app.py

import json
import sseclient
import requests
from flask import Flask, request, Response, stream_with_context
import random

app = Flask(__name__)

def generate_random_ip():
    return f"{random.randint(1,255)}.{random.randint(0,255)}.{random.randint(0,255)}.{random.randint(0,255)}"

def format_openai_response(content, finish_reason=None):
    return {
        "id": "chatcmpl-123",
        "object": "chat.completion.chunk",
        "created": 1677652288,
        "model": "gpt-4o",
        "choices": [{
            "delta": {"content": content} if content else {"finish_reason": finish_reason},
            "index": 0,
            "finish_reason": finish_reason
        }]
    }

@app.route('/hf/v1/chat/completions', methods=['POST'])
def chat_completions():
    data = request.json
    messages = data.get('messages', [])
    stream = data.get('stream', False)
    
    if not messages:
        return {"error": "No messages provided"}, 400
    
    # 将消息列表转换为单个字符串，保留对话历史
    conversation = "\n".join([f"{msg['role']}: {msg['content']}" for msg in messages])
    
    # 添加指导语
    conversation += "\n请关注并回复user最近的消息并避免总结对话历史的回答"

    model = data.get('model', 'gpt-4o')

    original_api_url = 'https://chatpro.ai-pro.org/api/ask/openAI'
    headers = {
        'content-type': 'application/json',
        'X-Forwarded-For': generate_random_ip(),
        'origin': 'https://chatpro.ai-pro.org',
        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.0.0 Safari/537.36'
    }
    payload = {
        "text": conversation,
        "endpoint": "openAI",
        "model": model
    }

    def generate():
        last_content = ""
        response = requests.post(original_api_url, headers=headers, json=payload, stream=True)
        client = sseclient.SSEClient(response)
        
        for event in client.events():
            if event.data.startswith('{"text":'):
                data = json.loads(event.data)
                new_content = data['text'][len(last_content):]
                last_content = data['text']
                
                if new_content:
                    yield f"data: {json.dumps(format_openai_response(new_content))}\n\n"
            
            elif '"final":true' in event.data:
                yield f"data: {json.dumps(format_openai_response('', 'stop'))}\n\n"
                yield "data: [DONE]\n\n"
                break

    if stream:
        return Response(stream_with_context(generate()), content_type='text/event-stream')
    else:
        full_response = ""
        for chunk in generate():
            if chunk.startswith("data: ") and not chunk.strip() == "data: [DONE]":
                response_data = json.loads(chunk[6:])
                if 'choices' in response_data and response_data['choices']:
                    delta = response_data['choices'][0].get('delta', {})
                    if 'content' in delta:
                        full_response += delta['content']

        return {
            "id": "chatcmpl-123",
            "object": "chat.completion",
            "created": 1677652288,
            "model": model,
            "choices": [{
                "index": 0,
                "message": {
                    "role": "assistant",
                    "content": full_response
                },
                "finish_reason": "stop"
            }],
            "usage": {
                "prompt_tokens": 0,
                "completion_tokens": 0,
                "total_tokens": 0
            }
        }

if __name__ == '__main__':
    app.run(debug=True, port=5000)