountService.get_reset_password_data(token)
+
+        if reset_data is None:
+            return {'is_valid': False, 'email': None}
+        return {'is_valid': True, 'email': reset_data.get('email')}
+
+
+class ForgotPasswordResetApi(Resource):
+
+    @setup_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('token', type=str, required=True, nullable=False, location='json')
+        parser.add_argument('new_password', type=valid_password, required=True, nullable=False, location='json')
+        parser.add_argument('password_confirm', type=valid_password, required=True, nullable=False, location='json')
+        args = parser.parse_args()
+
+        new_password = args['new_password']
+        password_confirm = args['password_confirm']
+
+        if str(new_password).strip() != str(password_confirm).strip():
+            raise PasswordMismatchError()
+
+        token = args['token']
+        reset_data = AccountService.get_reset_password_data(token)
+
+        if reset_data is None:
+            raise InvalidTokenError()
+
+        AccountService.revoke_reset_password_token(token)
+
+        salt = secrets.token_bytes(16)
+        base64_salt = base64.b64encode(salt).decode()
+
+        password_hashed = hash_password(new_password, salt)
+        base64_password_hashed = base64.b64encode(password_hashed).decode()
+
+        account = Account.query.filter_by(email=reset_data.get('email')).first()
+        account.password = base64_password_hashed
+        account.password_salt = base64_salt
+        db.session.commit()
+
+        return {'result': 'success'}
+
+
+api.add_resource(ForgotPasswordSendEmailApi, '/forgot-password')
+api.add_resource(ForgotPasswordCheckApi, '/forgot-password/validity')
+api.add_resource(ForgotPasswordResetApi, '/forgot-password/resets')

api/controllers/console/auth/login.py

@@ -1,7 +1,7 @@
 from typing import cast
 
 import flask_login
-from flask import current_app, request
+from flask import request
 from flask_restful import Resource, reqparse
 
 import services
@@ -56,14 +56,14 @@ class LogoutApi(Resource):
 class ResetPasswordApi(Resource):
     @setup_required
     def get(self):
-        parser = reqparse.RequestParser()
-        parser.add_argument('email', type=email, required=True, location='json')
-        args = parser.parse_args()
+        # parser = reqparse.RequestParser()
+        # parser.add_argument('email', type=email, required=True, location='json')
+        # args = parser.parse_args()
 
         # import mailchimp_transactional as MailchimpTransactional
         # from mailchimp_transactional.api_client import ApiClientError
 
-        account = {'email': args['email']}
+        # account = {'email': args['email']}
         # account = AccountService.get_by_email(args['email'])
         # if account is None:
         #     raise ValueError('Email not found')
@@ -71,22 +71,22 @@ class ResetPasswordApi(Resource):
         # AccountService.update_password(account, new_password)
 
         # todo: Send email
-        MAILCHIMP_API_KEY = current_app.config['MAILCHIMP_TRANSACTIONAL_API_KEY']
+        # MAILCHIMP_API_KEY = current_app.config['MAILCHIMP_TRANSACTIONAL_API_KEY']
         # mailchimp = MailchimpTransactional(MAILCHIMP_API_KEY)
 
-        message = {
-            'from_email': 'noreply@example.com',
-            'to': [{'email': account.email}],
-            'subject': 'Reset your Dify password',
-            'html': """
-                <p>Dear User,</p>
-                <p>The Dify team has generated a new password for you, details as follows:</p> 
-                <p><strong>{new_password}</strong></p>
-                <p>Please change your password to log in as soon as possible.</p>
-                <p>Regards,</p>
-                <p>The Dify Team</p> 
-            """
-        }
+        # message = {
+        #     'from_email': 'noreply@example.com',
+        #     'to': [{'email': account['email']}],
+        #     'subject': 'Reset your Dify password',
+        #     'html': """
+        #         <p>Dear User,</p>
+        #         <p>The Dify team has generated a new password for you, details as follows:</p> 
+        #         <p><strong>{new_password}</strong></p>
+        #         <p>Please change your password to log in as soon as possible.</p>
+        #         <p>Regards,</p>
+        #         <p>The Dify Team</p> 
+        #     """
+        # }
 
         # response = mailchimp.messages.send({
         #     'message': message,

api/controllers/console/auth/oauth.py

@@ -6,6 +6,7 @@ import requests
 from flask import current_app, redirect, request
 from flask_restful import Resource
 
+from configs import dify_config
 from constants.languages import languages
 from extensions.ext_database import db
 from libs.helper import get_remote_ip
@@ -18,22 +19,24 @@ from .. import api
 
 def get_oauth_providers():
     with current_app.app_context():
-        github_oauth = GitHubOAuth(client_id=current_app.config.get('GITHUB_CLIENT_ID'),
-                                   client_secret=current_app.config.get(
-                                       'GITHUB_CLIENT_SECRET'),
-                                   redirect_uri=current_app.config.get(
-                                       'CONSOLE_API_URL') + '/console/api/oauth/authorize/github')
+        if not dify_config.GITHUB_CLIENT_ID or not dify_config.GITHUB_CLIENT_SECRET:
+            github_oauth = None
+        else:
+            github_oauth = GitHubOAuth(
+                client_id=dify_config.GITHUB_CLIENT_ID,
+                client_secret=dify_config.GITHUB_CLIENT_SECRET,
+                redirect_uri=dify_config.CONSOLE_API_URL + '/console/api/oauth/authorize/github',
+            )
+        if not dify_config.GOOGLE_CLIENT_ID or not dify_config.GOOGLE_CLIENT_SECRET:
+            google_oauth = None
+        else:
+            google_oauth = GoogleOAuth(
+                client_id=dify_config.GOOGLE_CLIENT_ID,
+                client_secret=dify_config.GOOGLE_CLIENT_SECRET,
+                redirect_uri=dify_config.CONSOLE_API_URL + '/console/api/oauth/authorize/google',
+            )
 
-        google_oauth = GoogleOAuth(client_id=current_app.config.get('GOOGLE_CLIENT_ID'),
-                                   client_secret=current_app.config.get(
-                                       'GOOGLE_CLIENT_SECRET'),
-                                   redirect_uri=current_app.config.get(
-                                       'CONSOLE_API_URL') + '/console/api/oauth/authorize/google')
-
-        OAUTH_PROVIDERS = {
-            'github': github_oauth,
-            'google': google_oauth
-        }
+        OAUTH_PROVIDERS = {'github': github_oauth, 'google': google_oauth}
         return OAUTH_PROVIDERS
 
 
@@ -63,8 +66,7 @@ class OAuthCallback(Resource):
             token = oauth_provider.get_access_token(code)
             user_info = oauth_provider.get_user_info(token)
         except requests.exceptions.HTTPError as e:
-            logging.exception(
-                f"An error occurred during the OAuth process with {provider}: {e.response.text}")
+            logging.exception(f'An error occurred during the OAuth process with {provider}: {e.response.text}')
             return {'error': 'OAuth process failed'}, 400
 
         account = _generate_account(provider, user_info)
@@ -81,7 +83,7 @@ class OAuthCallback(Resource):
 
         token = AccountService.login(account, ip_address=get_remote_ip(request))
 
-        return redirect(f'{current_app.config.get("CONSOLE_WEB_URL")}?console_token={token}')
+        return redirect(f'{dify_config.CONSOLE_WEB_URL}?console_token={token}')
 
 
 def _get_account_by_openid_or_email(provider: str, user_info: OAuthUserInfo) -> Optional[Account]:
@@ -101,11 +103,7 @@ def _generate_account(provider: str, user_info: OAuthUserInfo):
         # Create account
         account_name = user_info.name if user_info.name else 'Dify'
         account = RegisterService.register(
-            email=user_info.email,
-            name=account_name,
-            password=None,
-            open_id=user_info.id,
-            provider=provider
+            email=user_info.email, name=account_name, password=None, open_id=user_info.id, provider=provider
         )
 
         # Set interface language

api/controllers/console/workspace/account.py

@@ -245,6 +245,8 @@ class AccountIntegrateApi(Resource):
         return {'data': integrate_data}
 
 
+
+
 # Register API resources
 api.add_resource(AccountInitApi, '/account/init')
 api.add_resource(AccountProfileApi, '/account/profile')

api/core/app/app_config/entities.py

@@ -114,6 +114,10 @@ class VariableEntity(BaseModel):
     default: Optional[str] = None
     hint: Optional[str] = None
 
+    @property
+    def name(self) -> str:
+        return self.variable
+
 
 class ExternalDataVariableEntity(BaseModel):
     """

api/core/app/apps/base_app_generator.py

@@ -1,52 +1,56 @@
+from collections.abc import Mapping
+from typing import Any, Optional
+
 from core.app.app_config.entities import AppConfig, VariableEntity
 
 
 class BaseAppGenerator:
-    def _get_cleaned_inputs(self, user_inputs: dict, app_config: AppConfig):
-        if user_inputs is None:
-            user_inputs = {}
-
-        filtered_inputs = {}
-
+    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
-        for variable_config in variables:
-            variable = variable_config.variable
-
-            if (variable not in user_inputs
-                    or user_inputs[variable] is None
-                    or (isinstance(user_inputs[variable], str) and user_inputs[variable] == '')):
-                if variable_config.required:
-                    raise ValueError(f"{variable} is required in input form")
-                else:
-                    filtered_inputs[variable] = variable_config.default if variable_config.default is not None else ""
-                    continue
-
-            value = user_inputs[variable]
-
-            if value is not None:
-                if variable_config.type != VariableEntity.Type.NUMBER and not isinstance(value, str):
-                    raise ValueError(f"{variable} in input form must be a string")
-                elif variable_config.type == VariableEntity.Type.NUMBER and isinstance(value, str):
-                    if '.' in value:
-                        value = float(value)
-                    else:
-                        value = int(value)
-
-            if variable_config.type == VariableEntity.Type.SELECT:
-                options = variable_config.options if variable_config.options is not None else []
-                if value not in options:
-                    raise ValueError(f"{variable} in input form must be one of the following: {options}")
-            elif variable_config.type in [VariableEntity.Type.TEXT_INPUT, VariableEntity.Type.PARAGRAPH]:
-                if variable_config.max_length is not None:
-                    max_length = variable_config.max_length
-                    if len(value) > max_length:
-                        raise ValueError(f'{variable} in input form must be less than {max_length} characters')
-
-            if value and isinstance(value, str):
-                filtered_inputs[variable] = value.replace('\x00', '')
-            else:
-                filtered_inputs[variable] = value if value is not None else None
-
+        filtered_inputs = {var.name: 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
 
+    def _validate_input(self, *, inputs: Mapping[str, Any], var: VariableEntity):
+        user_input_value = inputs.get(var.name)
+        if var.required and not user_input_value:
+            raise ValueError(f'{var.name} 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 (
+                VariableEntity.Type.TEXT_INPUT,
+                VariableEntity.Type.SELECT,
+                VariableEntity.Type.PARAGRAPH,
+            )
+            and user_input_value
+            and not isinstance(user_input_value, str)
+        ):
+            raise ValueError(f"(type '{var.type}') {var.name} in input form must be a string")
+        if var.type == VariableEntity.Type.NUMBER and isinstance(user_input_value, str):
+            # may raise ValueError if user_input_value is not a valid number
+            try:
+                if '.' in user_input_value:
+                    return float(user_input_value)
+                else:
+                    return int(user_input_value)
+            except ValueError:
+                raise ValueError(f"{var.name} in input form must be a valid number")
+        if var.type == VariableEntity.Type.SELECT:
+            options = var.options or []
+            if user_input_value not in options:
+                raise ValueError(f'{var.name} in input form must be one of the following: {options}')
+        elif var.type in (VariableEntity.Type.TEXT_INPUT, VariableEntity.Type.PARAGRAPH):
+            if var.max_length and user_input_value and len(user_input_value) > var.max_length:
+                raise ValueError(f'{var.name} in input form must be less than {var.max_length} characters')
+
+        return user_input_value
+
+    def _sanitize_value(self, value: Any) -> Any:
+        if isinstance(value, str):
+            return value.replace('\x00', '')
+        return value

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

@@ -94,7 +94,6 @@ class WorkflowAppGenerator(BaseAppGenerator):
             application_generate_entity=application_generate_entity,
             invoke_from=invoke_from,
             stream=stream,
-            call_depth=call_depth,
         )
 
     def _generate(
@@ -104,7 +103,6 @@ class WorkflowAppGenerator(BaseAppGenerator):
         application_generate_entity: WorkflowAppGenerateEntity,
         invoke_from: InvokeFrom,
         stream: bool = True,
-        call_depth: int = 0
     ) -> Union[dict, Generator[dict, None, None]]:
         """
         Generate App response.
@@ -166,10 +164,10 @@ class WorkflowAppGenerator(BaseAppGenerator):
         """
         if not node_id:
             raise ValueError('node_id is required')
-        
+
         if args.get('inputs') is None:
             raise ValueError('inputs is required')
-        
+
         extras = {
             "auto_generate_conversation_name": False
         }

api/core/extension/api_based_extension_requestor.py

@@ -1,7 +1,6 @@
-import os
-
 import requests
 
+from configs import dify_config
 from models.api_based_extension import APIBasedExtensionPoint
 
 
@@ -31,10 +30,10 @@ class APIBasedExtensionRequestor:
         try:
             # proxy support for security
             proxies = None
-            if os.environ.get("SSRF_PROXY_HTTP_URL") and os.environ.get("SSRF_PROXY_HTTPS_URL"):
+            if dify_config.SSRF_PROXY_HTTP_URL and dify_config.SSRF_PROXY_HTTPS_URL:
                 proxies = {
-                    'http': os.environ.get("SSRF_PROXY_HTTP_URL"),
-                    'https': os.environ.get("SSRF_PROXY_HTTPS_URL"),
+                    'http': dify_config.SSRF_PROXY_HTTP_URL,
+                    'https': dify_config.SSRF_PROXY_HTTPS_URL,
                 }
 
             response = requests.request(

api/core/file/message_file_parser.py

@@ -186,7 +186,7 @@ class MessageFileParser:
             }
 
             response = requests.head(url, headers=headers, allow_redirects=True)
-            if response.status_code == 200:
+            if response.status_code in {200, 304}:
                 return True, ""
             else:
                 return False, "URL does not exist."

api/core/helper/code_executor/code_executor.py

@@ -1,5 +1,4 @@
 import logging
-import os
 import time
 from enum import Enum
 from threading import Lock
@@ -9,6 +8,7 @@ from httpx import get, post
 from pydantic import BaseModel
 from yarl import URL
 
+from configs import dify_config
 from core.helper.code_executor.entities import CodeDependency
 from core.helper.code_executor.javascript.javascript_transformer import NodeJsTemplateTransformer
 from core.helper.code_executor.jinja2.jinja2_transformer import Jinja2TemplateTransformer
@@ -18,8 +18,8 @@ from core.helper.code_executor.template_transformer import TemplateTransformer
 logger = logging.getLogger(__name__)
 
 # Code Executor
-CODE_EXECUTION_ENDPOINT = os.environ.get('CODE_EXECUTION_ENDPOINT', 'http://sandbox:8194')
-CODE_EXECUTION_API_KEY = os.environ.get('CODE_EXECUTION_API_KEY', 'dify-sandbox')
+CODE_EXECUTION_ENDPOINT = dify_config.CODE_EXECUTION_ENDPOINT
+CODE_EXECUTION_API_KEY = dify_config.CODE_EXECUTION_API_KEY
 
 CODE_EXECUTION_TIMEOUT= (10, 60)
 

api/core/memory/token_buffer_memory.py

@@ -12,7 +12,8 @@ from core.model_runtime.entities.message_entities import (
     UserPromptMessage,
 )
 from extensions.ext_database import db
-from models.model import AppMode, Conversation, Message
+from models.model import AppMode, Conversation, Message, MessageFile
+from models.workflow import WorkflowRun
 
 
 class TokenBufferMemory:
@@ -30,33 +31,46 @@ class TokenBufferMemory:
         app_record = self.conversation.app
 
         # fetch limited messages, and return reversed
-        query = db.session.query(Message).filter(
+        query = db.session.query(
+            Message.id,
+            Message.query,
+            Message.answer,
+            Message.created_at,
+            Message.workflow_run_id
+        ).filter(
             Message.conversation_id == self.conversation.id,
             Message.answer != ''
         ).order_by(Message.created_at.desc())
 
         if message_limit and message_limit > 0:
-            messages = query.limit(message_limit).all()
+            message_limit = message_limit if message_limit <= 500 else 500
         else:
-            messages = query.all()
+            message_limit = 500
+
+        messages = query.limit(message_limit).all()
 
         messages = list(reversed(messages))
         message_file_parser = MessageFileParser(
             tenant_id=app_record.tenant_id,
             app_id=app_record.id
         )
-
         prompt_messages = []
         for message in messages:
-            files = message.message_files
+            files = db.session.query(MessageFile).filter(MessageFile.message_id == message.id).all()
             if files:
+                file_extra_config = None
                 if self.conversation.mode not in [AppMode.ADVANCED_CHAT.value, AppMode.WORKFLOW.value]:
-                    file_extra_config = FileUploadConfigManager.convert(message.app_model_config.to_dict())
+                    file_extra_config = FileUploadConfigManager.convert(self.conversation.model_config)
                 else:
-                    file_extra_config = FileUploadConfigManager.convert(
-                        message.workflow_run.workflow.features_dict,
-                        is_vision=False
-                    )
+                    if message.workflow_run_id:
+                        workflow_run = (db.session.query(WorkflowRun)
+                                        .filter(WorkflowRun.id == message.workflow_run_id).first())
+
+                        if workflow_run:
+                            file_extra_config = FileUploadConfigManager.convert(
+                                workflow_run.workflow.features_dict,
+                                is_vision=False
+                            )
 
                 if file_extra_config:
                     file_objs = message_file_parser.transform_message_files(
@@ -136,4 +150,4 @@ class TokenBufferMemory:
                 message = f"{role}: {m.content}"
                 string_messages.append(message)
 
-        return "\n".join(string_messages)
+        return "\n".join(string_messages)

api/core/model_runtime/model_providers/bedrock/llm/anthropic.claude-3-haiku-v1.yaml

@@ -5,6 +5,8 @@ model_type: llm
 features:
   - agent-thought
   - vision
+  - tool-call
+  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 200000

api/core/model_runtime/model_providers/bedrock/llm/anthropic.claude-3-opus-v1.yaml

@@ -5,6 +5,8 @@ model_type: llm
 features:
   - agent-thought
   - vision
+  - tool-call
+  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 200000

api/core/model_runtime/model_providers/bedrock/llm/anthropic.claude-3-sonnet-v1.5.yaml

@@ -5,6 +5,8 @@ model_type: llm
 features:
   - agent-thought
   - vision
+  - tool-call
+  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 200000

api/core/model_runtime/model_providers/bedrock/llm/anthropic.claude-3-sonnet-v1.yaml

@@ -5,6 +5,8 @@ model_type: llm
 features:
   - agent-thought
   - vision
+  - tool-call
+  - stream-tool-call
 model_properties:
   mode: chat
   context_size: 200000

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

@@ -29,6 +29,7 @@ from core.model_runtime.entities.message_entities import (
     PromptMessageTool,
     SystemPromptMessage,
     TextPromptMessageContent,
+    ToolPromptMessage,
     UserPromptMessage,
 )
 from core.model_runtime.errors.invoke import (
@@ -68,7 +69,7 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         # TODO: consolidate different invocation methods for models based on base model capabilities
         # invoke anthropic models via boto3 client
         if "anthropic" in model:
-            return self._generate_anthropic(model, credentials, prompt_messages, model_parameters, stop, stream, user)
+            return self._generate_anthropic(model, credentials, prompt_messages, model_parameters, stop, stream, user, tools)
         # invoke Cohere models via boto3 client
         if "cohere.command-r" in model:
             return self._generate_cohere_chat(model, credentials, prompt_messages, model_parameters, stop, stream, user, tools)
@@ -151,7 +152,7 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
 
 
     def _generate_anthropic(self, model: str, credentials: dict, prompt_messages: list[PromptMessage], model_parameters: dict,
-                stop: Optional[list[str]] = None, stream: bool = True, user: Optional[str] = None) -> Union[LLMResult, Generator]:
+                stop: Optional[list[str]] = None, stream: bool = True, user: Optional[str] = None, tools: Optional[list[PromptMessageTool]] = None,) -> Union[LLMResult, Generator]:
         """
         Invoke Anthropic large language model
 
@@ -171,23 +172,24 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         system, prompt_message_dicts = self._convert_converse_prompt_messages(prompt_messages)
         inference_config, additional_model_fields = self._convert_converse_api_model_parameters(model_parameters, stop)
 
+        parameters = {
+            'modelId': model,
+            'messages': prompt_message_dicts,
+            'inferenceConfig': inference_config,
+            'additionalModelRequestFields': additional_model_fields,
+        }
+
+        if system and len(system) > 0:
+            parameters['system'] = system
+
+        if tools:
+            parameters['toolConfig'] = self._convert_converse_tool_config(tools=tools)
+
         if stream:
-            response = bedrock_client.converse_stream(
-                modelId=model,
-                messages=prompt_message_dicts,
-                system=system,
-                inferenceConfig=inference_config,
-                additionalModelRequestFields=additional_model_fields
-            )
+            response = bedrock_client.converse_stream(**parameters)
             return self._handle_converse_stream_response(model, credentials, response, prompt_messages)
         else:
-            response = bedrock_client.converse(
-                modelId=model,
-                messages=prompt_message_dicts,
-                system=system,
-                inferenceConfig=inference_config,
-                additionalModelRequestFields=additional_model_fields
-            )
+            response = bedrock_client.converse(**parameters)
             return self._handle_converse_response(model, credentials, response, prompt_messages)
 
     def _handle_converse_response(self, model: str, credentials: dict, response: dict,
@@ -246,12 +248,18 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
             output_tokens = 0
             finish_reason = None
             index = 0
+            tool_calls: list[AssistantPromptMessage.ToolCall] = []
+            tool_use = {}
 
             for chunk in response['stream']:
                 if 'messageStart' in chunk:
                     return_model = model
                 elif 'messageStop' in chunk:
                     finish_reason = chunk['messageStop']['stopReason']
+                elif 'contentBlockStart' in chunk:
+                    tool = chunk['contentBlockStart']['start']['toolUse']
+                    tool_use['toolUseId'] = tool['toolUseId']
+                    tool_use['name'] = tool['name']
                 elif 'metadata' in chunk:
                     input_tokens = chunk['metadata']['usage']['inputTokens']
                     output_tokens = chunk['metadata']['usage']['outputTokens']
@@ -260,29 +268,49 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
                         model=return_model,
                         prompt_messages=prompt_messages,
                         delta=LLMResultChunkDelta(
-                            index=index + 1,
+                            index=index,
                             message=AssistantPromptMessage(
-                                content=''
+                                content='',
+                                tool_calls=tool_calls
                             ),
                             finish_reason=finish_reason,
                             usage=usage
                         )
                     )
                 elif 'contentBlockDelta' in chunk:
-                    chunk_text = chunk['contentBlockDelta']['delta']['text'] if chunk['contentBlockDelta']['delta']['text'] else ''
-                    full_assistant_content += chunk_text
-                    assistant_prompt_message = AssistantPromptMessage(
-                        content=chunk_text if chunk_text else '',
-                    )
-                    index = chunk['contentBlockDelta']['contentBlockIndex']
-                    yield LLMResultChunk(
-                        model=model,
-                        prompt_messages=prompt_messages,
-                        delta=LLMResultChunkDelta(
-                            index=index,
-                            message=assistant_prompt_message,
+                    delta = chunk['contentBlockDelta']['delta']
+                    if 'text' in delta:
+                        chunk_text = delta['text'] if delta['text'] else ''
+                        full_assistant_content += chunk_text
+                        assistant_prompt_message = AssistantPromptMessage(
+                            content=chunk_text if chunk_text else '',
                         )
-                    )
+                        index = chunk['contentBlockDelta']['contentBlockIndex']
+                        yield LLMResultChunk(
+                            model=model,
+                            prompt_messages=prompt_messages,
+                            delta=LLMResultChunkDelta(
+                                index=index+1,
+                                message=assistant_prompt_message,
+                            )
+                        )
+                    elif 'toolUse' in delta:
+                        if 'input' not in tool_use:
+                            tool_use['input'] = ''
+                        tool_use['input'] += delta['toolUse']['input']
+                elif 'contentBlockStop' in chunk:
+                    if 'input' in tool_use:
+                        tool_call = AssistantPromptMessage.ToolCall(
+                            id=tool_use['toolUseId'],
+                            type='function',
+                            function=AssistantPromptMessage.ToolCall.ToolCallFunction(
+                                name=tool_use['name'],
+                                arguments=tool_use['input']
+                            )
+                        )
+                        tool_calls.append(tool_call)
+                        tool_use = {}
+
         except Exception as ex:
             raise InvokeError(str(ex))
     
@@ -312,16 +340,10 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         """
 
         system = []
-        first_loop = True
         for message in prompt_messages:
             if isinstance(message, SystemPromptMessage):
                 message.content=message.content.strip()
-                if first_loop:
-                    system=message.content
-                    first_loop=False
-                else:
-                    system+="\n"
-                    system+=message.content
+                system.append({"text": message.content})
 
         prompt_message_dicts = []
         for message in prompt_messages:
@@ -330,6 +352,25 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
 
         return system, prompt_message_dicts
 
+    def _convert_converse_tool_config(self, tools: Optional[list[PromptMessageTool]] = None) -> dict:
+        tool_config = {}
+        configs = []
+        if tools:
+            for tool in tools:
+                configs.append(
+                    {
+                        "toolSpec": {
+                            "name": tool.name,
+                            "description": tool.description,
+                            "inputSchema": {
+                                "json": tool.parameters
+                            }
+                        }
+                    }
+                )
+            tool_config["tools"] = configs
+            return tool_config
+    
     def _convert_prompt_message_to_dict(self, message: PromptMessage) -> dict:
         """
         Convert PromptMessage to dict
@@ -379,10 +420,32 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
                 message_dict = {"role": "user", "content": sub_messages}
         elif isinstance(message, AssistantPromptMessage):
             message = cast(AssistantPromptMessage, message)
-            message_dict = {"role": "assistant", "content": [{'text': message.content}]}
+            if message.tool_calls:
+                message_dict = {
+                    "role": "assistant", "content":[{
+                        "toolUse": {
+                            "toolUseId": message.tool_calls[0].id,
+                            "name": message.tool_calls[0].function.name,
+                            "input": json.loads(message.tool_calls[0].function.arguments)
+                        }
+                    }]
+                }
+            else:
+                message_dict = {"role": "assistant", "content": [{'text': message.content}]}
         elif isinstance(message, SystemPromptMessage):
             message = cast(SystemPromptMessage, message)
             message_dict = [{'text': message.content}]
+        elif isinstance(message, ToolPromptMessage):
+            message = cast(ToolPromptMessage, message)
+            message_dict = {
+                "role": "user",
+                "content": [{
+                    "toolResult": {
+                        "toolUseId": message.tool_call_id,
+                        "content": [{"json": {"text": message.content}}]
+                    }                   
+                }]
+            }
         else:
             raise ValueError(f"Got unknown type {message}")
 
@@ -401,11 +464,13 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         """
         prefix = model.split('.')[0]
         model_name = model.split('.')[1]
+        
         if isinstance(prompt_messages, str):
             prompt = prompt_messages
         else:
             prompt = self._convert_messages_to_prompt(prompt_messages, prefix, model_name)
 
+
         return self._get_num_tokens_by_gpt2(prompt)
 
     def validate_credentials(self, model: str, credentials: dict) -> None:
@@ -489,11 +554,16 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         content = message.content
 
         if isinstance(message, UserPromptMessage):
-            message_text = f"{human_prompt_prefix} {content} {human_prompt_postfix}"
+            body = content
+            if (isinstance(content, list)):
+                body = "".join([c.data for c in content if c.type == PromptMessageContentType.TEXT])
+            message_text = f"{human_prompt_prefix} {body} {human_prompt_postfix}"
         elif isinstance(message, AssistantPromptMessage):
             message_text = f"{ai_prompt} {content}"
         elif isinstance(message, SystemPromptMessage):
             message_text = content
+        elif isinstance(message, ToolPromptMessage):
+            message_text = f"{human_prompt_prefix} {message.content}"
         else:
             raise ValueError(f"Got unknown type {message}")
 

api/core/model_runtime/model_providers/volcengine_maas/llm/models.py

@@ -111,5 +111,71 @@ ModelConfigs = {
             'mode': 'chat',
         },
         'features': [],
+    },
+    'Moonshot-v1-8k': {
+        'req_params': {
+            'max_prompt_tokens': 8192,
+            'max_new_tokens': 4096,
+        },
+        'model_properties': {
+            'context_size': 8192,
+            'mode': 'chat',
+        },
+        'features': [],
+    },
+    'Moonshot-v1-32k': {
+        'req_params': {
+            'max_prompt_tokens': 32768,
+            'max_new_tokens': 16384,
+        },
+        'model_properties': {
+            'context_size': 32768,
+            'mode': 'chat',
+        },
+        'features': [],
+    },
+    'Moonshot-v1-128k': {
+        'req_params': {
+            'max_prompt_tokens': 131072,
+            'max_new_tokens': 65536,
+        },
+        'model_properties': {
+            'context_size': 131072,
+            'mode': 'chat',
+        },
+        'features': [],
+    },
+    'GLM3-130B': {
+        'req_params': {
+            'max_prompt_tokens': 8192,
+            'max_new_tokens': 4096,
+        },
+        'model_properties': {
+            'context_size': 8192,
+            'mode': 'chat',
+        },
+        'features': [],
+    },
+    'GLM3-130B-Fin': {
+        'req_params': {
+            'max_prompt_tokens': 8192,
+            'max_new_tokens': 4096,
+        },
+        'model_properties': {
+            'context_size': 8192,
+            'mode': 'chat',
+        },
+        'features': [],
+    },
+    'Mistral-7B': {
+        'req_params': {
+            'max_prompt_tokens': 8192,
+            'max_new_tokens': 2048,
+        },
+        'model_properties': {
+            'context_size': 8192,
+            'mode': 'chat',
+        },
+        'features': [],
     }
 }

api/core/model_runtime/model_providers/volcengine_maas/volcengine_maas.yaml

@@ -120,12 +120,6 @@ model_credential_schema:
           show_on:
             - variable: __model_type
               value: llm
-        - label:
-            en_US: Skylark2-pro-4k
-          value: Skylark2-pro-4k
-          show_on:
-            - variable: __model_type
-              value: llm
         - label:
             en_US: Llama3-8B
           value: Llama3-8B
@@ -138,6 +132,42 @@ model_credential_schema:
           show_on:
             - variable: __model_type
               value: llm
+        - label:
+            en_US: Moonshot-v1-8k
+          value: Moonshot-v1-8k
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: Moonshot-v1-32k
+          value: Moonshot-v1-32k
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: Moonshot-v1-128k
+          value: Moonshot-v1-128k
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: GLM3-130B
+          value: GLM3-130B
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: GLM3-130B-Fin
+          value: GLM3-130B-Fin
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: Mistral-7B
+          value: Mistral-7B
+          show_on:
+            - variable: __model_type
+              value: llm
         - label:
             en_US: Doubao-embedding
           value: Doubao-embedding
@@ -181,7 +211,7 @@ model_credential_schema:
         zh_Hans: 模型上下文长度
         en_US: Model Context Size
       type: text-input
-      default: '4096'
+      default: "4096"
       placeholder:
         zh_Hans: 输入您的模型上下文长度
         en_US: Enter your Model Context Size
@@ -195,7 +225,7 @@ model_credential_schema:
       label:
         zh_Hans: 最大 token 上限
         en_US: Upper Bound for Max Tokens
-      default: '4096'
+      default: "4096"
       type: text-input
       placeholder:
         zh_Hans: 输入您的模型最大 token 上限

api/core/model_runtime/model_providers/zhipuai/zhipuai_sdk/_client.py

@@ -29,10 +29,8 @@ class ZhipuAI(HttpClient):
             http_client: httpx.Client | None = None,
             custom_headers: Mapping[str, str] | None = None
     ) -> None:
-        # if api_key is None:
-        #     api_key = os.environ.get("ZHIPUAI_API_KEY")
         if api_key is None:
-            raise ZhipuAIError("未提供api_key，请通过参数或环境变量提供")
+            raise ZhipuAIError("No api_key provided, please provide it through parameters or environment variables")
         self.api_key = api_key
 
         if base_url is None:

api/core/model_runtime/model_providers/zhipuai/zhipuai_sdk/core/_http_client.py

@@ -7,6 +7,8 @@ from typing import Any, Union, cast
 import httpx
 import pydantic
 from httpx import URL, Timeout
+from tenacity import retry
+from tenacity.stop import stop_after_attempt
 
 from . import _errors
 from ._base_type import NOT_GIVEN, Body, Data, Headers, NotGiven, Query, RequestFiles, ResponseT
@@ -221,6 +223,7 @@ class HttpClient:
     def __exit__(self, exc_type, exc_val, exc_tb):
         self.close()
 
+    @retry(stop=stop_after_attempt(ZHIPUAI_DEFAULT_MAX_RETRIES))
     def request(
             self,
             *,

api/core/model_runtime/utils/_compat.py

@@ -1,21 +0,0 @@
-from typing import Any, Literal
-
-from pydantic import BaseModel
-from pydantic.version import VERSION as PYDANTIC_VERSION
-
-PYDANTIC_V2 = PYDANTIC_VERSION.startswith("2.")
-
-if PYDANTIC_V2:
-    from pydantic_core import Url as Url
-
-    def _model_dump(
-        model: BaseModel, mode: Literal["json", "python"] = "json", **kwargs: Any
-    ) -> Any:
-        return model.model_dump(mode=mode, **kwargs)
-else:
-    from pydantic import AnyUrl as Url  # noqa: F401
-
-    def _model_dump(
-        model: BaseModel, mode: Literal["json", "python"] = "json", **kwargs: Any
-    ) -> Any:
-        return model.dict(**kwargs)

api/core/model_runtime/utils/encoders.py

@@ -8,16 +8,20 @@ from ipaddress import IPv4Address, IPv4Interface, IPv4Network, IPv6Address, IPv6
 from pathlib import Path, PurePath
 from re import Pattern
 from types import GeneratorType
-from typing import Any, Optional, Union
+from typing import Any, Literal, Optional, Union
 from uuid import UUID
 
 from pydantic import BaseModel
 from pydantic.networks import AnyUrl, NameEmail
 from pydantic.types import SecretBytes, SecretStr
+from pydantic_core import Url
 from pydantic_extra_types.color import Color
 
-from ._compat import PYDANTIC_V2, Url, _model_dump
 
+def _model_dump(
+    model: BaseModel, mode: Literal["json", "python"] = "json", **kwargs: Any
+) -> Any:
+    return model.model_dump(mode=mode, **kwargs)
 
 # Taken from Pydantic v1 as is
 def isoformat(o: Union[datetime.date, datetime.time]) -> str:
@@ -109,12 +113,6 @@ def jsonable_encoder(
                 if isinstance(obj, encoder_type):
                     return encoder_instance(obj)
     if isinstance(obj, BaseModel):
-        # TODO: remove when deprecating Pydantic v1
-        encoders: dict[Any, Any] = {}
-        if not PYDANTIC_V2:
-            encoders = getattr(obj.__config__, "json_encoders", {})  # type: ignore[attr-defined]
-            if custom_encoder:
-                encoders.update(custom_encoder)
         obj_dict = _model_dump(
             obj,
             mode="json",
@@ -131,8 +129,6 @@ def jsonable_encoder(
             obj_dict,
             exclude_none=exclude_none,
             exclude_defaults=exclude_defaults,
-            # TODO: remove when deprecating Pydantic v1
-            custom_encoder=encoders,
             sqlalchemy_safe=sqlalchemy_safe,
         )
     if dataclasses.is_dataclass(obj):

api/core/ops/entities/config_entity.py

@@ -27,7 +27,7 @@ class LangfuseConfig(BaseTracingConfig):
     def set_value(cls, v, info: ValidationInfo):
         if v is None or v == "":
             v = 'https://api.langfuse.com'
-        if not v.startswith('https://') or not v.startswith('http://'):
+        if not v.startswith('https://') and not v.startswith('http://'):
             raise ValueError('host must start with https:// or http://')
 
         return v

api/core/ops/langfuse_trace/langfuse_trace.py

@@ -107,7 +107,20 @@ class LangFuseDataTrace(BaseTraceInstance):
 
         # through workflow_run_id get all_nodes_execution
         workflow_nodes_executions = (
-            db.session.query(WorkflowNodeExecution)
+            db.session.query(
+                WorkflowNodeExecution.id,
+                WorkflowNodeExecution.tenant_id,
+                WorkflowNodeExecution.app_id,
+                WorkflowNodeExecution.title,
+                WorkflowNodeExecution.node_type,
+                WorkflowNodeExecution.status,
+                WorkflowNodeExecution.inputs,
+                WorkflowNodeExecution.outputs,
+                WorkflowNodeExecution.created_at,
+                WorkflowNodeExecution.elapsed_time,
+                WorkflowNodeExecution.process_data,
+                WorkflowNodeExecution.execution_metadata,
+            )
             .filter(WorkflowNodeExecution.workflow_run_id == trace_info.workflow_run_id)
             .all()
         )

api/core/ops/langsmith_trace/langsmith_trace.py

@@ -100,7 +100,20 @@ class LangSmithDataTrace(BaseTraceInstance):
 
         # through workflow_run_id get all_nodes_execution
         workflow_nodes_executions = (
-            db.session.query(WorkflowNodeExecution)
+            db.session.query(
+                WorkflowNodeExecution.id,
+                WorkflowNodeExecution.tenant_id,
+                WorkflowNodeExecution.app_id,
+                WorkflowNodeExecution.title,
+                WorkflowNodeExecution.node_type,
+                WorkflowNodeExecution.status,
+                WorkflowNodeExecution.inputs,
+                WorkflowNodeExecution.outputs,
+                WorkflowNodeExecution.created_at,
+                WorkflowNodeExecution.elapsed_time,
+                WorkflowNodeExecution.process_data,
+                WorkflowNodeExecution.execution_metadata,
+            )
             .filter(WorkflowNodeExecution.workflow_run_id == trace_info.workflow_run_id)
             .all()
         )

api/core/rag/extractor/notion_extractor.py

@@ -140,11 +140,10 @@ class NotionExtractor(BaseExtractor):
 
     def _get_notion_block_data(self, page_id: str) -> list[str]:
         result_lines_arr = []
-        cur_block_id = page_id
+        start_cursor = None
+        block_url = BLOCK_CHILD_URL_TMPL.format(block_id=page_id)
         while True:
-            block_url = BLOCK_CHILD_URL_TMPL.format(block_id=cur_block_id)
-            query_dict: dict[str, Any] = {}
-
+            query_dict: dict[str, Any] = {} if not start_cursor else {'start_cursor': start_cursor}
             res = requests.request(
                 "GET",
                 block_url,
@@ -153,7 +152,7 @@ class NotionExtractor(BaseExtractor):
                     "Content-Type": "application/json",
                     "Notion-Version": "2022-06-28",
                 },
-                json=query_dict
+                params=query_dict
             )
             data = res.json()
             for result in data["results"]:
@@ -191,16 +190,16 @@ class NotionExtractor(BaseExtractor):
             if data["next_cursor"] is None:
                 break
             else:
-                cur_block_id = data["next_cursor"]
+                start_cursor = data["next_cursor"]
         return result_lines_arr
 
     def _read_block(self, block_id: str, num_tabs: int = 0) -> str:
         """Read a block."""
         result_lines_arr = []
-        cur_block_id = block_id
+        start_cursor = None
+        block_url = BLOCK_CHILD_URL_TMPL.format(block_id=block_id)
         while True:
-            block_url = BLOCK_CHILD_URL_TMPL.format(block_id=cur_block_id)
-            query_dict: dict[str, Any] = {}
+            query_dict: dict[str, Any] = {} if not start_cursor else {'start_cursor': start_cursor}
 
             res = requests.request(
                 "GET",
@@ -210,7 +209,7 @@ class NotionExtractor(BaseExtractor):
                     "Content-Type": "application/json",
                     "Notion-Version": "2022-06-28",
                 },
-                json=query_dict
+                params=query_dict
             )
             data = res.json()
             if 'results' not in data or data["results"] is None:
@@ -249,7 +248,7 @@ class NotionExtractor(BaseExtractor):
             if data["next_cursor"] is None:
                 break
             else:
-                cur_block_id = data["next_cursor"]
+                start_cursor = data["next_cursor"]
 
         result_lines = "\n".join(result_lines_arr)
         return result_lines
@@ -258,10 +257,10 @@ class NotionExtractor(BaseExtractor):
         """Read table rows."""
         done = False
         result_lines_arr = []
-        cur_block_id = block_id
+        start_cursor = None
+        block_url = BLOCK_CHILD_URL_TMPL.format(block_id=block_id)
         while not done:
-            block_url = BLOCK_CHILD_URL_TMPL.format(block_id=cur_block_id)
-            query_dict: dict[str, Any] = {}
+            query_dict: dict[str, Any] = {} if not start_cursor else {'start_cursor': start_cursor}
 
             res = requests.request(
                 "GET",
@@ -271,7 +270,7 @@ class NotionExtractor(BaseExtractor):
                     "Content-Type": "application/json",
                     "Notion-Version": "2022-06-28",
                 },
-                json=query_dict
+                params=query_dict
             )
             data = res.json()
             # get table headers text
@@ -300,7 +299,7 @@ class NotionExtractor(BaseExtractor):
                 done = True
                 break
             else:
-                cur_block_id = data["next_cursor"]
+                start_cursor = data["next_cursor"]
 
         result_lines = "\n".join(result_lines_arr)
         return result_lines

api/core/tools/docs/en_US/tool_scale_out.md

@@ -5,6 +5,7 @@ Here, we will use GoogleSearch as an example to demonstrate how to quickly integ
 ## 1. Prepare the Tool Provider yaml
 
 ### Introduction
+
 This yaml declares a new tool provider, and includes information like the provider's name, icon, author, and other details that are fetched by the frontend for display.
 
 ### Example
@@ -28,9 +29,11 @@ identity: # Basic information of the tool provider
     - search
 
 ```
- - The `identity` field is mandatory, it contains the basic information of the tool provider, including author, name, label, description, icon, etc.
-    - The icon needs to be placed in the `_assets` folder of the current module, you can refer to [here](../../provider/builtin/google/_assets/icon.svg).
-    - The `tags` field is optional, it is used to classify the provider, and the frontend can filter the provider according to the tag, for all tags, they have been listed below:
+
+- The `identity` field is mandatory, it contains the basic information of the tool provider, including author, name, label, description, icon, etc.
+  - The icon needs to be placed in the `_assets` folder of the current module, you can refer to [here](../../provider/builtin/google/_assets/icon.svg).
+  - The `tags` field is optional, it is used to classify the provider, and the frontend can filter the provider according to the tag, for all tags, they have been listed below:
+
     ```python
     class ToolLabelEnum(Enum):
       SEARCH = 'search'
@@ -56,6 +59,7 @@ identity: # Basic information of the tool provider
 Google, as a third-party tool, uses the API provided by SerpApi, which requires an API Key to use. This means that this tool needs a credential to use. For tools like `wikipedia`, there is no need to fill in the credential field, you can refer to [here](../../provider/builtin/wikipedia/wikipedia.yaml).
 
 After configuring the credential field, the effect is as follows:
+
 ```yaml
 identity:
   author: Dify
@@ -87,6 +91,7 @@ credentials_for_provider: # Credential field
 - `type`: Credential field type, currently can be either `secret-input`, `text-input`, or `select` , corresponding to password input box, text input box, and drop-down box, respectively. If set to `secret-input`, it will mask the input content on the frontend, and the backend will encrypt the input content.
 
 ## 3. Prepare Tool yaml
+
 A provider can have multiple tools, each tool needs a yaml file to describe, this file contains the basic information, parameters, output, etc. of the tool.
 
 Still taking GoogleSearch as an example, we need to create a `tools` module under the `google` module, and create `tools/google_search.yaml`, the content is as follows.
@@ -140,21 +145,22 @@ parameters: # Parameter list
 
 - The `identity` field is mandatory, it contains the basic information of the tool, including name, author, label, description, etc.
 - `parameters` Parameter list
-    - `name` Parameter name, unique, no duplication with other parameters
-    - `type` Parameter type, currently supports `string`, `number`, `boolean`, `select`, `secret-input` four types, corresponding to string, number, boolean, drop-down box, and encrypted input box, respectively. For sensitive information, we recommend using `secret-input` type
-    - `required` Required or not
-        - In `llm` mode, if the parameter is required, the Agent is required to infer this parameter
-        - In `form` mode, if the parameter is required, the user is required to fill in this parameter on the frontend before the conversation starts
-    - `options` Parameter options
-        - In `llm` mode, Dify will pass all options to LLM, LLM can infer based on these options
-        - In `form` mode, when `type` is `select`, the frontend will display these options
-    - `default` Default value
-    - `label` Parameter label, for frontend display
-    - `human_description` Introduction for frontend display, supports multiple languages
-    - `llm_description` Introduction passed to LLM, in order to make LLM better understand this parameter, we suggest to write as detailed information about this parameter as possible here, so that LLM can understand this parameter
-    - `form` Form type, currently supports `llm`, `form` two types, corresponding to Agent self-inference and frontend filling
+  - `name` Parameter name, unique, no duplication with other parameters
+  - `type` Parameter type, currently supports `string`, `number`, `boolean`, `select`, `secret-input` four types, corresponding to string, number, boolean, drop-down box, and encrypted input box, respectively. For sensitive information, we recommend using `secret-input` type
+  - `required` Required or not
+    - In `llm` mode, if the parameter is required, the Agent is required to infer this parameter
+    - In `form` mode, if the parameter is required, the user is required to fill in this parameter on the frontend before the conversation starts
+  - `options` Parameter options
+    - In `llm` mode, Dify will pass all options to LLM, LLM can infer based on these options
+    - In `form` mode, when `type` is `select`, the frontend will display these options
+  - `default` Default value
+  - `label` Parameter label, for frontend display
+  - `human_description` Introduction for frontend display, supports multiple languages
+  - `llm_description` Introduction passed to LLM, in order to make LLM better understand this parameter, we suggest to write as detailed information about this parameter as possible here, so that LLM can understand this parameter
+  - `form` Form type, currently supports `llm`, `form` two types, corresponding to Agent self-inference and frontend filling
 
 ## 4. Add Tool Logic
+
 After completing the tool configuration, we can start writing the tool code that defines how it is invoked.
 
 Create `google_search.py` under the `google/tools` module, the content is as follows.
@@ -176,7 +182,7 @@ class GoogleSearchTool(BuiltinTool):
         query = tool_parameters['query']
         result_type = tool_parameters['result_type']
         api_key = self.runtime.credentials['serpapi_api_key']
-        # TODO: search with serpapi
+        # Search with serpapi
         result = SerpAPI(api_key).run(query, result_type=result_type)
 
         if result_type == 'text':
@@ -185,12 +191,15 @@ class GoogleSearchTool(BuiltinTool):
 ```
 
 ### Parameters
+
 The overall logic of the tool is in the `_invoke` method, this method accepts two parameters: `user_id` and `tool_parameters`, which represent the user ID and tool parameters respectively
 
 ### Return Data
+
 When the tool returns, you can choose to return one message or multiple messages, here we return one message, using `create_text_message` and `create_link_message` can create a text message or a link message.
 
 ## 5. Add Provider Code
+
 Finally, we need to create a provider class under the provider module to implement the provider's credential verification logic. If the credential verification fails, it will throw a `ToolProviderCredentialValidationError` exception.
 
 Create `google.py` under the `google` module, the content is as follows.
@@ -227,8 +236,9 @@ class GoogleProvider(BuiltinToolProviderController):
 ```
 
 ## Completion
+
 After the above steps are completed, we can see this tool on the frontend, and it can be used in the Agent.
 
 Of course, because google_search needs a credential, before using it, you also need to input your credentials on the frontend.
 
-![Alt text](../zh_Hans/images/index/image-2.png)
+![Alt text](../zh_Hans/images/index/image-2.png)

api/core/tools/provider/builtin/duckduckgo/tools/ddgo_ai.yaml

@@ -31,6 +31,12 @@ parameters:
       - value: claude-3-haiku
         label:
           en_US: Claude 3
+      - value: llama-3-70b
+        label:
+          en_US: Llama 3
+      - value: mixtral-8x7b
+        label:
+          en_US: Mixtral
     default: gpt-3.5
     label:
       en_US: Choose Model

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

@@ -6,7 +6,7 @@ identity:
     zh_CN: Firecrawl
   description:
     en_US: Firecrawl API integration for web crawling and scraping.
-    zh_CN: Firecrawl API 集成，用于网页爬取和数据抓取。
+    zh_Hans: Firecrawl API 集成，用于网页爬取和数据抓取。
   icon: icon.svg
   tags:
     - search
@@ -17,20 +17,22 @@ credentials_for_provider:
     required: true
     label:
       en_US: Firecrawl API Key
-      zh_CN: Firecrawl API 密钥
+      zh_Hans: Firecrawl API 密钥
     placeholder:
       en_US: Please input your Firecrawl API key
-      zh_CN: 请输入您的 Firecrawl API 密钥
+      zh_Hans: 请输入您的 Firecrawl API 密钥，如果是自托管版本，可以随意填写密钥
     help:
-      en_US: Get your Firecrawl API key from your Firecrawl account settings.
-      zh_CN: 从您的 Firecrawl 账户设置中获取 Firecrawl API 密钥。
+      en_US: Get your Firecrawl API key from your Firecrawl account settings.If you are using a self-hosted version, you may enter any key at your convenience.
+      zh_Hans: 从您的 Firecrawl 账户设置中获取 Firecrawl API 密钥。如果是自托管版本，可以随意填写密钥。
     url: https://www.firecrawl.dev/account
   base_url:
     type: text-input
     required: false
     label:
       en_US: Firecrawl server's Base URL
+      zh_Hans: Firecrawl服务器的API URL
       pt_BR: Firecrawl server's Base URL
     placeholder:
       en_US: https://www.firecrawl.dev
+      zh_HansL: https://www.firecrawl.dev
       pt_BR: https://www.firecrawl.dev

api/core/tools/provider/builtin/firecrawl/firecrawl_appx.py

@@ -1,98 +1,93 @@
 import time
+from collections.abc import Mapping
+from typing import Any
 
 import requests
+from requests.exceptions import HTTPError
 
 
 class FirecrawlApp:
-    def __init__(self, api_key=None, base_url=None):
+    def __init__(self, api_key: str | None = None, base_url: str | None = None):
         self.api_key = api_key
         self.base_url = base_url or 'https://api.firecrawl.dev'
-        if self.api_key is None and self.base_url == 'https://api.firecrawl.dev':
-            raise ValueError('No API key provided')
+        if not self.api_key:
+            raise ValueError("API key is required")
 
-    def scrape_url(self, url, params=None) -> dict:
+    def _prepare_headers(self, idempotency_key: str | None = None):
         headers = {
             'Content-Type': 'application/json',
             'Authorization': f'Bearer {self.api_key}'
         }
-        json_data = {'url': url}
-        if params:
-            json_data.update(params)
-        response = requests.post(
-            f'{self.base_url}/v0/scrape',
-            headers=headers,
-            json=json_data
-        )
-        if response.status_code == 200:
-            response = response.json()
-            if response['success'] == True:
-                return response['data']
-            else:
-                raise Exception(f'Failed to scrape URL. Error: {response["error"]}')
+        if idempotency_key:
+            headers['Idempotency-Key'] = idempotency_key
+        return headers
 
-        elif response.status_code in [402, 409, 500]:
-            error_message = response.json().get('error', 'Unknown error occurred')
-            raise Exception(f'Failed to scrape URL. Status code: {response.status_code}. Error: {error_message}')
-        else:
-            raise Exception(f'Failed to scrape URL. Status code: {response.status_code}')
-
-    def crawl_url(self, url, params=None, wait_until_done=True, timeout=2) -> str:
-        headers = self._prepare_headers()
-        json_data = {'url': url}
-        if params:
-            json_data.update(params)
-        response = self._post_request(f'{self.base_url}/v0/crawl', json_data, headers)
-        if response.status_code == 200:
-            job_id = response.json().get('jobId')
-            if wait_until_done:
-                return self._monitor_job_status(job_id, headers, timeout)
-            else:
-                return {'jobId': job_id}
-        else:
-            self._handle_error(response, 'start crawl job')
-
-    def check_crawl_status(self, job_id) -> dict:
-        headers = self._prepare_headers()
-        response = self._get_request(f'{self.base_url}/v0/crawl/status/{job_id}', headers)
-        if response.status_code == 200:
-            return response.json()
-        else:
-            self._handle_error(response, 'check crawl status')
-
-    def _prepare_headers(self):
-        return {
-            'Content-Type': 'application/json',
-            'Authorization': f'Bearer {self.api_key}'
-        }
-
-    def _post_request(self, url, data, headers):
-        return requests.post(url, headers=headers, json=data)
-
-    def _get_request(self, url, headers):
-        return requests.get(url, headers=headers)
-
-    def _monitor_job_status(self, job_id, headers, timeout):
-        while True:
-            status_response = self._get_request(f'{self.base_url}/v0/crawl/status/{job_id}', headers)
-            if status_response.status_code == 200:
-                status_data = status_response.json()
-                if status_data['status'] == 'completed':
-                    if 'data' in status_data:
-                        return status_data['data']
-                    else:
-                        raise Exception('Crawl job completed but no data was returned')
-                elif status_data['status'] in ['active', 'paused', 'pending', 'queued']:
-                    if timeout < 2:
-                        timeout = 2
-                    time.sleep(timeout)  # Wait for the specified timeout before checking again
+    def _request(
+        self,
+        method: str,
+        url: str,
+        data: Mapping[str, Any] | None = None,
+        headers: Mapping[str, str] | None = None,
+        retries: int = 3,
+        backoff_factor: float = 0.3,
+    ) -> Mapping[str, Any] | None:
+        for i in range(retries):
+            try:
+                response = requests.request(method, url, json=data, headers=headers)
+                response.raise_for_status()
+                return response.json()
+            except requests.exceptions.RequestException as e:
+                if i < retries - 1:
+                    time.sleep(backoff_factor * (2 ** i))
                 else:
-                    raise Exception(f'Crawl job failed or was stopped. Status: {status_data["status"]}')
-            else:
-                self._handle_error(status_response, 'check crawl status')
+                    raise
+        return None
 
-    def _handle_error(self, response, action):
-        if response.status_code in [402, 409, 500]:
-            error_message = response.json().get('error', 'Unknown error occurred')
-            raise Exception(f'Failed to {action}. Status code: {response.status_code}. Error: {error_message}')
-        else:
-            raise Exception(f'Unexpected error occurred while trying to {action}. Status code: {response.status_code}')
+    def scrape_url(self, url: str, **kwargs):
+        endpoint = f'{self.base_url}/v0/scrape'
+        headers = self._prepare_headers()
+        data = {'url': url, **kwargs}
+        response = self._request('POST', endpoint, data, headers)
+        if response is None:
+            raise HTTPError("Failed to scrape URL after multiple retries")
+        return response
+
+    def search(self, query: str, **kwargs):
+        endpoint = f'{self.base_url}/v0/search'
+        headers = self._prepare_headers()
+        data = {'query': query, **kwargs}
+        response = self._request('POST', endpoint, data, headers)
+        if response is None:
+            raise HTTPError("Failed to perform search after multiple retries")
+        return response
+
+    def crawl_url(
+        self, url: str, wait: bool = False, poll_interval: int = 5, idempotency_key: str | None = None, **kwargs
+    ):
+        endpoint = f'{self.base_url}/v0/crawl'
+        headers = self._prepare_headers(idempotency_key)
+        data = {'url': url, **kwargs}
+        response = self._request('POST', endpoint, data, headers)
+        if response is None:
+            raise HTTPError("Failed to initiate crawl after multiple retries")
+        job_id: str = response['jobId']
+        if wait:
+            return self._monitor_job_status(job_id=job_id, poll_interval=poll_interval)
+        return job_id
+
+    def check_crawl_status(self, job_id: str):
+        endpoint = f'{self.base_url}/v0/crawl/status/{job_id}'
+        headers = self._prepare_headers()
+        response = self._request('GET', endpoint, headers=headers)
+        if response is None:
+            raise HTTPError(f"Failed to check status for job {job_id} after multiple retries")
+        return response
+
+    def _monitor_job_status(self, job_id: str, poll_interval: int):
+        while True:
+            status = self.check_crawl_status(job_id)
+            if status['status'] == 'completed':
+                return status
+            elif status['status'] == 'failed':
+                raise HTTPError(f'Job {job_id} failed: {status["error"]}')
+            time.sleep(poll_interval)

api/core/tools/provider/builtin/firecrawl/tools/crawl.py

@@ -1,3 +1,4 @@
+import json
 from typing import Any, Union
 
 from core.tools.entities.tool_entities import ToolInvokeMessage
@@ -7,7 +8,6 @@ from core.tools.tool.builtin_tool import BuiltinTool
 
 class CrawlTool(BuiltinTool):
     def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
-        # initialize the app object with the api key
         app = FirecrawlApp(api_key=self.runtime.credentials['firecrawl_api_key'], base_url=self.runtime.credentials['base_url'])
 
         options = {
@@ -21,29 +21,16 @@ class CrawlTool(BuiltinTool):
             }
         }
 
-        # crawl the url
         crawl_result = app.crawl_url(
             url=tool_parameters['url'], 
             params=options,
-            wait_until_done=True, 
+            wait=True
         )
-        
-        # reformat crawl result
-        crawl_output = "**Crawl Result**\n\n"
-        try:
-            for result in crawl_result:
-                crawl_output += f"**- Title:** {result.get('metadata', {}).get('title', '')}\n"
-                crawl_output += f"**- Description:** {result.get('metadata', {}).get('description', '')}\n"
-                crawl_output += f"**- URL:** {result.get('metadata', {}).get('ogUrl', '')}\n\n"
-                crawl_output += f"**- Web Content:**\n{result.get('markdown', '')}\n\n"
-                crawl_output += "---\n\n"
-        except Exception as e:
-            crawl_output += f"An error occurred: {str(e)}\n"
-            crawl_output += f"**- Title:** {result.get('metadata', {}).get('title', '')}\n"
-            crawl_output += f"**- Description:** {result.get('metadata', {}).get('description','')}\n"
-            crawl_output += f"**- URL:** {result.get('metadata', {}).get('ogUrl', '')}\n\n"
-            crawl_output += f"**- Web Content:**\n{result.get('markdown', '')}\n\n"
-            crawl_output += "---\n\n"
 
+        if not isinstance(crawl_result, str):
+            crawl_result = json.dumps(crawl_result, ensure_ascii=False, indent=4)
 
-        return self.create_text_message(crawl_output)
+        if not crawl_result:
+            return self.create_text_message("Crawl request failed.")
+
+        return self.create_text_message(crawl_result)

api/core/tools/provider/builtin/firecrawl/tools/scrape.py

@@ -0,0 +1,26 @@
+import json
+from typing import Any, Union
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.provider.builtin.firecrawl.firecrawl_appx import FirecrawlApp
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class ScrapeTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        app = FirecrawlApp(api_key=self.runtime.credentials['firecrawl_api_key'], base_url=self.runtime.credentials['base_url'])
+
+        crawl_result = app.scrape_url(
+            url=tool_parameters['url'],
+            wait=True
+        )
+
+        if isinstance(crawl_result, dict):
+            result_message = json.dumps(crawl_result, ensure_ascii=False, indent=4)
+        else:
+            result_message = str(crawl_result)
+
+        if not crawl_result:
+            return self.create_text_message("Scrape request failed.")
+
+        return self.create_text_message(result_message)

api/core/tools/provider/builtin/firecrawl/tools/scrape.yaml

@@ -0,0 +1,23 @@
+identity:
+  name: scrape
+  author: ahasasjeb
+  label:
+    en_US: Scrape
+    zh_Hans: 抓取
+description:
+  human:
+    en_US: Extract data from a single URL.
+    zh_Hans: 从单个URL抓取数据。
+  llm: This tool is designed to scrape URL and output the content in Markdown format.
+parameters:
+  - name: url
+    type: string
+    required: true
+    label:
+      en_US: URL to scrape
+      zh_Hans: 要抓取的URL
+    human_description:
+      en_US: The URL of the website to scrape and extract data from.
+      zh_Hans: 要抓取并提取数据的网站URL。
+    llm_description: The URL of the website that needs to be crawled. This is a required parameter.
+    form: llm

api/core/tools/provider/builtin/firecrawl/tools/search.py

@@ -0,0 +1,26 @@
+import json
+from typing import Any, Union
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.provider.builtin.firecrawl.firecrawl_appx import FirecrawlApp
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class SearchTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        app = FirecrawlApp(api_key=self.runtime.credentials['firecrawl_api_key'], base_url=self.runtime.credentials['base_url'])
+
+        crawl_result = app.search(
+            query=tool_parameters['keyword'],
+            wait=True
+        )
+
+        if isinstance(crawl_result, dict):
+            result_message = json.dumps(crawl_result, ensure_ascii=False, indent=4)
+        else:
+            result_message = str(crawl_result)
+
+        if not crawl_result:
+            return self.create_text_message("Search request failed.")
+
+        return self.create_text_message(result_message)

api/core/tools/provider/builtin/firecrawl/tools/search.yaml

@@ -0,0 +1,23 @@
+identity:
+  name: search
+  author: ahasasjeb
+  label:
+    en_US: Search
+    zh_Hans: 搜索
+description:
+  human:
+    en_US: Search, and output in Markdown format
+    zh_Hans: 搜索，并且以Markdown格式输出
+  llm: This tool can perform online searches and convert the results to Markdown format.
+parameters:
+  - name: keyword
+    type: string
+    required: true
+    label:
+      en_US: keyword
+      zh_Hans: 关键词
+    human_description:
+      en_US: Input keywords to use Firecrawl API for search.
+      zh_Hans: 输入关键词即可使用Firecrawl API进行搜索。
+    llm_description: Efficiently extract keywords from user text.
+    form: llm

api/core/tools/provider/builtin/jina/tools/jina_reader.py

@@ -1,3 +1,4 @@
+import json
 from typing import Any, Union
 
 from yarl import URL
@@ -26,6 +27,15 @@ class JinaReaderTool(BuiltinTool):
         if 'api_key' in self.runtime.credentials and self.runtime.credentials.get('api_key'):
             headers['Authorization'] = "Bearer " + self.runtime.credentials.get('api_key')
 
+        request_params = tool_parameters.get('request_params')
+        if request_params is not None and request_params != '':
+            try:
+                request_params = json.loads(request_params)
+                if not isinstance(request_params, dict):
+                    raise ValueError("request_params must be a JSON object")
+            except (json.JSONDecodeError, ValueError) as e:
+                raise ValueError(f"Invalid request_params: {e}")
+
         target_selector = tool_parameters.get('target_selector')
         if target_selector is not None and target_selector != '':
             headers['X-Target-Selector'] = target_selector
@@ -53,7 +63,8 @@ class JinaReaderTool(BuiltinTool):
         response = ssrf_proxy.get(
             str(URL(self._jina_reader_endpoint + url)),
             headers=headers,
-            timeout=(10, 60)
+            params=request_params,
+            timeout=(10, 60),
         )
 
         if tool_parameters.get('summary', False):

api/core/tools/provider/builtin/jina/tools/jina_reader.yaml

@@ -25,6 +25,22 @@ parameters:
       pt_BR: used for linking to webpages
     llm_description: url for scraping
     form: llm
+  - name: request_params
+    type: string
+    required: false
+    label:
+      en_US: Request params
+      zh_Hans: 请求参数
+      pt_BR: Request params
+    human_description:
+      en_US: |
+        request parameters, format: {"key1": "value1", "key2": "value2"}
+      zh_Hans: |
+        请求参数，格式：{"key1": "value1", "key2": "value2"}
+      pt_BR: |
+        request parameters, format: {"key1": "value1", "key2": "value2"}
+    llm_description: request parameters
+    form: llm
   - name: target_selector
     type: string
     required: false

api/core/tools/provider/builtin/slack/tools/slack_webhook.py

@@ -20,7 +20,7 @@ class SlackWebhookTool(BuiltinTool):
 
         webhook_url = tool_parameters.get('webhook_url', '')
 
-        if not webhook_url.startswith('https://hooks.slack.com/services/'):
+        if not webhook_url.startswith('https://hooks.slack.com/'):
             return self.create_text_message(
                 f'Invalid parameter webhook_url ${webhook_url}, not a valid Slack webhook URL')
 

api/core/tools/tool/api_tool.py

@@ -249,9 +249,12 @@ class ApiTool(Tool):
                 elif property['type'] == 'null':
                     if value is None:
                         return None
-                elif property['type'] == 'object':
+                elif property['type'] == 'object' or property['type'] == 'array':
                     if isinstance(value, str):
                         try:
+                            # an array str like '[1,2]' also can convert to list [1,2] through json.loads
+                            # json not support single quote, but we can support it
+                            value = value.replace("'", '"')
                             return json.loads(value)
                         except ValueError:
                             return value

api/core/tools/tool_manager.py

@@ -154,7 +154,7 @@ class ToolManager:
                 'invoke_from': invoke_from,
                 'tool_invoke_from': tool_invoke_from,
             })
-        
+
         elif provider_type == 'api':
             if tenant_id is None:
                 raise ValueError('tenant id is required for api provider')
@@ -201,7 +201,7 @@ class ToolManager:
             init runtime parameter
         """
         parameter_value = parameters.get(parameter_rule.name)
-        if not parameter_value:
+        if not parameter_value and parameter_value != 0:
             # get default value
             parameter_value = parameter_rule.default
             if not parameter_value and parameter_rule.required:
@@ -321,14 +321,14 @@ class ToolManager:
         if cls._builtin_providers_loaded:
             yield from list(cls._builtin_providers.values())
             return
-        
+
         with cls._builtin_provider_lock:
             if cls._builtin_providers_loaded:
                 yield from list(cls._builtin_providers.values())
                 return
-            
+
             yield from cls._list_builtin_providers()
-    
+
     @classmethod
     def _list_builtin_providers(cls) -> Generator[BuiltinToolProviderController, None, None]:
         """
@@ -492,7 +492,7 @@ class ToolManager:
 
         controller = ApiToolProviderController.from_db(
             provider,
-            ApiProviderAuthType.API_KEY if provider.credentials['auth_type'] == 'api_key' else 
+            ApiProviderAuthType.API_KEY if provider.credentials['auth_type'] == 'api_key' else
             ApiProviderAuthType.NONE
         )
         controller.load_bundled_tools(provider.tools)

api/core/workflow/nodes/code/code_node.py

@@ -1,6 +1,6 @@
-import os
 from typing import Optional, Union, cast
 
+from configs import dify_config
 from core.helper.code_executor.code_executor import CodeExecutionException, CodeExecutor, CodeLanguage
 from core.helper.code_executor.code_node_provider import CodeNodeProvider
 from core.helper.code_executor.javascript.javascript_code_provider import JavascriptCodeProvider
@@ -11,14 +11,14 @@ from core.workflow.nodes.base_node import BaseNode
 from core.workflow.nodes.code.entities import CodeNodeData
 from models.workflow import WorkflowNodeExecutionStatus
 
-MAX_NUMBER = int(os.environ.get('CODE_MAX_NUMBER', '9223372036854775807'))
-MIN_NUMBER = int(os.environ.get('CODE_MIN_NUMBER', '-9223372036854775808'))
+MAX_NUMBER = dify_config.CODE_MAX_NUMBER
+MIN_NUMBER = dify_config.CODE_MIN_NUMBER
 MAX_PRECISION = 20
 MAX_DEPTH = 5
-MAX_STRING_LENGTH = int(os.environ.get('CODE_MAX_STRING_LENGTH', '80000'))
-MAX_STRING_ARRAY_LENGTH = int(os.environ.get('CODE_MAX_STRING_ARRAY_LENGTH', '30'))
-MAX_OBJECT_ARRAY_LENGTH = int(os.environ.get('CODE_MAX_OBJECT_ARRAY_LENGTH', '30'))
-MAX_NUMBER_ARRAY_LENGTH = int(os.environ.get('CODE_MAX_NUMBER_ARRAY_LENGTH', '1000'))
+MAX_STRING_LENGTH = dify_config.CODE_MAX_STRING_LENGTH
+MAX_STRING_ARRAY_LENGTH = dify_config.CODE_MAX_STRING_ARRAY_LENGTH
+MAX_OBJECT_ARRAY_LENGTH = dify_config.CODE_MAX_OBJECT_ARRAY_LENGTH
+MAX_NUMBER_ARRAY_LENGTH = dify_config.CODE_MAX_NUMBER_ARRAY_LENGTH
 
 
 class CodeNode(BaseNode):

api/core/workflow/nodes/http_request/entities.py

@@ -1,57 +1,61 @@
-import os
 from typing import Literal, Optional, Union
 
 from pydantic import BaseModel, ValidationInfo, field_validator
 
+from configs import dify_config
 from core.workflow.entities.base_node_data_entities import BaseNodeData
 
-MAX_CONNECT_TIMEOUT = int(os.environ.get('HTTP_REQUEST_MAX_CONNECT_TIMEOUT', '300'))
-MAX_READ_TIMEOUT = int(os.environ.get('HTTP_REQUEST_MAX_READ_TIMEOUT', '600'))
-MAX_WRITE_TIMEOUT = int(os.environ.get('HTTP_REQUEST_MAX_WRITE_TIMEOUT', '600'))
+MAX_CONNECT_TIMEOUT = dify_config.HTTP_REQUEST_MAX_CONNECT_TIMEOUT
+MAX_READ_TIMEOUT = dify_config.HTTP_REQUEST_MAX_READ_TIMEOUT
+MAX_WRITE_TIMEOUT = dify_config.HTTP_REQUEST_MAX_WRITE_TIMEOUT
+
+
+class HttpRequestNodeAuthorizationConfig(BaseModel):
+    type: Literal[None, 'basic', 'bearer', 'custom']
+    api_key: Union[None, str] = None
+    header: Union[None, str] = None
+
+
+class HttpRequestNodeAuthorization(BaseModel):
+    type: Literal['no-auth', 'api-key']
+    config: Optional[HttpRequestNodeAuthorizationConfig] = None
+
+    @field_validator('config', mode='before')
+    @classmethod
+    def check_config(cls, v: HttpRequestNodeAuthorizationConfig, values: ValidationInfo):
+        """
+        Check config, if type is no-auth, config should be None, otherwise it should be a dict.
+        """
+        if values.data['type'] == 'no-auth':
+            return None
+        else:
+            if not v or not isinstance(v, dict):
+                raise ValueError('config should be a dict')
+
+            return v
+
+
+class HttpRequestNodeBody(BaseModel):
+    type: Literal['none', 'form-data', 'x-www-form-urlencoded', 'raw-text', 'json']
+    data: Union[None, str] = None
+
+
+class HttpRequestNodeTimeout(BaseModel):
+    connect: int = MAX_CONNECT_TIMEOUT
+    read: int = MAX_READ_TIMEOUT
+    write: int = MAX_WRITE_TIMEOUT
+
 
 class HttpRequestNodeData(BaseNodeData):
     """
     Code Node Data.
     """
-    class Authorization(BaseModel):
-        # TODO[pydantic]: The `Config` class inherits from another class, please create the `model_config` manually.
-        # Check https://docs.pydantic.dev/dev-v2/migration/#changes-to-config for more information.
-        class Config(BaseModel):
-            type: Literal[None, 'basic', 'bearer', 'custom']
-            api_key: Union[None, str] = None
-            header: Union[None, str] = None
-
-        type: Literal['no-auth', 'api-key']
-        config: Optional[Config] = None
-
-        @field_validator('config', mode='before')
-        @classmethod
-        def check_config(cls, v: Config, values: ValidationInfo):
-            """
-            Check config, if type is no-auth, config should be None, otherwise it should be a dict.
-            """
-            if values.data['type'] == 'no-auth':
-                return None
-            else:
-                if not v or not isinstance(v, dict):
-                    raise ValueError('config should be a dict')
-                
-                return v
-
-    class Body(BaseModel):
-        type: Literal['none', 'form-data', 'x-www-form-urlencoded', 'raw-text', 'json']
-        data: Union[None, str] = None
-
-    class Timeout(BaseModel):
-        connect: Optional[int] = MAX_CONNECT_TIMEOUT
-        read:  Optional[int] = MAX_READ_TIMEOUT
-        write:  Optional[int] = MAX_WRITE_TIMEOUT
 
     method: Literal['get', 'post', 'put', 'patch', 'delete', 'head']
     url: str
-    authorization: Authorization
+    authorization: HttpRequestNodeAuthorization
     headers: str
     params: str
-    body: Optional[Body] = None
-    timeout: Optional[Timeout] = None
+    body: Optional[HttpRequestNodeBody] = None
+    timeout: Optional[HttpRequestNodeTimeout] = None
     mask_authorization_header: Optional[bool] = True

api/core/workflow/nodes/http_request/http_executor.py

@@ -1,5 +1,4 @@
 import json
-import os
 from copy import deepcopy
 from random import randint
 from typing import Any, Optional, Union
@@ -8,22 +7,28 @@ from urllib.parse import urlencode
 import httpx
 
 import core.helper.ssrf_proxy as ssrf_proxy
+from configs import dify_config
 from core.workflow.entities.variable_entities import VariableSelector
 from core.workflow.entities.variable_pool import ValueType, VariablePool
-from core.workflow.nodes.http_request.entities import HttpRequestNodeData
+from core.workflow.nodes.http_request.entities import (
+    HttpRequestNodeAuthorization,
+    HttpRequestNodeBody,
+    HttpRequestNodeData,
+    HttpRequestNodeTimeout,
+)
 from core.workflow.utils.variable_template_parser import VariableTemplateParser
 
-MAX_BINARY_SIZE = int(os.environ.get('HTTP_REQUEST_NODE_MAX_BINARY_SIZE', 1024 * 1024 * 10))  # 10MB
-READABLE_MAX_BINARY_SIZE = f'{MAX_BINARY_SIZE / 1024 / 1024:.2f}MB'
-MAX_TEXT_SIZE = int(os.environ.get('HTTP_REQUEST_NODE_MAX_TEXT_SIZE', 1024 * 1024))  # 1MB
-READABLE_MAX_TEXT_SIZE = f'{MAX_TEXT_SIZE / 1024 / 1024:.2f}MB'
+MAX_BINARY_SIZE = dify_config.HTTP_REQUEST_NODE_MAX_BINARY_SIZE
+READABLE_MAX_BINARY_SIZE = dify_config.HTTP_REQUEST_NODE_READABLE_MAX_BINARY_SIZE
+MAX_TEXT_SIZE = dify_config.HTTP_REQUEST_NODE_MAX_TEXT_SIZE
+READABLE_MAX_TEXT_SIZE = dify_config.HTTP_REQUEST_NODE_READABLE_MAX_TEXT_SIZE
 
 
 class HttpExecutorResponse:
     headers: dict[str, str]
     response: httpx.Response
 
-    def __init__(self, response: httpx.Response = None):
+    def __init__(self, response: httpx.Response):
         self.response = response
         self.headers = dict(response.headers) if isinstance(self.response, httpx.Response) else {}
 
@@ -40,7 +45,6 @@ class HttpExecutorResponse:
     def get_content_type(self) -> str:
         return self.headers.get('content-type', '')
 
-
     def extract_file(self) -> tuple[str, bytes]:
         """
         extract file from response if content type is file related
@@ -88,17 +92,21 @@ class HttpExecutorResponse:
 class HttpExecutor:
     server_url: str
     method: str
-    authorization: HttpRequestNodeData.Authorization
+    authorization: HttpRequestNodeAuthorization
     params: dict[str, Any]
     headers: dict[str, Any]
     body: Union[None, str]
     files: Union[None, dict[str, Any]]
     boundary: str
     variable_selectors: list[VariableSelector]
-    timeout: HttpRequestNodeData.Timeout
+    timeout: HttpRequestNodeTimeout
 
-    def __init__(self, node_data: HttpRequestNodeData, timeout: HttpRequestNodeData.Timeout,
-                 variable_pool: Optional[VariablePool] = None):
+    def __init__(
+        self,
+        node_data: HttpRequestNodeData,
+        timeout: HttpRequestNodeTimeout,
+        variable_pool: Optional[VariablePool] = None,
+    ):
         self.server_url = node_data.url
         self.method = node_data.method
         self.authorization = node_data.authorization
@@ -113,11 +121,11 @@ class HttpExecutor:
         self._init_template(node_data, variable_pool)
 
     @staticmethod
-    def _is_json_body(body: HttpRequestNodeData.Body):
+    def _is_json_body(body: HttpRequestNodeBody):
         """
         check if body is json
         """
-        if body and body.type == 'json':
+        if body and body.type == 'json' and body.data:
             try:
                 json.loads(body.data)
                 return True
@@ -146,7 +154,6 @@ class HttpExecutor:
         return result
 
     def _init_template(self, node_data: HttpRequestNodeData, variable_pool: Optional[VariablePool] = None):
-
         # extract all template in url
         self.server_url, server_url_variable_selectors = self._format_template(node_data.url, variable_pool)
 
@@ -178,9 +185,7 @@ class HttpExecutor:
                 body = self._to_dict(body_data)
 
                 if node_data.body.type == 'form-data':
-                    self.files = {
-                        k: ('', v) for k, v in body.items()
-                    }
+                    self.files = {k: ('', v) for k, v in body.items()}
                     random_str = lambda n: ''.join([chr(randint(97, 122)) for _ in range(n)])
                     self.boundary = f'----WebKitFormBoundary{random_str(16)}'
 
@@ -192,13 +197,24 @@ class HttpExecutor:
             elif node_data.body.type == 'none':
                 self.body = ''
 
-        self.variable_selectors = (server_url_variable_selectors + params_variable_selectors
-                                   + headers_variable_selectors + body_data_variable_selectors)
+        self.variable_selectors = (
+            server_url_variable_selectors
+            + params_variable_selectors
+            + headers_variable_selectors
+            + body_data_variable_selectors
+        )
 
     def _assembling_headers(self) -> dict[str, Any]:
         authorization = deepcopy(self.authorization)
         headers = deepcopy(self.headers) or {}
         if self.authorization.type == 'api-key':
+            if self.authorization.config is None:
+                raise ValueError('self.authorization config is required')
+            if authorization.config is None:
+                raise ValueError('authorization config is required')
+            if authorization.config.type != 'bearer' and authorization.config.header is None:
+                raise ValueError('authorization config header is required')
+
             if self.authorization.config.api_key is None:
                 raise ValueError('api_key is required')
 
@@ -216,7 +232,7 @@ class HttpExecutor:
 
     def _validate_and_parse_response(self, response: httpx.Response) -> HttpExecutorResponse:
         """
-            validate the response
+        validate the response
         """
         if isinstance(response, httpx.Response):
             executor_response = HttpExecutorResponse(response)
@@ -226,24 +242,26 @@ class HttpExecutor:
         if executor_response.is_file:
             if executor_response.size > MAX_BINARY_SIZE:
                 raise ValueError(
-                    f'File size is too large, max size is {READABLE_MAX_BINARY_SIZE}, but current size is {executor_response.readable_size}.')
+                    f'File size is too large, max size is {READABLE_MAX_BINARY_SIZE}, but current size is {executor_response.readable_size}.'
+                )
         else:
             if executor_response.size > MAX_TEXT_SIZE:
                 raise ValueError(
-                    f'Text size is too large, max size is {READABLE_MAX_TEXT_SIZE}, but current size is {executor_response.readable_size}.')
+                    f'Text size is too large, max size is {READABLE_MAX_TEXT_SIZE}, but current size is {executor_response.readable_size}.'
+                )
 
         return executor_response
 
     def _do_http_request(self, headers: dict[str, Any]) -> httpx.Response:
         """
-            do http request depending on api bundle
+        do http request depending on api bundle
         """
         kwargs = {
             'url': self.server_url,
             'headers': headers,
             'params': self.params,
             'timeout': (self.timeout.connect, self.timeout.read, self.timeout.write),
-            'follow_redirects': True
+            'follow_redirects': True,
         }
 
         if self.method in ('get', 'head', 'post', 'put', 'delete', 'patch'):
@@ -306,8 +324,9 @@ class HttpExecutor:
 
         return raw_request
 
-    def _format_template(self, template: str, variable_pool: VariablePool, escape_quotes: bool = False) \
-            -> tuple[str, list[VariableSelector]]:
+    def _format_template(
+        self, template: str, variable_pool: Optional[VariablePool], escape_quotes: bool = False
+    ) -> tuple[str, list[VariableSelector]]:
         """
         format template
         """
@@ -318,14 +337,13 @@ class HttpExecutor:
             variable_value_mapping = {}
             for variable_selector in variable_selectors:
                 value = variable_pool.get_variable_value(
-                    variable_selector=variable_selector.value_selector,
-                    target_value_type=ValueType.STRING
+                    variable_selector=variable_selector.value_selector, target_value_type=ValueType.STRING
                 )
 
                 if value is None:
                     raise ValueError(f'Variable {variable_selector.variable} not found')
 
-                if escape_quotes:
+                if escape_quotes and isinstance(value, str):
                     value = value.replace('"', '\\"')
 
                 variable_value_mapping[variable_selector.variable] = value

api/core/workflow/nodes/http_request/http_request_node.py

@@ -5,6 +5,7 @@ from typing import cast
 
 from core.file.file_obj import FileTransferMethod, FileType, FileVar
 from core.tools.tool_file_manager import ToolFileManager
+from core.workflow.entities.base_node_data_entities import BaseNodeData
 from core.workflow.entities.node_entities import NodeRunResult, NodeType
 from core.workflow.entities.variable_pool import VariablePool
 from core.workflow.nodes.base_node import BaseNode
@@ -13,49 +14,50 @@ from core.workflow.nodes.http_request.entities import (
     MAX_READ_TIMEOUT,
     MAX_WRITE_TIMEOUT,
     HttpRequestNodeData,
+    HttpRequestNodeTimeout,
 )
 from core.workflow.nodes.http_request.http_executor import HttpExecutor, HttpExecutorResponse
 from models.workflow import WorkflowNodeExecutionStatus
 
-HTTP_REQUEST_DEFAULT_TIMEOUT = HttpRequestNodeData.Timeout(connect=min(10, MAX_CONNECT_TIMEOUT),
-                                                           read=min(60, MAX_READ_TIMEOUT),
-                                                           write=min(20, MAX_WRITE_TIMEOUT))
+HTTP_REQUEST_DEFAULT_TIMEOUT = HttpRequestNodeTimeout(
+    connect=min(10, MAX_CONNECT_TIMEOUT),
+    read=min(60, MAX_READ_TIMEOUT),
+    write=min(20, MAX_WRITE_TIMEOUT),
+)
 
 
 class HttpRequestNode(BaseNode):
     _node_data_cls = HttpRequestNodeData
-    node_type = NodeType.HTTP_REQUEST
+    _node_type = NodeType.HTTP_REQUEST
 
     @classmethod
-    def get_default_config(cls) -> dict:
+    def get_default_config(cls, filters: dict | None = None) -> dict:
         return {
-            "type": "http-request",
-            "config": {
-                "method": "get",
-                "authorization": {
-                    "type": "no-auth",
+            'type': 'http-request',
+            'config': {
+                'method': 'get',
+                'authorization': {
+                    'type': 'no-auth',
                 },
-                "body": {
-                    "type": "none"
-                },
-                "timeout": {
+                'body': {'type': 'none'},
+                'timeout': {
                     **HTTP_REQUEST_DEFAULT_TIMEOUT.model_dump(),
-                    "max_connect_timeout": MAX_CONNECT_TIMEOUT,
-                    "max_read_timeout": MAX_READ_TIMEOUT,
-                    "max_write_timeout": MAX_WRITE_TIMEOUT,
-                }
+                    'max_connect_timeout': MAX_CONNECT_TIMEOUT,
+                    'max_read_timeout': MAX_READ_TIMEOUT,
+                    'max_write_timeout': MAX_WRITE_TIMEOUT,
+                },
             },
         }
 
     def _run(self, variable_pool: VariablePool) -> NodeRunResult:
-        node_data: HttpRequestNodeData = cast(self._node_data_cls, self.node_data)
+        node_data: HttpRequestNodeData = cast(HttpRequestNodeData, self.node_data)
 
         # init http executor
         http_executor = None
         try:
-            http_executor = HttpExecutor(node_data=node_data,
-                                         timeout=self._get_request_timeout(node_data),
-                                         variable_pool=variable_pool)
+            http_executor = HttpExecutor(
+                node_data=node_data, timeout=self._get_request_timeout(node_data), variable_pool=variable_pool
+            )
 
             # invoke http executor
             response = http_executor.invoke()
@@ -70,7 +72,7 @@ class HttpRequestNode(BaseNode):
             return NodeRunResult(
                 status=WorkflowNodeExecutionStatus.FAILED,
                 error=str(e),
-                process_data=process_data
+                process_data=process_data,
             )
 
         files = self.extract_files(http_executor.server_url, response)
@@ -85,34 +87,32 @@ class HttpRequestNode(BaseNode):
             },
             process_data={
                 'request': http_executor.to_raw_request(
-                    mask_authorization_header=node_data.mask_authorization_header
+                    mask_authorization_header=node_data.mask_authorization_header,
                 ),
-            }
+            },
         )
 
-    def _get_request_timeout(self, node_data: HttpRequestNodeData) -> HttpRequestNodeData.Timeout:
+    def _get_request_timeout(self, node_data: HttpRequestNodeData) -> HttpRequestNodeTimeout:
         timeout = node_data.timeout
         if timeout is None:
             return HTTP_REQUEST_DEFAULT_TIMEOUT
 
-        if timeout.connect is None:
-            timeout.connect = HTTP_REQUEST_DEFAULT_TIMEOUT.connect
+        timeout.connect = timeout.connect or HTTP_REQUEST_DEFAULT_TIMEOUT.connect
         timeout.connect = min(timeout.connect, MAX_CONNECT_TIMEOUT)
-        if timeout.read is None:
-            timeout.read = HTTP_REQUEST_DEFAULT_TIMEOUT.read
+        timeout.read = timeout.read or HTTP_REQUEST_DEFAULT_TIMEOUT.read
         timeout.read = min(timeout.read, MAX_READ_TIMEOUT)
-        if timeout.write is None:
-            timeout.write = HTTP_REQUEST_DEFAULT_TIMEOUT.write
+        timeout.write = timeout.write or HTTP_REQUEST_DEFAULT_TIMEOUT.write
         timeout.write = min(timeout.write, MAX_WRITE_TIMEOUT)
         return timeout
 
     @classmethod
-    def _extract_variable_selector_to_variable_mapping(cls, node_data: HttpRequestNodeData) -> dict[str, list[str]]:
+    def _extract_variable_selector_to_variable_mapping(cls, node_data: BaseNodeData) -> dict[str, list[str]]:
         """
         Extract variable selector to variable mapping
         :param node_data: node data
         :return:
         """
+        node_data = cast(HttpRequestNodeData, node_data)
         try:
             http_executor = HttpExecutor(node_data=node_data, timeout=HTTP_REQUEST_DEFAULT_TIMEOUT)
 
@@ -124,7 +124,7 @@ class HttpRequestNode(BaseNode):
 
             return variable_mapping
         except Exception as e:
-            logging.exception(f"Failed to extract variable selector to variable mapping: {e}")
+            logging.exception(f'Failed to extract variable selector to variable mapping: {e}')
             return {}
 
     def extract_files(self, url: str, response: HttpExecutorResponse) -> list[FileVar]:
@@ -144,21 +144,23 @@ class HttpRequestNode(BaseNode):
             extension = guess_extension(mimetype) or '.bin'
 
             tool_file = ToolFileManager.create_file_by_raw(
-                user_id=self.user_id, 
-                tenant_id=self.tenant_id, 
-                conversation_id=None, 
-                file_binary=file_binary, 
+                user_id=self.user_id,
+                tenant_id=self.tenant_id,
+                conversation_id=None,
+                file_binary=file_binary,
                 mimetype=mimetype,
             )
 
-            files.append(FileVar(
-                tenant_id=self.tenant_id,
-                type=FileType.IMAGE,
-                transfer_method=FileTransferMethod.TOOL_FILE,
-                related_id=tool_file.id,
-                filename=filename,
-                extension=extension,
-                mime_type=mimetype,
-            ))
+            files.append(
+                FileVar(
+                    tenant_id=self.tenant_id,
+                    type=FileType.IMAGE,
+                    transfer_method=FileTransferMethod.TOOL_FILE,
+                    related_id=tool_file.id,
+                    filename=filename,
+                    extension=extension,
+                    mime_type=mimetype,
+                )
+            )
 
         return files

api/core/workflow/nodes/parameter_extractor/parameter_extractor_node.py

@@ -365,7 +365,7 @@ class ParameterExtractorNode(LLMNode):
             files=[],
             context='',
             memory_config=node_data.memory,
-            memory=memory,
+            memory=None,
             model_config=model_config
         )
 

api/extensions/storage/azure_storage.py

@@ -1,5 +1,4 @@
 from collections.abc import Generator
-from contextlib import closing
 from datetime import datetime, timedelta, timezone
 
 from azure.storage.blob import AccountSasPermissions, BlobServiceClient, ResourceTypes, generate_account_sas
@@ -38,10 +37,9 @@ class AzureStorage(BaseStorage):
 
         def generate(filename: str = filename) -> Generator:
             blob = client.get_blob_client(container=self.bucket_name, blob=filename)
-            with closing(blob.download_blob()) as blob_stream:
-                while chunk := blob_stream.readall():
-                    yield chunk
-
+            blob_data = blob.download_blob()
+            for chunk in blob_data.chunks():
+                yield from chunk
         return generate()
 
     def download(self, filename, target_filepath):

api/libs/helper.py

@@ -1,18 +1,23 @@
 import json
+import logging
 import random
 import re
 import string
 import subprocess
+import time
 import uuid
 from collections.abc import Generator
 from datetime import datetime
 from hashlib import sha256
-from typing import Union
+from typing import Any, Optional, Union
 from zoneinfo import available_timezones
 
-from flask import Response, stream_with_context
+from flask import Response, current_app, stream_with_context
 from flask_restful import fields
 
+from extensions.ext_redis import redis_client
+from models.account import Account
+
 
 def run(script):
     return subprocess.getstatusoutput('source /root/.bashrc && ' + script)
@@ -46,12 +51,12 @@ def uuid_value(value):
         error = ('{value} is not a valid uuid.'
                  .format(value=value))
         raise ValueError(error)
-    
+
 def alphanumeric(value: str):
     # check if the value is alphanumeric and underlined
     if re.match(r'^[a-zA-Z0-9_]+$', value):
         return value
-    
+
     raise ValueError(f'{value} is not a valid alphanumeric value')
 
 def timestamp_value(timestamp):
@@ -163,3 +168,97 @@ def compact_generate_response(response: Union[dict, Generator]) -> Response:
 
         return Response(stream_with_context(generate()), status=200,
                         mimetype='text/event-stream')
+
+
+class TokenManager:
+
+    @classmethod
+    def generate_token(cls, account: Account, token_type: str, additional_data: dict = None) -> str:
+        old_token = cls._get_current_token_for_account(account.id, token_type)
+        if old_token:
+            if isinstance(old_token, bytes):
+                old_token = old_token.decode('utf-8')
+            cls.revoke_token(old_token, token_type)
+
+        token = str(uuid.uuid4())
+        token_data = {
+            'account_id': account.id,
+            'email': account.email,
+            'token_type': token_type
+        }
+        if additional_data:
+            token_data.update(additional_data)
+
+        expiry_hours = current_app.config[f'{token_type.upper()}_TOKEN_EXPIRY_HOURS']
+        token_key = cls._get_token_key(token, token_type)
+        redis_client.setex(
+            token_key,
+            expiry_hours * 60 * 60,
+            json.dumps(token_data)
+        )
+
+        cls._set_current_token_for_account(account.id, token, token_type, expiry_hours)
+        return token
+
+    @classmethod
+    def _get_token_key(cls, token: str, token_type: str) -> str:
+        return f'{token_type}:token:{token}'
+
+    @classmethod
+    def revoke_token(cls, token: str, token_type: str):
+        token_key = cls._get_token_key(token, token_type)
+        redis_client.delete(token_key)
+
+    @classmethod
+    def get_token_data(cls, token: str, token_type: str) -> Optional[dict[str, Any]]:
+        key = cls._get_token_key(token, token_type)
+        token_data_json = redis_client.get(key)
+        if token_data_json is None:
+            logging.warning(f"{token_type} token {token} not found with key {key}")
+            return None
+        token_data = json.loads(token_data_json)
+        return token_data
+
+    @classmethod
+    def _get_current_token_for_account(cls, account_id: str, token_type: str) -> Optional[str]:
+        key = cls._get_account_token_key(account_id, token_type)
+        current_token = redis_client.get(key)
+        return current_token
+
+    @classmethod
+    def _set_current_token_for_account(cls, account_id: str, token: str, token_type: str, expiry_hours: int):
+        key = cls._get_account_token_key(account_id, token_type)
+        redis_client.setex(key, expiry_hours * 60 * 60, token)
+
+    @classmethod
+    def _get_account_token_key(cls, account_id: str, token_type: str) -> str:
+        return f'{token_type}:account:{account_id}'
+
+
+class RateLimiter:
+    def __init__(self, prefix: str, max_attempts: int, time_window: int):
+        self.prefix = prefix
+        self.max_attempts = max_attempts
+        self.time_window = time_window
+
+    def _get_key(self, email: str) -> str:
+        return f"{self.prefix}:{email}"
+
+    def is_rate_limited(self, email: str) -> bool:
+        key = self._get_key(email)
+        current_time = int(time.time())
+        window_start_time = current_time - self.time_window
+
+        redis_client.zremrangebyscore(key, '-inf', window_start_time)
+        attempts = redis_client.zcard(key)
+
+        if attempts and int(attempts) >= self.max_attempts:
+            return True
+        return False
+
+    def increment_rate_limit(self, email: str):
+        key = self._get_key(email)
+        current_time = int(time.time())
+
+        redis_client.zadd(key, {current_time: current_time})
+        redis_client.expire(key, self.time_window * 2)

api/pyproject.toml

@@ -115,7 +115,6 @@ chardet = "~5.1.0"
 cohere = "~5.2.4"
 cos-python-sdk-v5 = "1.9.30"
 dashscope = { version = "~1.17.0", extras = ["tokenizer"] }
-firecrawl-py = "0.0.5"
 flask = "~3.0.1"
 flask-compress = "~1.14"
 flask-cors = "~4.0.0"
@@ -164,7 +163,6 @@ redis = { version = "~5.0.3", extras = ["hiredis"] }
 replicate = "~0.22.0"
 resend = "~0.7.0"
 safetensors = "~0.4.3"
-scikit-learn = "1.2.2"
 sentry-sdk = { version = "~1.39.2", extras = ["flask"] }
 sqlalchemy = "~2.0.29"
 tencentcloud-sdk-python-hunyuan = "~3.0.1158"
@@ -186,7 +184,7 @@ zhipuai = "1.0.7"
 arxiv = "2.1.0"
 matplotlib = "~3.8.2"
 newspaper3k = "0.2.8"
-duckduckgo-search = "~6.1.5"
+duckduckgo-search = "^6.1.8"
 jsonpath-ng = "1.6.1"
 numexpr = "~2.9.0"
 opensearch-py = "2.4.0"

api/services/account_service.py

@@ -13,6 +13,7 @@ from werkzeug.exceptions import Unauthorized
 from constants.languages import language_timezone_mapping, languages
 from events.tenant_event import tenant_was_created
 from extensions.ext_redis import redis_client
+from libs.helper import RateLimiter, TokenManager
 from libs.passport import PassportService
 from libs.password import compare_password, hash_password, valid_password
 from libs.rsa import generate_key_pair
@@ -29,14 +30,22 @@ from services.errors.account import (
     LinkAccountIntegrateError,
     MemberNotInTenantError,
     NoPermissionError,
+    RateLimitExceededError,
     RoleAlreadyAssignedError,
     TenantNotFound,
 )
 from tasks.mail_invite_member_task import send_invite_member_mail_task
+from tasks.mail_reset_password_task import send_reset_password_mail_task
 
 
 class AccountService:
 
+    reset_password_rate_limiter = RateLimiter(
+        prefix="reset_password_rate_limit",
+        max_attempts=5,
+        time_window=60 * 60
+    )
+
     @staticmethod
     def load_user(user_id: str) -> Account:
         account = Account.query.filter_by(id=user_id).first()
@@ -222,9 +231,33 @@ class AccountService:
             return None
         return AccountService.load_user(account_id)
 
+    @classmethod
+    def send_reset_password_email(cls, account):
+        if cls.reset_password_rate_limiter.is_rate_limited(account.email):
+            raise RateLimitExceededError(f"Rate limit exceeded for email: {account.email}. Please try again later.")
+
+        token = TokenManager.generate_token(account, 'reset_password')
+        send_reset_password_mail_task.delay(
+            language=account.interface_language,
+            to=account.email,
+            token=token
+        )
+        cls.reset_password_rate_limiter.increment_rate_limit(account.email)
+        return token
+
+    @classmethod
+    def revoke_reset_password_token(cls, token: str):
+        TokenManager.revoke_token(token, 'reset_password')
+
+    @classmethod
+    def get_reset_password_data(cls, token: str) -> Optional[dict[str, Any]]:
+        return TokenManager.get_token_data(token, 'reset_password')
+
+
 def _get_login_cache_key(*, account_id: str, token: str):
     return f"account_login:{account_id}:{token}"
 
+
 class TenantService:
 
     @staticmethod

api/services/errors/account.py

@@ -51,3 +51,8 @@ class MemberNotInTenantError(BaseServiceError):
 
 class RoleAlreadyAssignedError(BaseServiceError):
     pass
+
+
+class RateLimitExceededError(BaseServiceError):
+    pass
+

api/services/hit_testing_service.py

@@ -1,13 +1,6 @@
 import logging
 import time
 
-import numpy as np
-from sklearn.manifold import TSNE
-
-from core.embedding.cached_embedding import CacheEmbedding
-from core.model_manager import ModelManager
-from core.model_runtime.entities.model_entities import ModelType
-from core.rag.datasource.entity.embedding import Embeddings
 from core.rag.datasource.retrieval_service import RetrievalService
 from core.rag.models.document import Document
 from core.rag.retrieval.retrival_methods import RetrievalMethod
@@ -45,17 +38,6 @@ class HitTestingService:
         if not retrieval_model:
             retrieval_model = dataset.retrieval_model if dataset.retrieval_model else default_retrieval_model
 
-        # get embedding model
-        model_manager = ModelManager()
-        embedding_model = model_manager.get_model_instance(
-            tenant_id=dataset.tenant_id,
-            model_type=ModelType.TEXT_EMBEDDING,
-            provider=dataset.embedding_model_provider,
-            model=dataset.embedding_model
-        )
-
-        embeddings = CacheEmbedding(embedding_model)
-
         all_documents = RetrievalService.retrieve(retrival_method=retrieval_model['search_method'],
                                                   dataset_id=dataset.id,
                                                   query=query,
@@ -80,20 +62,10 @@ class HitTestingService:
         db.session.add(dataset_query)
         db.session.commit()
 
-        return cls.compact_retrieve_response(dataset, embeddings, query, all_documents)
+        return cls.compact_retrieve_response(dataset, query, all_documents)
 
     @classmethod
-    def compact_retrieve_response(cls, dataset: Dataset, embeddings: Embeddings, query: str, documents: list[Document]):
-        text_embeddings = [
-            embeddings.embed_query(query)
-        ]
-
-        text_embeddings.extend(embeddings.embed_documents([document.page_content for document in documents]))
-
-        tsne_position_data = cls.get_tsne_positions_from_embeddings(text_embeddings)
-
-        query_position = tsne_position_data.pop(0)
-
+    def compact_retrieve_response(cls, dataset: Dataset, query: str, documents: list[Document]):
         i = 0
         records = []
         for document in documents:
@@ -113,7 +85,6 @@ class HitTestingService:
             record = {
                 "segment": segment,
                 "score": document.metadata.get('score', None),
-                "tsne_position": tsne_position_data[i]
             }
 
             records.append(record)
@@ -123,34 +94,10 @@ class HitTestingService:
         return {
             "query": {
                 "content": query,
-                "tsne_position": query_position,
             },
             "records": records
         }
 
-    @classmethod
-    def get_tsne_positions_from_embeddings(cls, embeddings: list):
-        embedding_length = len(embeddings)
-        if embedding_length <= 1:
-            return [{'x': 0, 'y': 0}]
-
-        noise = np.random.normal(0, 1e-4, np.array(embeddings).shape)
-        concatenate_data = np.array(embeddings) + noise
-        concatenate_data = concatenate_data.reshape(embedding_length, -1)
-
-        perplexity = embedding_length / 2 + 1
-        if perplexity >= embedding_length:
-            perplexity = max(embedding_length - 1, 1)
-
-        tsne = TSNE(n_components=2, perplexity=perplexity, early_exaggeration=12.0)
-        data_tsne = tsne.fit_transform(concatenate_data)
-
-        tsne_position_data = []
-        for i in range(len(data_tsne)):
-            tsne_position_data.append({'x': float(data_tsne[i][0]), 'y': float(data_tsne[i][1])})
-
-        return tsne_position_data
-
     @classmethod
     def hit_testing_args_check(cls, args):
         query = args['query']

api/services/recommended_app_service.py

@@ -110,7 +110,12 @@ class RecommendedAppService:
         if response.status_code != 200:
             raise ValueError(f'fetch recommended apps failed, status code: {response.status_code}')
 
-        return response.json()
+        result = response.json()
+
+        if "categories" in result:
+            result["categories"] = sorted(result["categories"])
+        
+        return result
 
     @classmethod
     def _fetch_recommended_apps_from_builtin(cls, language: str) -> dict:

api/tasks/mail_invite_member_task.py

@@ -39,16 +39,15 @@ def send_invite_member_mail_task(language: str, to: str, token: str, inviter_nam
             mail.send(to=to, subject="立即加入 Dify 工作空间", html=html_content)
         else:
             html_content = render_template('invite_member_mail_template_en-US.html',
-                                        to=to,
-                                        inviter_name=inviter_name, 
-                                        workspace_name=workspace_name,
-                                        url=url)
+                                           to=to,
+                                           inviter_name=inviter_name,
+                                           workspace_name=workspace_name,
+                                           url=url)
             mail.send(to=to, subject="Join Dify Workspace Now", html=html_content)
-        
 
         end_at = time.perf_counter()
         logging.info(
             click.style('Send invite member mail to {} succeeded: latency: {}'.format(to, end_at - start_at),
                         fg='green'))
     except Exception:
-        logging.exception("Send invite member mail to {} failed".format(to))
+        logging.exception("Send invite member mail to {} failed".format(to))

api/tasks/mail_reset_password_task.py

@@ -0,0 +1,44 @@
+import logging
+import time
+
+import click
+from celery import shared_task
+from flask import current_app, render_template
+
+from extensions.ext_mail import mail
+
+
+@shared_task(queue='mail')
+def send_reset_password_mail_task(language: str, to: str, token: str):
+    """
+    Async Send reset password mail
+    :param language: Language in which the email should be sent (e.g., 'en', 'zh')
+    :param to: Recipient email address
+    :param token: Reset password token to be included in the email
+    """
+    if not mail.is_inited():
+        return
+
+    logging.info(click.style('Start password reset mail to {}'.format(to), fg='green'))
+    start_at = time.perf_counter()
+
+    # send reset password mail using different languages
+    try:
+        url = f'{current_app.config.get("CONSOLE_WEB_URL")}/forgot-password?token={token}'
+        if language == 'zh-Hans':
+            html_content = render_template('reset_password_mail_template_zh-CN.html',
+                                           to=to,
+                                           url=url)
+            mail.send(to=to, subject="重置您的 Dify 密码", html=html_content)
+        else:
+            html_content = render_template('reset_password_mail_template_en-US.html',
+                                           to=to,
+                                           url=url)
+            mail.send(to=to, subject="Reset Your Dify Password", html=html_content)
+
+        end_at = time.perf_counter()
+        logging.info(
+            click.style('Send password reset mail to {} succeeded: latency: {}'.format(to, end_at - start_at),
+                        fg='green'))
+    except Exception:
+        logging.exception("Send password reset mail to {} failed".format(to))

api/templates/reset_password_mail_template_en-US.html

@@ -0,0 +1,72 @@
+<!DOCTYPE html>
+<html>
+<head>
+    <style>
+        body {
+            font-family: 'Arial', sans-serif;
+            line-height: 16pt;
+            color: #374151;
+            background-color: #E5E7EB;
+            margin: 0;
+            padding: 0;
+        }
+        .container {
+            width: 100%;
+            max-width: 560px;
+            margin: 40px auto;
+            padding: 20px;
+            background-color: #F3F4F6;
+            border-radius: 8px;
+            box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
+        }
+        .header {
+            text-align: center;
+            margin-bottom: 20px;
+        }
+        .header img {
+            max-width: 100px;
+            height: auto;
+        }
+        .button {
+            display: inline-block;
+            padding: 12px 24px;
+            background-color: #2970FF;
+            color: white;
+            text-decoration: none;
+            border-radius: 4px;
+            text-align: center;
+            transition: background-color 0.3s ease;
+        }
+        .button:hover {
+            background-color: #265DD4;
+        }
+        .footer {
+            font-size: 0.9em;
+            color: #777777;
+            margin-top: 30px;
+        }
+        .content {
+            margin-top: 20px;
+        }
+    </style>
+</head>
+
+<body>
+    <div class="container">
+        <div class="header">
+            <img src="https://cloud.dify.ai/logo/logo-site.png" alt="Dify Logo">
+        </div>
+        <div class="content">
+            <p>Dear {{ to }},</p>
+            <p>We have received a request to reset your password. If you initiated this request, please click the button below to reset your password:</p>
+            <p style="text-align: center;"><a style="color: #fff; text-decoration: none" class="button" href="{{ url }}">Reset Password</a></p>
+            <p>If you did not request a password reset, please ignore this email and your account will remain secure.</p>
+        </div>
+        <div class="footer">
+            <p>Best regards,</p>
+            <p>Dify Team</p>
+            <p>Please do not reply directly to this email; it is automatically sent by the system.</p>
+        </div>
+    </div>
+</body>
+</html>

api/templates/reset_password_mail_template_zh-CN.html

@@ -0,0 +1,72 @@
+<!DOCTYPE html>
+<html>
+<head>
+    <style>
+        body {
+            font-family: 'Arial', sans-serif;
+            line-height: 16pt;
+            color: #374151;
+            background-color: #E5E7EB;
+            margin: 0;
+            padding: 0;
+        }
+        .container {
+            width: 100%;
+            max-width: 560px;
+            margin: 40px auto;
+            padding: 20px;
+            background-color: #F3F4F6;
+            border-radius: 8px;
+            box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
+        }
+        .header {
+            text-align: center;
+            margin-bottom: 20px;
+        }
+        .header img {
+            max-width: 100px;
+            height: auto;
+        }
+        .button {
+            display: inline-block;
+            padding: 12px 24px;
+            background-color: #2970FF;
+            color: white;
+            text-decoration: none;
+            border-radius: 4px;
+            text-align: center;
+            transition: background-color 0.3s ease;
+        }
+        .button:hover {
+            background-color: #265DD4;
+        }
+        .footer {
+            font-size: 0.9em;
+            color: #777777;
+            margin-top: 30px;
+        }
+        .content {
+            margin-top: 20px;
+        }
+    </style>
+</head>
+
+<body>
+    <div class="container">
+        <div class="header">
+            <img src="https://cloud.dify.ai/logo/logo-site.png" alt="Dify Logo">
+        </div>
+        <div class="content">
+            <p>尊敬的 {{ to }}，</p>
+            <p>我们收到了您关于重置密码的请求。如果是您本人操作，请点击以下按钮重置您的密码：</p>
+            <p style="text-align: center;"><a style="color: #fff; text-decoration: none" class="button" href="{{ url }}">重置密码</a></p>
+            <p>如果您没有请求重置密码，请忽略此邮件，您的账户信息将保持安全。</p>
+        </div>
+        <div class="footer">
+            <p>此致，</p>
+            <p>Dify 团队</p>
+            <p>请不要直接回复此电子邮件；由系统自动发送。</p>
+        </div>
+    </div>
+</body>
+</html>