支持gpt-4-gizmo-XXX，动态配置GPTS

.github/workflows/xyhelper-deploy.yml

@@ -1,46 +0,0 @@
-name: xyhelper Build and Push Docker Image
-
-on:
-  release:
-    types: [created]
-  workflow_dispatch:
-    inputs:
-      tag:
-        description: 'Tag Name'
-        required: true
-
-
-jobs:
-  build-and-push:
-    runs-on: ubuntu-latest
-    steps:
-    - uses: actions/checkout@v2
-
-    - name: Set up Docker Buildx
-      uses: docker/setup-buildx-action@v1
-
-    - name: Login to Docker Hub
-      uses: docker/login-action@v1
-      with:
-        username: ${{ secrets.DOCKERHUB_USERNAME }}
-        password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
-    - name: Set tag name
-      id: tag_name
-      run: |
-        if [ "${{ github.event_name }}" = "release" ]; then
-          echo "::set-output name=tag::${GITHUB_REF#refs/tags/}"
-        elif [ "${{ github.event_name }}" = "workflow_dispatch" ]; then
-          echo "::set-output name=tag::${{ github.event.inputs.tag }}"
-        fi
-
-    - name: Build and push Docker image with Release tag
-      uses: docker/build-push-action@v2
-      with:
-        context: .
-        file: ./Dockerfile
-        push: true
-        tags: |
-          yangclivia/pandora-to-api:${{ steps.tag_name.outputs.tag }}
-        platforms: linux/amd64,linux/arm64
-        build-args: TARGETPLATFORM=${{ matrix.platform }}

Readme.md

@@ -2,28 +2,20 @@
 
 > [!IMPORTANT]
 >
-> Respect `xyhelper` ,Respect `ninja` , Respect `Wizerd`！
+> Respect Zhile大佬 , Respect Wizerd！
 
-感谢xyhelper、ninja和Wizerd大佬们的付出，敬礼！！！
+感谢pandoraNext和Wizerd的付出，敬礼！！！
 
 本项目支持：
 
-1. 将 xyhelper  `proxy` 模式下的 `backend-api免费接口` 转为 `/v1/chat/completions` 接口，支持流式和非流式响应。
+1. 将 ninja  `proxy` 模式下的 `backend-api` 转为 `/v1/chat/completions` 接口，支持流式和非流式响应。
 
-2. 将 xyhelper  `proxy` 模式下的 `backend-api免费接口` 转为 `/v1/images/generations` 接口
+2. 将 ninja  `proxy` 模式下的 `backend-api` 转为 `/v1/images/generations` 接口
 
 3. 支持直接把refresh_token作为请求key，方便接入one_api
 
 4. 支持 gpt-4-mobile 、gpt-4-s 、基本所有的GPTS
 
-* **xyhelper 的 免费 backend-api 接口，无需打码**
-
-* **xyhelper接口每分钟最多请求30次，介意请绕行**
-    * 我是开发者,我想自行修改功能->前往源码库 https://github.com/xyhelper/chatgpt-mirror-server
-    * 我没服务器,也没有官网账号,只想使用->前往官网购买我们运营的会员服务 https://www.xyhelper.com.cn
-    * 我想做商业用途,我想自己运营->老板里面请 https://www.xyhelper.com.cn/access
-    * 我有服务器,我想自己部署->请继续阅读本文档(有条件的话给个star吧)
-
 如果本项目对你有帮助的话，请点个小星星吧~
 
 如果有什么在项目的使用过程中的疑惑或需求，欢迎加入 Community Telegram Channel: [Inker 的魔法世界](https://t.me/InkerWorld) 来和大家一起交流一下~
@@ -67,13 +59,13 @@
 ## 注意
 
 > [!CAUTION]
-> 1. 本项目的运行需要 xyhelper 的免费接口
+> 1. 本项目的运行需要 ninja 
 >
-> 2. 本项目实际为将来自 `/v1/chat/completions` 的请求转发到xyhelper免费接口的 `/backend-api/conversation` 接口，因此本项目并不支持高并发操作，请不要接入如 `沉浸式翻译` 等高并发项目。
+> 2. 本项目实际为将来自 `/v1/chat/completions` 的请求转发到ninja的 `/backend-api/conversation` 接口，因此本项目并不支持高并发操作，请不要接入如 `沉浸式翻译` 等高并发项目。
 >
 > 3. 本项目支持使用apple平台的refresh_token作为请求key. 
 >
-> 4. 本项目并不能绕过 OpenAI 和 xyhelper 官方的限制，只提供便利，不提供绕过。
+> 4. 本项目并不能绕过 OpenAI 和 ninja 官方的限制，只提供便利，不提供绕过。
 >
 > 5. 提问的艺术：当出现项目不能正常运行时，请携带 `DEBUG` 级别的日志在 `Issue` 或者社区群内提问，否则将开启算命模式~
 
@@ -103,9 +95,9 @@
 
 - `process_threads`: 用于设置线程数，如果不需要设置，可以保持不变，如果需要设置，可以设置为需要设置的值，如果设置为 `1`，则会强制设置为单线程模式。
 
-- `upstream_base_url`: xyhelper 的免费接口地址，如：`https://demo.xyhelper.cn`，注意：不要以 `/` 结尾。
+- `upstream_base_url`: ninja 的部署地址，如：`https://pandoranext.com`，注意：不要以 `/` 结尾。可以填写为本项目可以访问到的 PandoraNext 的内网地址。
 
-- `upstream_api_prefix`: 默认为""
+- `upstream_api_prefix`: PandoraNext Proxy 模式下的 API 前缀
 
 - `backend_container_url`: 用于dalle模型生成图片的时候展示所用，需要设置为使用如 [ChatGPT-Next-Web](https://github.com/ChatGPTNextWebTeam/ChatGPT-Next-Web) 的用户可以访问到的本项目地址，如：`http://1.2.3.4:50011`，同原环境变量中的 `UPLOAD_BASE_URL`
 
@@ -151,8 +143,7 @@ PS. 注意，arkose_urls中的地址需要支持PandoraNext的Arkose Token获取
   
     - `enableOai`:用于设置是否使用官网通过refresh_token刷新access_token，仅在 `enableOai` 为 `true` 时生效。
       
-    - `xyhelper_refreshToAccess_Url`:用于设置使用xyhelper来进行使用refresh_token刷新access_token,enableOai为false的时候必填
-        - 默认为"https://demo.xyhelper.cn/applelogin"
+    - `ninja_refreshToAccess_Url`:用于设置使用ninja来进行使用refresh_token刷新access_token,enableOai为false的时候必填。
 
 - `redis`
 
@@ -191,7 +182,7 @@ PS. 注意，arkose_urls中的地址需要支持PandoraNext的Arkose Token获取
 
 请求方式：`POST`
 
-请求头：正常携带 `Authorization` 和 `Content-Type` 即可，`Authorization` 的值为 `Bearer <refresh_token>`，`Content-Type` 的值为 `application/json`
+请求头：正常携带 `Authorization` 和 `Content-Type` 即可，`Authorization` 的值为 `Bearer <ninja 的 fk>`，`Content-Type` 的值为 `application/json`
 
 请求体格式示例：
 
@@ -388,4 +379,4 @@ services:
 
 ## Star History
 
-[![Star History Chart](https://api.star-history.com/svg?repos=Yanyutin753/xyhelperV1Api_refresh&type=Date)](https://star-history.com/#Yanyutin753/xyhelperV1Api_refresh&Date)
+[![Star History Chart](https://api.star-history.com/svg?repos=Yanyutin753/ninjaToV1Api_refresh&type=Date)](https://star-history.com/#Yanyutin753/ninjaToV1Api_refresh&Date)

data/config.json

@@ -4,7 +4,7 @@
     "process_workers": 2,
     "process_threads": 2,
     "proxy": "",
-    "upstream_base_url": "https://demo.xyhelper.cn",
+    "upstream_base_url": "",
     "upstream_api_prefix": "",
     "backend_container_url": "",
     "backend_container_api_prefix": "",
@@ -27,7 +27,7 @@
     "refresh_ToAccess": {
         "stream_sleep_time": 0,
         "enableOai":"true",
-        "xyhelper_refreshToAccess_Url": "https://demo.xyhelper.cn/applelogin"
+        "ninja_refreshToAccess_Url": ""
     },
     "redis": {
         "host": "redis",

data/gpts.json

@@ -1 +1,5 @@
-{}
+{
+    "gpt-4-classic": {
+        "id":"g-YyyyMT9XH-chatgpt-classic"
+    }
+}

docker-compose.yml

@@ -2,7 +2,7 @@ version: '3'
 
 services:
   backend-to-api:
-    image: yangclivia/pandora-to-api:0.7.8
+    image: yangclivia/pandora-to-api:0.7.7
     restart: always
     ports:
       - "50011:33333"

main.py

@@ -62,10 +62,10 @@ BOT_MODE_ENABLED_CODE_BLOCK_OUTPUT = BOT_MODE.get('enabled_plugin_output', 'fals
 
 BOT_MODE_ENABLED_PLAIN_IMAGE_URL_OUTPUT = BOT_MODE.get('enabled_plain_image_url_output', 'false').lower() == 'true'
 
-# xyhelperToV1Api_refresh
+# ninjaToV1Api_refresh
 REFRESH_TOACCESS = CONFIG.get('refresh_ToAccess', {})
 REFRESH_TOACCESS_ENABLEOAI = REFRESH_TOACCESS.get('enableOai', 'true').lower() == 'true'
-REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL = REFRESH_TOACCESS.get('xyhelper_refreshToAccess_Url', '')
+REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL = REFRESH_TOACCESS.get('ninja_refreshToAccess_Url', '')
 STEAM_SLEEP_TIME = REFRESH_TOACCESS.get('steam_sleep_time', 0)
 
 NEED_DELETE_CONVERSATION_AFTER_RESPONSE = CONFIG.get('need_delete_conversation_after_response',
@@ -198,15 +198,12 @@ def oaiGetAccessToken(refresh_token):
     return None
 
 
-# xyhelper获得access_token
-def xyhelperGetAccessToken(getAccessTokenUrl, refresh_token):
+# ninja获得access_token
+def ninjaGetAccessToken(getAccessTokenUrl, refresh_token):
     try:
         logger.info("将通过这个网址请求access_token：" + getAccessTokenUrl)
-        
-        data = {
-            'refresh_token': refresh_token,
-        }
-        response = requests.post(getAccessTokenUrl, data=data)
+        headers = {"Authorization": "Bearer " + refresh_token}
+        response = requests.post(getAccessTokenUrl, headers=headers)
         if not response.ok:
             logger.error("Request 失败: " + response.text.strip())
             return None
@@ -230,7 +227,7 @@ def updateGptsKey():
         if REFRESH_TOACCESS_ENABLEOAI:
             access_token = oaiGetAccessToken(KEY_FOR_GPTS_INFO)
         else:
-            access_token = xyhelperGetAccessToken(REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL, KEY_FOR_GPTS_INFO)
+            access_token = ninjaGetAccessToken(REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL, KEY_FOR_GPTS_INFO)
         if access_token.startswith("eyJhb"):
             KEY_FOR_GPTS_INFO_ACCESS_TOKEN = access_token
             logging.info("KEY_FOR_GPTS_INFO_ACCESS_TOKEN被更新:" + KEY_FOR_GPTS_INFO_ACCESS_TOKEN)
@@ -276,17 +273,15 @@ def add_config_to_global_list(base_url, proxy_api_prefix, gpts_data):
             if gizmo_info:
                 redis_client.set(model_id, str(gizmo_info))
                 logger.info(f"Cached gizmo info for {model_name}, {model_id}")
-
-        if gizmo_info:
-            # 检查模型名称是否已经在列表中
-            if not any(d['name'] == model_name for d in gpts_configurations):
-                gpts_configurations.append({
-                    'name': model_name,
-                    'id': model_id,
-                    'config': gizmo_info
-                })
-            else:
-                logger.info(f"Model already exists in the list, skipping...")
+                # 检查模型名称是否已经在列表中
+                if not any(d['name'] == model_name for d in gpts_configurations):
+                    gpts_configurations.append({
+                        'name': model_name,
+                        'id': model_id,
+                        'config': gizmo_info
+                    })
+                else:
+                    logger.info(f"Model already exists in the list, skipping...")
 
 
 def generate_gpts_payload(model, messages):
@@ -324,9 +319,9 @@ scheduler.start()
 # PANDORA_UPLOAD_URL = 'files.pandoranext.com'
 
 
-VERSION = '0.7.8'
+VERSION = '0.7.7.1'
 # VERSION = 'test'
-UPDATE_INFO = '项目将脱离ninja，使用xyhelper，xyhelper_refreshToAccess_Url等配置需修改'
+UPDATE_INFO = '支持gpt-4-gizmo-XXX，动态配置GPTS'
 # UPDATE_INFO = '【仅供临时测试使用】 '
 
 # 解析响应中的信息
@@ -405,7 +400,7 @@ with app.app_context():
 
     logger.info(f"REFRESH_TOACCESS_ENABLEOAI: {REFRESH_TOACCESS_ENABLEOAI}")
     if not REFRESH_TOACCESS_ENABLEOAI:
-        logger.info(f"REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL: {REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL}")
+        logger.info(f"REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL: {REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL}")
 
     if BOT_MODE_ENABLED:
         logger.info(f"enabled_markdown_image_output: {BOT_MODE_ENABLED_MARKDOWN_IMAGE_OUTPUT}")
@@ -413,10 +408,10 @@ with app.app_context():
         logger.info(f"enabled_bing_reference_output: {BOT_MODE_ENABLED_BING_REFERENCE_OUTPUT}")
         logger.info(f"enabled_plugin_output: {BOT_MODE_ENABLED_CODE_BLOCK_OUTPUT}")
 
-    # xyhelperToV1Api_refresh
+    # ninjaToV1Api_refresh
 
     logger.info(f"REFRESH_TOACCESS_ENABLEOAI: {REFRESH_TOACCESS_ENABLEOAI}")
-    logger.info(f"REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL: {REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL}")
+    logger.info(f"REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL: {REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL}")
     logger.info(f"STEAM_SLEEP_TIME: {STEAM_SLEEP_TIME}")
 
     if not BASE_URL:
@@ -864,10 +859,14 @@ def send_text_prompt_and_get_response(messages, api_key, stream, model):
 
     # 查找模型配置
     model_config = find_model_config(model)
-    if model_config:
+    if model_config or 'gpt-4-gizmo-' in model:
         # 检查是否有 ori_name
-        ori_model_name = model_config.get('ori_name', model)
-        logger.info(f"原模型名: {ori_model_name}")
+        if model_config:
+            ori_model_name = model_config.get('ori_name', model)
+            logger.info(f"原模型名: {ori_model_name}")
+        else:
+            logger.info(f"请求模型名: {model}")
+            ori_model_name = model
         if ori_model_name == 'gpt-4-s':
             payload = {
                 # 构建 payload
@@ -918,6 +917,32 @@ def send_text_prompt_and_get_response(messages, api_key, stream, model):
                 "force_paragen": False,
                 "force_rate_limit": False
             }
+        elif 'gpt-4-gizmo-' in model:
+            payload = generate_gpts_payload(model, formatted_messages)
+            if not payload:
+                global gpts_configurations
+                # 假设 model是 'gpt-4-gizmo-123'
+                split_name = model.split('gpt-4-gizmo-')
+                model_id = split_name[1] if len(split_name) > 1 else None
+                gizmo_info = fetch_gizmo_info(BASE_URL, PROXY_API_PREFIX, model_id)
+                logging.info(gizmo_info)
+
+                # 如果成功获取到数据，则将其存入 Redis
+                if gizmo_info:
+                    redis_client.set(model_id, str(gizmo_info))
+                    logger.info(f"Cached gizmo info for {model}, {model_id}")
+                    # 检查模型名称是否已经在列表中
+                    if not any(d['name'] == model for d in gpts_configurations):
+                        gpts_configurations.append({
+                            'name': model,
+                            'id': model_id,
+                            'config': gizmo_info
+                        })
+                    else:
+                        logger.info(f"Model already exists in the list, skipping...")
+                    payload = generate_gpts_payload(model, formatted_messages)
+                else:
+                    raise Exception('KEY_FOR_GPTS_INFO is not accessible')
         else:
             payload = generate_gpts_payload(model, formatted_messages)
             if not payload:
@@ -2248,7 +2273,7 @@ def chat_completions():
     messages = data.get('messages')
     model = data.get('model')
     accessible_model_list = get_accessible_model_list()
-    if model not in accessible_model_list:
+    if model not in accessible_model_list and not 'gpt-4-gizmo-' in model:
         return jsonify({"error": "model is not accessible"}), 401
 
     stream = data.get('stream', False)
@@ -2266,7 +2291,7 @@ def chat_completions():
             if REFRESH_TOACCESS_ENABLEOAI:
                 api_key = oaiGetAccessToken(api_key)
             else:
-                api_key = xyhelperGetAccessToken(REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL, api_key)
+                api_key = ninjaGetAccessToken(REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL, api_key)
             if not api_key.startswith("eyJhb"):
                 return jsonify({"error": "refresh_token is wrong or refresh_token url is wrong!"}), 401
             add_to_dict(refresh_token, api_key)
@@ -2406,7 +2431,7 @@ def images_generations():
     # messages = data.get('messages')
     model = data.get('model')
     accessible_model_list = get_accessible_model_list()
-    if model not in accessible_model_list:
+    if model not in accessible_model_list and not 'gpt-4-gizmo-' in model:
         return jsonify({"error": "model is not accessible"}), 401
 
     prompt = data.get('prompt', '')
@@ -2431,7 +2456,7 @@ def images_generations():
                 refresh_token = api_key
                 api_key = oaiGetAccessToken(api_key)
             else:
-                api_key = xyhelperGetAccessToken(REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL, api_key)
+                api_key = ninjaGetAccessToken(REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL, api_key)
             if not api_key.startswith("eyJhb"):
                 return jsonify({"error": "refresh_token is wrong or refresh_token url is wrong!"}), 401
             add_to_dict(refresh_token, api_key)
@@ -2611,7 +2636,7 @@ def updateRefresh_dict():
         if REFRESH_TOACCESS_ENABLEOAI:
             access_token = oaiGetAccessToken(key)
         else:
-            access_token = xyhelperGetAccessToken(REFRESH_TOACCESS_XYHELPER_REFRESHTOACCESS_URL, key)
+            access_token = ninjaGetAccessToken(REFRESH_TOACCESS_NINJA_REFRESHTOACCESS_URL, key)
         if not access_token.startswith("eyJhb"):
             logger.debug("refresh_token is wrong or refresh_token url is wrong!")
             error_num += 1
@@ -2619,18 +2644,17 @@ def updateRefresh_dict():
         success_num += 1
     logging.info("更新成功: " + str(success_num) + ", 失败: " + str(error_num))
     logger.info(f"==========================================")
-    logging.info("开始更新KEY_FOR_GPTS_INFO_ACCESS_TOKEN和GPTS配置信息.......")
+    logging.info("开始更新KEY_FOR_GPTS_INFO_ACCESS_TOKEN和GPTS配置信息......")
     # 加载配置并添加到全局列表
     gpts_data = load_gpts_config("./data/gpts.json")
     add_config_to_global_list(BASE_URL, PROXY_API_PREFIX, gpts_data)
-
+    
     accessible_model_list = get_accessible_model_list()
-    logger.info(f"当前可用 GPTS 列表: {accessible_model_list}")
-
     # 检查列表中是否有重复的模型名称
     if len(accessible_model_list) != len(set(accessible_model_list)):
-        raise Exception("检测到重复的模型名称，请检查环境变量或配置文件......")
+        raise Exception("检测到重复的模型名称，请检查环境变量或配置文件。")
     logging.info("更新KEY_FOR_GPTS_INFO_ACCESS_TOKEN和GPTS配置信息成功......")
+    logger.info(f"当前可用 GPTS 列表: {accessible_model_list}")
     logger.info(f"==========================================")