feat: [backend] vision support (#1510)

Co-authored-by: Garfield Dai <dai.hai@foxmail.com>
2025-12-14 13:26:52 +08:00 · 2023-11-13 22:05:46 +08:00
parent d0e1ea8f06
commit 41d0a8b295
61 changed files with 1563 additions and 300 deletions
--- a/api/config.py
+++ b/api/config.py
@@ -26,6 +26,7 @@ DEFAULTS = {
    'SERVICE_API_URL': 'https://api.dify.ai',
    'APP_WEB_URL': 'https://udify.app',
    'APP_API_URL': 'https://udify.app',
+    'FILES_URL': '',
    'STORAGE_TYPE': 'local',
    'STORAGE_LOCAL_PATH': 'storage',
    'CHECK_UPDATE_URL': 'https://updates.dify.ai',
@@ -57,7 +58,9 @@ DEFAULTS = {
    'CLEAN_DAY_SETTING': 30,
    'UPLOAD_FILE_SIZE_LIMIT': 15,
    'UPLOAD_FILE_BATCH_LIMIT': 5,
-    'OUTPUT_MODERATION_BUFFER_SIZE': 300
+    'UPLOAD_IMAGE_FILE_SIZE_LIMIT': 10,
+    'OUTPUT_MODERATION_BUFFER_SIZE': 300,
+    'MULTIMODAL_SEND_IMAGE_FORMAT': 'base64'
 }


@@ -84,15 +87,9 @@ class Config:
    """Application configuration class."""

    def __init__(self):
-        # app settings
-        self.CONSOLE_API_URL = get_env('CONSOLE_URL') if get_env('CONSOLE_URL') else get_env('CONSOLE_API_URL')
-        self.CONSOLE_WEB_URL = get_env('CONSOLE_URL') if get_env('CONSOLE_URL') else get_env('CONSOLE_WEB_URL')
-        self.SERVICE_API_URL = get_env('API_URL') if get_env('API_URL') else get_env('SERVICE_API_URL')
-        self.APP_WEB_URL = get_env('APP_URL') if get_env('APP_URL') else get_env('APP_WEB_URL')
-        self.APP_API_URL = get_env('APP_URL') if get_env('APP_URL') else get_env('APP_API_URL')
-        self.CONSOLE_URL = get_env('CONSOLE_URL')
-        self.API_URL = get_env('API_URL')
-        self.APP_URL = get_env('APP_URL')
+        # ------------------------
+        # General Configurations.
+        # ------------------------
        self.CURRENT_VERSION = "0.3.29"
        self.COMMIT_SHA = get_env('COMMIT_SHA')
        self.EDITION = "SELF_HOSTED"
@@ -100,70 +97,55 @@ class Config:
        self.TESTING = False
        self.LOG_LEVEL = get_env('LOG_LEVEL')

+        # The backend URL prefix of the console API.
+        # used to concatenate the login authorization callback or notion integration callback.
+        self.CONSOLE_API_URL = get_env('CONSOLE_URL') if get_env('CONSOLE_URL') else get_env('CONSOLE_API_URL')
+
+        # The front-end URL prefix of the console web.
+        # used to concatenate some front-end addresses and for CORS configuration use.
+        self.CONSOLE_WEB_URL = get_env('CONSOLE_URL') if get_env('CONSOLE_URL') else get_env('CONSOLE_WEB_URL')
+
+        # WebApp API backend Url prefix.
+        # used to declare the back-end URL for the front-end API.
+        self.APP_API_URL = get_env('APP_URL') if get_env('APP_URL') else get_env('APP_API_URL')
+
+        # WebApp Url prefix.
+        # used to display WebAPP API Base Url to the front-end.
+        self.APP_WEB_URL = get_env('APP_URL') if get_env('APP_URL') else get_env('APP_WEB_URL')
+
+        # Service API Url prefix.
+        # used to display Service API Base Url to the front-end.
+        self.SERVICE_API_URL = get_env('API_URL') if get_env('API_URL') else get_env('SERVICE_API_URL')
+
+        # File preview or download Url prefix.
+        # used to display File preview or download Url to the front-end or as Multi-model inputs;
+        # Url is signed and has expiration time.
+        self.FILES_URL = get_env('FILES_URL') if get_env('FILES_URL') else self.CONSOLE_API_URL
+
+        # Fallback Url prefix.
+        # Will be deprecated in the future.
+        self.CONSOLE_URL = get_env('CONSOLE_URL')
+        self.API_URL = get_env('API_URL')
+        self.APP_URL = get_env('APP_URL')
+
        # Your App secret key will be used for securely signing the session cookie
        # Make sure you are changing this key for your deployment with a strong key.
        # You can generate a strong key using `openssl rand -base64 42`.
        # Alternatively you can set it with `SECRET_KEY` environment variable.
        self.SECRET_KEY = get_env('SECRET_KEY')

-        # redis settings
-        self.REDIS_HOST = get_env('REDIS_HOST')
-        self.REDIS_PORT = get_env('REDIS_PORT')
-        self.REDIS_USERNAME = get_env('REDIS_USERNAME')
-        self.REDIS_PASSWORD = get_env('REDIS_PASSWORD')
-        self.REDIS_DB = get_env('REDIS_DB')
-        self.REDIS_USE_SSL = get_bool_env('REDIS_USE_SSL')
-
-        # storage settings
-        self.STORAGE_TYPE = get_env('STORAGE_TYPE')
-        self.STORAGE_LOCAL_PATH = get_env('STORAGE_LOCAL_PATH')
-        self.S3_ENDPOINT = get_env('S3_ENDPOINT')
-        self.S3_BUCKET_NAME = get_env('S3_BUCKET_NAME')
-        self.S3_ACCESS_KEY = get_env('S3_ACCESS_KEY')
-        self.S3_SECRET_KEY = get_env('S3_SECRET_KEY')
-        self.S3_REGION = get_env('S3_REGION')
-
-        # vector store settings, only support weaviate, qdrant
-        self.VECTOR_STORE = get_env('VECTOR_STORE')
-
-        # weaviate settings
-        self.WEAVIATE_ENDPOINT = get_env('WEAVIATE_ENDPOINT')
-        self.WEAVIATE_API_KEY = get_env('WEAVIATE_API_KEY')
-        self.WEAVIATE_GRPC_ENABLED = get_bool_env('WEAVIATE_GRPC_ENABLED')
-        self.WEAVIATE_BATCH_SIZE = int(get_env('WEAVIATE_BATCH_SIZE'))
-
-        # qdrant settings
-        self.QDRANT_URL = get_env('QDRANT_URL')
-        self.QDRANT_API_KEY = get_env('QDRANT_API_KEY')
-
-        # milvus setting
-        self.MILVUS_HOST = get_env('MILVUS_HOST')
-        self.MILVUS_PORT = get_env('MILVUS_PORT')
-        self.MILVUS_USER = get_env('MILVUS_USER')
-        self.MILVUS_PASSWORD = get_env('MILVUS_PASSWORD')
-        self.MILVUS_SECURE = get_env('MILVUS_SECURE')
-
-
        # cors settings
        self.CONSOLE_CORS_ALLOW_ORIGINS = get_cors_allow_origins(
            'CONSOLE_CORS_ALLOW_ORIGINS', self.CONSOLE_WEB_URL)
        self.WEB_API_CORS_ALLOW_ORIGINS = get_cors_allow_origins(
            'WEB_API_CORS_ALLOW_ORIGINS', '*')

-        # mail settings
-        self.MAIL_TYPE = get_env('MAIL_TYPE')
-        self.MAIL_DEFAULT_SEND_FROM = get_env('MAIL_DEFAULT_SEND_FROM')
-        self.RESEND_API_KEY = get_env('RESEND_API_KEY')
-
-        # sentry settings
-        self.SENTRY_DSN = get_env('SENTRY_DSN')
-        self.SENTRY_TRACES_SAMPLE_RATE = float(get_env('SENTRY_TRACES_SAMPLE_RATE'))
-        self.SENTRY_PROFILES_SAMPLE_RATE = float(get_env('SENTRY_PROFILES_SAMPLE_RATE'))
-
        # check update url
        self.CHECK_UPDATE_URL = get_env('CHECK_UPDATE_URL')

-        # database settings
+        # ------------------------
+        # Database Configurations.
+        # ------------------------
        db_credentials = {
            key: get_env(key) for key in
            ['DB_USERNAME', 'DB_PASSWORD', 'DB_HOST', 'DB_PORT', 'DB_DATABASE']
@@ -177,14 +159,102 @@ class Config:

        self.SQLALCHEMY_ECHO = get_bool_env('SQLALCHEMY_ECHO')

-        # celery settings
+        # ------------------------
+        # Redis Configurations.
+        # ------------------------
+        self.REDIS_HOST = get_env('REDIS_HOST')
+        self.REDIS_PORT = get_env('REDIS_PORT')
+        self.REDIS_USERNAME = get_env('REDIS_USERNAME')
+        self.REDIS_PASSWORD = get_env('REDIS_PASSWORD')
+        self.REDIS_DB = get_env('REDIS_DB')
+        self.REDIS_USE_SSL = get_bool_env('REDIS_USE_SSL')
+
+        # ------------------------
+        # Celery worker Configurations.
+        # ------------------------
        self.CELERY_BROKER_URL = get_env('CELERY_BROKER_URL')
        self.CELERY_BACKEND = get_env('CELERY_BACKEND')
        self.CELERY_RESULT_BACKEND = 'db+{}'.format(self.SQLALCHEMY_DATABASE_URI) \
            if self.CELERY_BACKEND == 'database' else self.CELERY_BROKER_URL
        self.BROKER_USE_SSL = self.CELERY_BROKER_URL.startswith('rediss://')

-        # hosted provider credentials
+        # ------------------------
+        # File Storage Configurations.
+        # ------------------------
+        self.STORAGE_TYPE = get_env('STORAGE_TYPE')
+        self.STORAGE_LOCAL_PATH = get_env('STORAGE_LOCAL_PATH')
+        self.S3_ENDPOINT = get_env('S3_ENDPOINT')
+        self.S3_BUCKET_NAME = get_env('S3_BUCKET_NAME')
+        self.S3_ACCESS_KEY = get_env('S3_ACCESS_KEY')
+        self.S3_SECRET_KEY = get_env('S3_SECRET_KEY')
+        self.S3_REGION = get_env('S3_REGION')
+
+        # ------------------------
+        # Vector Store Configurations.
+        # Currently, only support: qdrant, milvus, zilliz, weaviate
+        # ------------------------
+        self.VECTOR_STORE = get_env('VECTOR_STORE')
+
+        # qdrant settings
+        self.QDRANT_URL = get_env('QDRANT_URL')
+        self.QDRANT_API_KEY = get_env('QDRANT_API_KEY')
+
+        # milvus / zilliz setting
+        self.MILVUS_HOST = get_env('MILVUS_HOST')
+        self.MILVUS_PORT = get_env('MILVUS_PORT')
+        self.MILVUS_USER = get_env('MILVUS_USER')
+        self.MILVUS_PASSWORD = get_env('MILVUS_PASSWORD')
+        self.MILVUS_SECURE = get_env('MILVUS_SECURE')
+
+        # weaviate settings
+        self.WEAVIATE_ENDPOINT = get_env('WEAVIATE_ENDPOINT')
+        self.WEAVIATE_API_KEY = get_env('WEAVIATE_API_KEY')
+        self.WEAVIATE_GRPC_ENABLED = get_bool_env('WEAVIATE_GRPC_ENABLED')
+        self.WEAVIATE_BATCH_SIZE = int(get_env('WEAVIATE_BATCH_SIZE'))
+
+        # ------------------------
+        # Mail Configurations.
+        # ------------------------
+        self.MAIL_TYPE = get_env('MAIL_TYPE')
+        self.MAIL_DEFAULT_SEND_FROM = get_env('MAIL_DEFAULT_SEND_FROM')
+        self.RESEND_API_KEY = get_env('RESEND_API_KEY')
+
+        # ------------------------
+        # Sentry Configurations.
+        # ------------------------
+        self.SENTRY_DSN = get_env('SENTRY_DSN')
+        self.SENTRY_TRACES_SAMPLE_RATE = float(get_env('SENTRY_TRACES_SAMPLE_RATE'))
+        self.SENTRY_PROFILES_SAMPLE_RATE = float(get_env('SENTRY_PROFILES_SAMPLE_RATE'))
+
+        # ------------------------
+        # Business Configurations.
+        # ------------------------
+
+        # multi model send image format, support base64, url, default is base64
+        self.MULTIMODAL_SEND_IMAGE_FORMAT = get_env('MULTIMODAL_SEND_IMAGE_FORMAT')
+
+        # Dataset Configurations.
+        self.TENANT_DOCUMENT_COUNT = get_env('TENANT_DOCUMENT_COUNT')
+        self.CLEAN_DAY_SETTING = get_env('CLEAN_DAY_SETTING')
+
+        # File upload Configurations.
+        self.UPLOAD_FILE_SIZE_LIMIT = int(get_env('UPLOAD_FILE_SIZE_LIMIT'))
+        self.UPLOAD_FILE_BATCH_LIMIT = int(get_env('UPLOAD_FILE_BATCH_LIMIT'))
+        self.UPLOAD_IMAGE_FILE_SIZE_LIMIT = int(get_env('UPLOAD_IMAGE_FILE_SIZE_LIMIT'))
+
+        # Moderation in app Configurations.
+        self.OUTPUT_MODERATION_BUFFER_SIZE = int(get_env('OUTPUT_MODERATION_BUFFER_SIZE'))
+
+        # Notion integration setting
+        self.NOTION_CLIENT_ID = get_env('NOTION_CLIENT_ID')
+        self.NOTION_CLIENT_SECRET = get_env('NOTION_CLIENT_SECRET')
+        self.NOTION_INTEGRATION_TYPE = get_env('NOTION_INTEGRATION_TYPE')
+        self.NOTION_INTERNAL_SECRET = get_env('NOTION_INTERNAL_SECRET')
+        self.NOTION_INTEGRATION_TOKEN = get_env('NOTION_INTEGRATION_TOKEN')
+
+        # ------------------------
+        # Platform Configurations.
+        # ------------------------
        self.HOSTED_OPENAI_ENABLED = get_bool_env('HOSTED_OPENAI_ENABLED')
        self.HOSTED_OPENAI_API_KEY = get_env('HOSTED_OPENAI_API_KEY')
        self.HOSTED_OPENAI_API_BASE = get_env('HOSTED_OPENAI_API_BASE')
@@ -212,26 +282,6 @@ class Config:
        self.HOSTED_MODERATION_ENABLED = get_bool_env('HOSTED_MODERATION_ENABLED')
        self.HOSTED_MODERATION_PROVIDERS = get_env('HOSTED_MODERATION_PROVIDERS')

-        self.STRIPE_API_KEY = get_env('STRIPE_API_KEY')
-        self.STRIPE_WEBHOOK_SECRET = get_env('STRIPE_WEBHOOK_SECRET')
-
-        # notion import setting
-        self.NOTION_CLIENT_ID = get_env('NOTION_CLIENT_ID')
-        self.NOTION_CLIENT_SECRET = get_env('NOTION_CLIENT_SECRET')
-        self.NOTION_INTEGRATION_TYPE = get_env('NOTION_INTEGRATION_TYPE')
-        self.NOTION_INTERNAL_SECRET = get_env('NOTION_INTERNAL_SECRET')
-        self.NOTION_INTEGRATION_TOKEN = get_env('NOTION_INTEGRATION_TOKEN')
-
-        self.TENANT_DOCUMENT_COUNT = get_env('TENANT_DOCUMENT_COUNT')
-        self.CLEAN_DAY_SETTING = get_env('CLEAN_DAY_SETTING')
-
-        # uploading settings
-        self.UPLOAD_FILE_SIZE_LIMIT = int(get_env('UPLOAD_FILE_SIZE_LIMIT'))
-        self.UPLOAD_FILE_BATCH_LIMIT = int(get_env('UPLOAD_FILE_BATCH_LIMIT'))
-
-        # moderation settings
-        self.OUTPUT_MODERATION_BUFFER_SIZE = int(get_env('OUTPUT_MODERATION_BUFFER_SIZE'))
-

 class CloudEditionConfig(Config):

@@ -246,18 +296,5 @@ class CloudEditionConfig(Config):
        self.GOOGLE_CLIENT_SECRET = get_env('GOOGLE_CLIENT_SECRET')
        self.OAUTH_REDIRECT_PATH = get_env('OAUTH_REDIRECT_PATH')

-
-class TestConfig(Config):
-
-    def __init__(self):
-        super().__init__()
-
-        self.EDITION = "SELF_HOSTED"
-        self.TESTING = True
-
-        db_credentials = {
-            key: get_env(key) for key in ['DB_USERNAME', 'DB_PASSWORD', 'DB_HOST', 'DB_PORT']
-        }
-
-        # use a different database for testing: dify_test
-        self.SQLALCHEMY_DATABASE_URI = f"postgresql://{db_credentials['DB_USERNAME']}:{db_credentials['DB_PASSWORD']}@{db_credentials['DB_HOST']}:{db_credentials['DB_PORT']}/dify_test"
+        self.STRIPE_API_KEY = get_env('STRIPE_API_KEY')
+        self.STRIPE_WEBHOOK_SECRET = get_env('STRIPE_WEBHOOK_SECRET')