支持最新的gpt-4-o模型

.github/workflows/oaifree-docker-image.yml

@@ -42,5 +42,6 @@ jobs:
         push: true
         tags: |
           yangclivia/pandora-to-api:${{ steps.tag_name.outputs.tag }}
+          yangclivia/pandora-to-api:0.7.9
         platforms: linux/amd64,linux/arm64
         build-args: TARGETPLATFORM=${{ matrix.platform }}

Dockerfile

@@ -10,15 +10,13 @@ COPY . /app
 # 设置环境变量
 ENV PYTHONUNBUFFERED=1
 
-RUN chmod +x /app/start.sh
-
-RUN apt update && apt install -y jq
+RUN chmod +x /app/main.py
 
 # # 设置 pip 源为清华大学镜像
 # RUN pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
 
 # 安装任何所需的依赖项
-RUN pip install --no-cache-dir flask flask_apscheduler gunicorn requests Pillow flask-cors tiktoken fake_useragent redis websocket-client pysocks requests[socks] websocket-client[optional]
+RUN pip install --no-cache-dir flask flask_apscheduler requests Pillow flask-cors tiktoken fake_useragent redis websocket-client pysocks requests[socks] websocket-client[optional]
 
 # 在容器启动时运行 Flask 应用
-CMD ["/app/start.sh"]
+CMD ["python3", "main.py"]

Readme.md

@@ -1,4 +1,7 @@
-## 项目简介
+# [RefreshToV1Api](https://github.com/Yanyutin753/RefreshToV1Api)
+![Docker Image Size (tag)](https://img.shields.io/docker/image-size/yangclivia/pandora-to-api/0.7.8)![Docker Pulls](https://img.shields.io/docker/pulls/yangclivia/pandora-to-api)[![GitHub Repo stars](https://img.shields.io/github/stars/Yanyutin753/RefreshToV1Api?style=social)](https://github.com/Yanyutin753/refresh-gpt-chat/stargazers)
+
+## [项目简介](https://github.com/Yanyutin753/RefreshToV1Api)
 
 > [!IMPORTANT]
 >
@@ -16,7 +19,9 @@
 
 4. 支持 gpt-4-mobile 、gpt-4-s 、基本所有的GPTS
 
-* **oaiFree 的 免费 backend-api 接口，无需打码**
+* **oaiFree 的 backend-api 接口，无需打码**
+
+* **oaiFree 的 backend-api 接口，只支持Chatgpt Plus账号**
 
 * 之后可能跟[Linux.do](https://linux.do/latest)论坛挂钩，请提前做好准备
 
@@ -85,20 +90,19 @@
 
 4. gpt-3.5-turbo
 
-## Docker-Compose 部署
+## 部署说明
+<details>
+
+### Docker-Compose 部署
 
 仓库内已包含相关文件和目录，拉到本地后修改 docker-compose.yml 文件里的环境变量后运行`docker-compose up -d`即可。
 
-## config.json 变量说明：
+### config.json 变量说明：
 
 - `log_level`: 用于设置日志等级，可选值为：`DEBUG`、`INFO`、`WARNING`、`ERROR`，默认为 `DEBUG`
 
 - `need_log_to_file`: 用于设置是否需要将日志输出到文件，可选值为：`true`、`false`，默认为 `true`，日志文件路径为：`./log/access.log`，默认每天会自动分割日志文件。
 
-- `process_workers`: 用于设置进程数，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，默认为 `2`。
-
-- `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，默认为 `2`。
-
 - `upstream_base_url`: oaiFree 的接口地址，如：`https://chat.oaifree.com`，注意：不要以 `/` 结尾。
 
 - `upstream_api_prefix`: 默认为["dad04481-fa3f-494e-b90c-b822128073e5"]，之后可多填
@@ -147,7 +151,7 @@ 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。
       
-    - `oaiFree_refreshToAccess_Url`:用于设置使用oaiFree来进行使用refresh_token刷新access_token,enableOai为false的时候必填
+    - `oaifree_refreshToAccess_Url`:用于设置使用oaiFree来进行使用refresh_token刷新access_token,enableOai为false的时候必填
         - 默认为"https://token.oaifree.com/api/auth/refresh"
 
 - `redis`
@@ -160,7 +164,7 @@ PS. 注意，arkose_urls中的地址需要支持PandoraNext的Arkose Token获取
 
     - `db`: Redis的数据库，默认：0，如有特殊需求，你可以将此值设置为其他数据库
 
-## GPTS配置说明
+### GPTS配置说明
 
 如果需要使用 GPTS，需要修改 `gpts.json` 文件，其中每个对象的key即为调用对应 GPTS 的时候使用的模型名称，而 `id` 则为对应的模型id，该 `id` 对应每个 GPTS 的链接的后缀。配置多个GPTS的时候用逗号隔开。
 
@@ -181,7 +185,7 @@ PS. 注意，arkose_urls中的地址需要支持PandoraNext的Arkose Token获取
 
 注意：使用该配置的时候需要保证正确填写 `docker-compose.yml` 的环境变量 `KEY_FOR_GPTS_INFO`，同时该变量设置的 `key` 允许访问所有配置的 GPTS。
 
-## 绘图接口使用说明
+### 绘图接口使用说明
 
 接口URI：`/v1/images/generations`
 
@@ -218,7 +222,7 @@ PS. 注意，arkose_urls中的地址需要支持PandoraNext的Arkose Token获取
 }
 ```
 
-## 文件识别接口使用说明
+### 文件识别接口使用说明
 
 调用方式同官方 `gpt-4-vision-preview` API 
 
@@ -317,8 +321,18 @@ PS. 注意，arkose_urls中的地址需要支持PandoraNext的Arkose Token获取
     }
 }
 ```
+### 获取ChatGPT-Account-ID接口
 
-## 示例
+接口URI：`/getAccountID`
+
+请求方式：`POST`
+
+```
+请求头加上
+Authorization：Bearer refresh_token 或 access_token
+```
+
+### 示例
 
 以ChatGPT-Next-Web项目的docker-compose部署为例，这里提供一个简单的部署配置文件示例：
 
@@ -335,6 +349,8 @@ services:
       - CUSTOM_MODELS=+gpt-4-s,+gpt-4-mobile,+<gpts.json 中的模型名>
 
 ```
+</details>
+
 
 ## 功能演示
 <details>
@@ -377,11 +393,12 @@ services:
 > * 本项目只提供转发接口🥰
 > * 开源项目不易，请点个星星吧！！！
 
-### 新增群聊，点了⭐️可以进群讨论部署，我把你们拉进群，无广，广子踢掉
-<img src="https://github.com/Yanyutin753/PandoraNext-TokensTool/assets/132346501/6544e8ed-6673-48f9-95a6-c13255acbab1" width="300" height="300">
+## Sponsor 
+
+### 如果你觉得我的开源项目对你有帮助，可以赞助我一杯咖啡嘛，十分感谢！！！
+<img src="https://github.com/Yanyutin753/RefreshToV1Api/assets/132346501/e5ab5e80-1cf2-4822-ae36-f9d0b11ed1b1" width="300" height="300">
 
-### 请给我一个免费的⭐吧！！！
 
 ## Star History
 
-[![Star History Chart](https://api.star-history.com/svg?repos=Yanyutin753/oaiFreeV1Api_refresh&type=Date)](https://star-history.com/#Yanyutin753/oaiFreeV1Api_refresh&Date)
+[![Star History Chart](https://api.star-history.com/svg?repos=Yanyutin753/RefreshToV1Api&type=Date)](https://star-history.com/#Yanyutin753/oaiFreeV1Api_refresh&Date)

data/config.json

@@ -1,8 +1,6 @@
 {
     "log_level": "INFO",
     "need_log_to_file": "true",
-    "process_workers": 2,
-    "process_threads": 2,
     "proxy": "",
     "upstream_base_url": "https://chat.oaifree.com",
     "upstream_api_prefix": ["dad04481-fa3f-494e-b90c-b822128073e5"],
@@ -12,6 +10,7 @@
     "gpt_4_s_new_name": "gpt-4-s",
     "gpt_4_mobile_new_name": "gpt-4-mobile,dall-e-3",
     "gpt_3_5_new_name": "gpt-3.5-turbo",
+    "gpt_4_o_new_name": "gpt-4-o",
     "need_delete_conversation_after_response": "true",
     "use_oaiusercontent_url": "false",
     "custom_arkose_url": "false",

main.py

@@ -51,6 +51,7 @@ API_PREFIX = CONFIG.get('backend_container_api_prefix', '')
 GPT_4_S_New_Names = CONFIG.get('gpt_4_s_new_name', 'gpt-4-s').split(',')
 GPT_4_MOBILE_NEW_NAMES = CONFIG.get('gpt_4_mobile_new_name', 'gpt-4-mobile').split(',')
 GPT_3_5_NEW_NAMES = CONFIG.get('gpt_3_5_new_name', 'gpt-3.5-turbo').split(',')
+GPT_4_O_NEW_NAMES = CONFIG.get('gpt_4_o_new_name', 'gpt-4-o').split(',')
 
 BOT_MODE = CONFIG.get('bot_mode', {})
 BOT_MODE_ENABLED = BOT_MODE.get('enabled', 'false').lower() == 'true'
@@ -221,7 +222,6 @@ def oaiFreeGetAccessToken(getAccessTokenUrl, refresh_token):
             'refresh_token': refresh_token,
         }
         response = requests.post(getAccessTokenUrl, data=data)
-        logging.info(response.text)
         if not response.ok:
             logger.error("Request 失败: " + response.text.strip())
             return None
@@ -336,9 +336,9 @@ scheduler.start()
 # PANDORA_UPLOAD_URL = 'files.pandoranext.com'
 
 
-VERSION = '0.7.9.0'
+VERSION = '0.7.9.2'
 # VERSION = 'test'
-UPDATE_INFO = '接入oaifree'
+UPDATE_INFO = '支持最新的gpt-4-o模型'
 # UPDATE_INFO = '【仅供临时测试使用】 '
 
 with app.app_context():
@@ -449,7 +449,11 @@ with app.app_context():
             "name": name.strip(),
             "ori_name": "gpt-3.5-turbo"
         })
-
+    for name in GPT_4_O_NEW_NAMES:
+        gpts_configurations.append({
+            "name": name.strip(),
+            "ori_name": "gpt-4-o"
+        })
     logger.info(f"GPTS 配置信息")
 
     # 加载配置并添加到全局列表
@@ -614,7 +618,7 @@ def get_file_metadata(file_content, mime_type, api_key, proxy_api_prefix):
     sha256_hash = hashlib.sha256(file_content).hexdigest()
     logger.debug(f"sha256_hash: {sha256_hash}")
     # 首先尝试从Redis中获取数据
-    cached_data = file_redis_client.get(sha256_hash)
+    cached_data = redis_client.get(sha256_hash)
     if cached_data is not None:
         # 如果在Redis中找到了数据，解码后直接返回
         logger.info(f"从Redis中获取到文件缓存数据")
@@ -653,7 +657,7 @@ def get_file_metadata(file_content, mime_type, api_key, proxy_api_prefix):
         new_file_data['height'] = height
 
     # 将新的文件数据存入Redis
-    file_redis_client.set(sha256_hash, json.dumps(new_file_data))
+    redis_client.set(sha256_hash, json.dumps(new_file_data))
 
     return new_file_data
 
@@ -702,7 +706,7 @@ my_files_types = [
 
 
 # 定义发送请求的函数
-def send_text_prompt_and_get_response(messages, api_key, stream, model, proxy_api_prefix):
+def send_text_prompt_and_get_response(messages, api_key, account_id, stream, model, proxy_api_prefix):
     url = f"{BASE_URL}{proxy_api_prefix}/backend-api/conversation"
     headers = {
         "Authorization": f"Bearer {api_key}"
@@ -720,7 +724,7 @@ def send_text_prompt_and_get_response(messages, api_key, stream, model, proxy_ap
         message_id = str(uuid.uuid4())
         content = message.get("content")
 
-        if isinstance(content, list) and ori_model_name != 'gpt-3.5-turbo':
+        if isinstance(content, list) and ori_model_name not in ['gpt-3.5-turbo', 'gpt-4-o']:
             logger.debug(f"gpt-vision 调用")
             new_parts = []
             attachments = []
@@ -880,6 +884,28 @@ def send_text_prompt_and_get_response(messages, api_key, stream, model, proxy_ap
                 "force_paragen": False,
                 "force_rate_limit": False
             }
+        elif ori_model_name == 'gpt-4-o':
+            payload = {
+                # 构建 payload
+                "action": "next",
+                "messages": formatted_messages,
+                "parent_message_id": str(uuid.uuid4()),
+                "model": "auto",
+                "timezone_offset_min": -480,
+                "suggestions": [
+                    "What are 5 creative things I could do with my kids' art? I don't want to throw them away, but it's also so much clutter.",
+                    "I want to cheer up my friend who's having a rough day. Can you suggest a couple short and sweet text messages to go with a kitten gif?",
+                    "Come up with 5 concepts for a retro-style arcade game.",
+                    "I have a photoshoot tomorrow. Can you recommend me some colors and outfit options that will look good on camera?"
+                ],
+                "history_and_training_disabled": False,
+                "arkose_token": None,
+                "conversation_mode": {
+                    "kind": "primary_assistant"
+                },
+                "force_paragen": False,
+                "force_rate_limit": False
+            }
         elif 'gpt-4-gizmo-' in model:
             payload = generate_gpts_payload(model, formatted_messages)
             if not payload:
@@ -914,12 +940,17 @@ def send_text_prompt_and_get_response(messages, api_key, stream, model, proxy_ap
         if NEED_DELETE_CONVERSATION_AFTER_RESPONSE:
             logger.debug(f"是否保留会话: {NEED_DELETE_CONVERSATION_AFTER_RESPONSE == False}")
             payload['history_and_training_disabled'] = True
-        if ori_model_name != 'gpt-3.5-turbo':
+        if ori_model_name not in ['gpt-3.5-turbo', 'gpt-4-o']:
             if CUSTOM_ARKOSE:
                 token = get_token()
                 payload["arkose_token"] = token
                 # 在headers中添加新字段
                 headers["Openai-Sentinel-Arkose-Token"] = token
+
+            # 用于调用ChatGPT Team次数
+            if account_id:
+                headers["ChatGPT-Account-ID"] = account_id
+
         logger.debug(f"headers: {headers}")
         logger.debug(f"payload: {payload}")
         response = requests.post(url, headers=headers, json=payload, stream=True)
@@ -1139,7 +1170,8 @@ def replace_sandbox(text, conversation_id, message_id, api_key, proxy_api_prefix
     return replaced_text
 
 
-def data_fetcher(upstream_response, data_queue, stop_event, last_data_time, api_key, chat_message_id, model, proxy_api_prefix):
+def data_fetcher(upstream_response, data_queue, stop_event, last_data_time, api_key, chat_message_id, model,
+                 proxy_api_prefix):
     all_new_text = ""
 
     first_output = True
@@ -1372,7 +1404,7 @@ def data_fetcher(upstream_response, data_queue, stop_event, last_data_time, api_
                                         if is_complete_sandbox_format(file_output_buffer):
                                             # 替换完整的引用格式
                                             replaced_text = replace_sandbox(file_output_buffer, conversation_id,
-                                                                            message_id, api_key,proxy_api_prefix)
+                                                                            message_id, api_key, proxy_api_prefix)
                                             # print(replaced_text)  # 输出替换后的文本
                                             new_text = replaced_text
                                             file_output_accumulating = False
@@ -1756,13 +1788,18 @@ def chat_completions():
     auth_header = request.headers.get('Authorization')
     if not auth_header or not auth_header.startswith('Bearer '):
         return jsonify({"error": "Authorization header is missing or invalid"}), 401
-    api_key = auth_header.split(' ')[1]
+    try:
+        api_key = auth_header.split(' ')[1].split(',')[0].strip()
+        account_id = auth_header.split(' ')[1].split(',')[1].strip()
+        logging.info(f"{api_key}:{account_id}")
+    except IndexError:
+        account_id = None
     if not api_key.startswith("eyJhb"):
+        refresh_token = api_key
         if api_key in refresh_dict:
             logger.info(f"从缓存读取到api_key.........。")
             api_key = refresh_dict.get(api_key)
         else:
-            refresh_token = api_key
             if REFRESH_TOACCESS_ENABLEOAI:
                 api_key = oaiGetAccessToken(api_key)
             else:
@@ -1772,7 +1809,8 @@ def chat_completions():
             add_to_dict(refresh_token, api_key)
     logger.info(f"api_key: {api_key}")
 
-    upstream_response = send_text_prompt_and_get_response(messages, api_key, stream, model, proxy_api_prefix)
+    upstream_response = send_text_prompt_and_get_response(messages, api_key, account_id, stream, model,
+                                                          proxy_api_prefix)
 
     # 在非流式响应的情况下，我们需要一个变量来累积所有的 new_text
     all_new_text = ""
@@ -1791,7 +1829,8 @@ def chat_completions():
 
         # 启动数据处理线程
         fetcher_thread = threading.Thread(target=data_fetcher, args=(
-            upstream_response, data_queue, stop_event, last_data_time, api_key, chat_message_id, model,proxy_api_prefix))
+            upstream_response, data_queue, stop_event, last_data_time, api_key, chat_message_id, model,
+            proxy_api_prefix))
         fetcher_thread.start()
 
         # 启动保活线程
@@ -1848,9 +1887,9 @@ def chat_completions():
             fetcher_thread.join()
             keep_alive_thread.join()
 
-            if conversation_id:
-                # print(f"准备删除的会话id： {conversation_id}")
-                delete_conversation(conversation_id, api_key,proxy_api_prefix)
+            # if conversation_id:
+            #     # print(f"准备删除的会话id： {conversation_id}")
+            #     delete_conversation(conversation_id, api_key,proxy_api_prefix)
 
     if not stream:
         # 执行流式响应的生成函数来累积 all_new_text
@@ -1924,14 +1963,19 @@ def images_generations():
     auth_header = request.headers.get('Authorization')
     if not auth_header or not auth_header.startswith('Bearer '):
         return jsonify({"error": "Authorization header is missing or invalid"}), 401
-    api_key = auth_header.split(' ')[1]
+    try:
+        api_key = auth_header.split(' ')[1].split(',')[0].strip()
+        account_id = auth_header.split(' ')[1].split(',')[1].strip()
+        logging.info(f"{api_key}:{account_id}")
+    except IndexError:
+        account_id = None
     if not api_key.startswith("eyJhb"):
+        refresh_token = api_key
         if api_key in refresh_dict:
             logger.info(f"从缓存读取到api_key.........")
             api_key = refresh_dict.get(api_key)
         else:
             if REFRESH_TOACCESS_ENABLEOAI:
-                refresh_token = api_key
                 api_key = oaiGetAccessToken(api_key)
             else:
                 api_key = oaiFreeGetAccessToken(REFRESH_TOACCESS_OAIFREE_REFRESHTOACCESS_URL, api_key)
@@ -1951,7 +1995,7 @@ def images_generations():
         }
     ]
 
-    upstream_response = send_text_prompt_and_get_response(messages, api_key, False, model,proxy_api_prefix)
+    upstream_response = send_text_prompt_and_get_response(messages, api_key, account_id, False, model, proxy_api_prefix)
 
     # 在非流式响应的情况下，我们需要一个变量来累积所有的 new_text
     all_new_text = ""
@@ -2367,6 +2411,56 @@ def get_file(filename):
     return send_from_directory('files', filename)
 
 
+@app.route(f'/{API_PREFIX}/getAccountID' if API_PREFIX else '/getAccountID', methods=['POST'])
+@cross_origin()  # 使用装饰器来允许跨域请求
+def getAccountID():
+    logger.info(f"New Img Request")
+    proxy_api_prefix = getPROXY_API_PREFIX(lock)
+    if proxy_api_prefix == None:
+        return jsonify({"error": "PROXY_API_PREFIX is not accessible"}), 401
+    auth_header = request.headers.get('Authorization')
+    if not auth_header or not auth_header.startswith('Bearer '):
+        return jsonify({"error": "Authorization header is missing or invalid"}), 401
+    try:
+        api_key = auth_header.split(' ')[1].split(',')[0].strip()
+        account_id = auth_header.split(' ')[1].split(',')[1].strip()
+    except IndexError:
+        account_id = None
+    if not api_key.startswith("eyJhb"):
+        refresh_token = api_key
+        if api_key in refresh_dict:
+            logger.info(f"从缓存读取到api_key.........")
+            api_key = refresh_dict.get(api_key)
+        else:
+            if REFRESH_TOACCESS_ENABLEOAI:
+                api_key = oaiGetAccessToken(api_key)
+            else:
+                api_key = oaiFreeGetAccessToken(REFRESH_TOACCESS_OAIFREE_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)
+    logger.info(f"api_key: {api_key}")
+
+    url = f"{BASE_URL}{proxy_api_prefix}/backend-api/accounts/check/v4-2023-04-27"
+    headers = {
+        "Authorization": "Bearer " + api_key
+    }
+    res = requests.get(url, headers=headers)
+    if res.status_code == 200:
+        data = res.json()
+        result = {"plus": set(), "team": set()}
+        for account_id, account_data in data["accounts"].items():
+            plan_type = account_data["account"]["plan_type"]
+            if plan_type == "team":
+                result[plan_type].add(account_id)
+            elif plan_type == "plus":
+                result[plan_type].add(account_id)
+        result = {plan_type: list(ids) for plan_type, ids in result.items()}
+        return jsonify(result)
+    else:
+        return jsonify({"error": "Request failed."}), 400
+
+
 # 内置自动刷新access_token
 def updateRefresh_dict():
     success_num = 0

start.sh

@@ -1,34 +0,0 @@
-#!/bin/bash
-
-# 记录当前日期和时间
-NOW=$(date +"%Y-%m-%d-%H-%M")
-
-# 尝试从环境变量获取参数，如果不存在，则从 config.json 文件中读取
-# 如果这些值仍然不存在，将它们设置为默认值
-
-if [ -z "$PROCESS_WORKERS" ]; then
-    PROCESS_WORKERS=$(jq -r '.process_workers // empty' /app/data/config.json)
-    export PROCESS_WORKERS
-
-    if [ -z "$PROCESS_WORKERS" ]; then
-        PROCESS_WORKERS=2
-    fi
-fi
-
-if [ -z "$PROCESS_THREADS" ]; then
-    PROCESS_THREADS=$(jq -r '.process_threads // empty' /app/data/config.json)
-    export PROCESS_THREADS
-
-    if [ -z "$PROCESS_THREADS" ]; then
-        PROCESS_THREADS=2
-    fi
-fi
-
-export PROCESS_WORKERS
-export PROCESS_THREADS
-
-echo "PROCESS_WORKERS: ${PROCESS_WORKERS}"
-echo "PROCESS_THREADS: ${PROCESS_THREADS}"
-
-# 启动 Gunicorn 并使用 tee 命令同时输出日志到文件和控制台
-exec gunicorn -w ${PROCESS_WORKERS} --threads ${PROCESS_THREADS} --bind 0.0.0.0:33333 main:app --access-logfile - --error-logfile -