Update app.py

app.py

@@ -1,103 +1,32 @@
-import requests
-import json
-import uuid
-import time
-from flask import Flask, request, jsonify, Response
-
-# 1. 初始化Flask应用
-app = Flask(__name__)
-
-# 2. gpt-oss.com API的固定配置
-GPT_OSS_API_URL = "https://api.gpt-oss.com/chatkit"
-GPT_OSS_HEADERS = {
-    'authority': 'api.gpt-oss.com',
-    'accept': 'text/event-stream',
-    'content-type': 'application/json',
-    'origin': 'https://gpt-oss.com',
-    'referer': 'https://gpt-oss.com/',
-    'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
-    'x-selected-model': 'gpt-oss-120b',
-}
-
-# 3. 新增：根路由，用于健康检查和显示欢迎信息
-@app.route('/', methods=['GET'])
-def root():
-    return jsonify({
-        "message": "欢迎使用 GPT-OSS to OpenAI 格式代理API",
-        "status": "ok",
-        "api_endpoint": "/v1/chat/completions"
-    })
-
-# 4. 核心：创建OpenAI兼容的API端点
-@app.route('/v1/chat/completions', methods=['POST'])
-def chat_completions_proxy():
-    try:
-        openai_request_data = request.json
-        messages = openai_request_data.get("messages", [])
-        user_prompt = next((m['content'] for m in reversed(messages) if m.get('role') == 'user'), None)
-        
-        if not user_prompt:
-            return jsonify({"error": "在请求中未找到用户消息。"}), 400
-            
-        stream_requested = openai_request_data.get("stream", False)
-    except Exception as e:
-        return jsonify({"error": f"请求格式无效: {e}"}), 400
-
-    random_user_id = str(uuid.uuid4())
-    gpt_oss_cookies = {'user_id': random_user_id}
-    gpt_oss_payload = {
-        "op": "threads.create",
-        "params": {"input": {"text": user_prompt, "content": [{"type": "input_text", "text": user_prompt}]}}
-    }
-    
-    def generate_stream():
-        try:
-            with requests.post(
-                GPT_OSS_API_URL,
-                headers=GPT_OSS_HEADERS,
-                cookies=gpt_oss_cookies,
-                json=gpt_oss_payload,
-                stream=True,
-                timeout=120
-            ) as response:
-                response.raise_for_status()
-
-                for line in response.iter_lines():
-                    if line:
-                        line_str = line.decode('utf-8')
-                        if line_str.startswith('data: '):
-                            json_data_str = line_str[6:]
-                            try:
-                                gpt_oss_data = json.loads(json_data_str)
-                                
-                                event_type = gpt_oss_data.get('type')
-                                if (event_type == 'thread.item_updated' and 
-                                    gpt_oss_data.get('update', {}).get('type') == 'assistant_message.content_part.text_delta'):
-                                    
-                                    delta_content = gpt_oss_data['update'].get('delta', '')
-                                    openai_chunk = {
-                                        "id": f"chatcmpl-{str(uuid.uuid4())}",
-                                        "object": "chat.completion.chunk",
-                                        "created": int(time.time()),
-                                        "model": "gpt-oss-120b",
-                                        "choices": [{"index": 0, "delta": {"content": delta_content}, "finish_reason": None}]
-                                    }
-                                    yield f"data: {json.dumps(openai_chunk)}\n\n"
-
-                            except json.JSONDecodeError:
-                                continue
-
-                yield "data: [DONE]\n\n"
-        except requests.exceptions.RequestException as e:
-            error_chunk = {"error": f"与后端服务通信失败: {e}"}
-            yield f"data: {json.dumps(error_chunk)}\n\n"
-
-    if stream_requested:
-        return Response(generate_stream(), mimetype='text/event-stream')
-    else:
-        return jsonify({"error": "非流式响应目前不受支持，请在请求中设置 'stream': true"}), 501
-
-# 5. 启动应用
-if __name__ == '__main__':
-    # 使用 gunicorn 部署时不会执行这里，但在本地测试时会用到
-    app.run(host='0.0.0.0', port=7860)
+---
+title: My OpenAI Proxy (Stream/Non-Stream)
+emoji: 🚀
+colorFrom: blue
+colorTo: green
+sdk: gradio
+sdk_version: 3.45.0
+python_version: 3.9
+app_file: app.py
+app_port: 7860
+---
+
+# GPT-OSS to OpenAI API Proxy
+
+这是一个代理服务，它将标准的OpenAI API请求转换为 `gpt-oss.com` 所需的格式。
+**此版本支持流式和非流式响应。**
+
+## 如何使用
+
+将您的API客户端的 `base_url` 指向本Space的URL，然后使用`/v1/chat/completions`端点。
+
+### 流式请求 (会返回思考过程)
+
+```bash
+curl https://YOUR_SPACE_URL/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -H "Authorization: Bearer sk-any-key-works" \
+  -d '{
+    "model": "gpt-oss-120b",
+    "messages": [{"role": "user", "content": "你好吗？"}],
+    "stream": true
+  }'