"""
GLM 适配器测试脚本
测试 GLMAdapter 的流式和非流式调用，包括联网搜索功能
"""

import asyncio
import os
import sys
from pathlib import Path

# Add project root to sys.path
root_dir = Path(__file__).parent.parent
sys.path.insert(0, str(root_dir))

from dotenv import load_dotenv

from adapters.glm_adapter import GLMAdapter
from adapters.base import ChatCompletionRequest

load_dotenv()


async def test_stream():
    """测试流式调用（联网搜索）"""
    adapter = GLMAdapter()

    if not adapter.is_available():
        print("错误：未配置 ZHIPU_API_KEY 或 GLM_API_KEY")
        return

    request = ChatCompletionRequest(
        model="glm-4.6v",
        messages=[{"role": "user", "content": "今天北京天气怎样？"}],
        stream=True,
        temperature=0.7,
        max_tokens=1024,
        web_search=True,
    )

    print("Testing stream with web_search...")
    response = await adapter.chat(request)

    # 流式响应是 StreamingResponse，需要手动读取
    async for chunk in response.body_iterator:
        # body_iterator 已经返回字符串
        print(chunk, end="")


async def test_sync():
    """测试非流式调用（联网搜索）"""
    adapter = GLMAdapter()

    if not adapter.is_available():
        print("错误：未配置 ZHIPU_API_KEY 或 GLM_API_KEY")
        return

    request = ChatCompletionRequest(
        model="glm-4-flash",
        messages=[{"role": "user", "content": "今天几号？武汉天气怎样？"}],
        stream=False,
        temperature=0.7,
        max_tokens=1024,
        web_search=True,
    )

    print("Testing sync with web_search...")
    response = await adapter.chat(request)

    # 非流式响应返回 JSONResponse
    if hasattr(response, "body"):
        import json

        data = json.loads(response.body)
        content = data.get("choices", [{}])[0].get("message", {}).get("content", "")
        print(f"Response: {content}")
    else:
        print(f"Response: {response}")


if __name__ == "__main__":
    # 运行流式测试
    # asyncio.run(test_stream())
    # 运行非流式测试
    asyncio.run(test_sync())