buffer_platform/app/api/ai_config.py

"""
AI模型配置接口 - 管理AI分析模型的配置
"""
import json
import logging
from pathlib import Path
from typing import Optional

from fastapi import APIRouter, Depends, HTTPException
from pydantic import BaseModel

logger = logging.getLogger(__name__)
router = APIRouter(prefix="/ai-config", tags=["AI模型配置"])


class AIModelConfig(BaseModel):
    """AI模型配置"""
    model_name: str
    api_key: str
    api_base: str = "https://api.openai.com/v1"
    model_id: str = "gpt-4"
    temperature: float = 0.7
    max_tokens: int = 2000
    enabled: bool = True


class AIConfigResponse(BaseModel):
    """AI配置响应"""
    success: bool
    data: Optional[dict] = None
    message: str = ""


class SaveAIConfigRequest(BaseModel):
    """保存AI配置请求"""
    models: list = []
    active_model: Optional[str] = None
    analysis_settings: Optional[dict] = None


CONFIG_DIR = Path(__file__).resolve().parent.parent.parent / "config"
AI_CONFIG_FILE = CONFIG_DIR / "ai_config.json"


def _ensure_config_dir():
    CONFIG_DIR.mkdir(parents=True, exist_ok=True)


def _load_ai_config() -> dict:
    """加载AI配置"""
    _ensure_config_dir()
    if not AI_CONFIG_FILE.exists():
        return {
            "models": [],
            "active_model": None,
            "analysis_settings": {
                "enable_technical_analysis": True,
                "enable_fundamental_analysis": False,
                "enable_sentiment_analysis": False,
                "risk_tolerance": "medium",
                "max_position_pct": 10
            }
        }
    with open(AI_CONFIG_FILE, "r", encoding="utf-8") as f:
        return json.load(f)


def _save_ai_config(config: dict):
    """保存AI配置"""
    _ensure_config_dir()
    with open(AI_CONFIG_FILE, "w", encoding="utf-8") as f:
        json.dump(config, f, ensure_ascii=False, indent=4)


@router.get("", response_model=AIConfigResponse)
def get_ai_config():
    """获取当前AI模型配置"""
    try:
        config = _load_ai_config()
        return {"success": True, "data": config}
    except Exception as e:
        logger.error(f"加载AI配置失败: {e}")
        return {"success": False, "message": str(e)}


@router.post("", response_model=AIConfigResponse)
def save_ai_config(config: SaveAIConfigRequest):
    """保存AI模型配置"""
    try:
        config_dict = {
            "models": config.models,
            "active_model": config.active_model,
            "analysis_settings": config.analysis_settings or {}
        }
        _save_ai_config(config_dict)
        return {"success": True, "message": "AI配置保存成功"}
    except Exception as e:
        logger.error(f"保存AI配置失败: {e}")
        return {"success": False, "message": str(e)}


@router.post("/test", response_model=AIConfigResponse)
def test_ai_connection(model_config: AIModelConfig):
    """测试AI模型连接"""
    try:
        import httpx
        
        headers = {
            "Authorization": f"Bearer {model_config.api_key}",
            "Content-Type": "application/json"
        }
        
        data = {
            "model": model_config.model_id,
            "messages": [{"role": "user", "content": "Hello"}],
            "max_tokens": 10
        }
        
        with httpx.Client(timeout=30) as client:
            response = client.post(
                f"{model_config.api_base}/chat/completions",
                headers=headers,
                json=data
            )
            
            if response.status_code == 200:
                return {"success": True, "message": "连接测试成功"}
            else:
                return {"success": False, "message": f"连接失败: {response.status_code} - {response.text}"}
                
    except Exception as e:
        logger.error(f"AI连接测试失败: {e}")
        return {"success": False, "message": f"连接测试失败: {str(e)}"}


@router.get("/providers")
def get_ai_providers():
    """获取支持的AI提供商列表"""
    providers = [
        {
            "id": "openai",
            "name": "OpenAI",
            "api_base": "https://api.openai.com/v1",
            "models": ["gpt-4o", "gpt-4-turbo", "gpt-3.5-turbo"]
        },
        {
            "id": "anthropic",
            "name": "Anthropic Claude",
            "api_base": "https://api.anthropic.com/v1",
            "models": ["claude-3-opus", "claude-3-sonnet", "claude-3-haiku"]
        },
        {
            "id": "google",
            "name": "Google Gemini",
            "api_base": "https://generativelanguage.googleapis.com/v1beta",
            "models": ["gemini-pro", "gemini-pro-vision"]
        },
        {
            "id": "aliyun",
            "name": "阿里云通义千问",
            "api_base": "https://dashscope.aliyuncs.com/compatible-mode/v1",
            "models": ["qwen-max", "qwen-plus", "qwen-turbo"]
        },
        {
            "id": "aliyun_coding",
            "name": "阿里云通义灵码",
            "api_base": "https://dashscope.aliyuncs.com/compatible-mode/v1",
            "models": ["qwen-coder-plus", "qwen-coder-turbo"]
        },
        {
            "id": "bailian",
            "name": "阿里百炼",
            "api_base": "https://coding.dashscope.aliyuncs.com/v1",
            "models": ["qwen3.6-plus", "qwen3.5-plus", "qwen3-max", "qwen3-coder-plus", "MiniMax-M2.5", "glm-4.7", "kimi-k2.5"]
        },
        {
            "id": "baidu",
            "name": "百度文心一言",
            "api_base": "https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop",
            "models": ["ernie-4.0", "ernie-3.5", "ernie-speed"]
        },
        {
            "id": "zhipu",
            "name": "智谱清言",
            "api_base": "https://open.bigmodel.cn/api/paas/v4",
            "models": ["glm-4", "glm-3-turbo"]
        }
    ]
    return {"success": True, "data": providers}