"""
模拟超时跳过逻辑
文件列表：file_A 正常(2s)，file_B 超时(20s)，file_C 正常(2s)
期望结果：B 超时后直接跳到 C，总耗时约 TIMEOUT+2s 而不是 20+2+2s
"""

import time
import threading
import concurrent.futures
from concurrent.futures import ThreadPoolExecutor

EXTRACT_TIMEOUT = 5   # 测试用 5 秒超时


def fake_extract(file_name, cancel_event):
    """模拟抽取：file_B 故意睡 20 秒"""
    sleep_time = 20 if file_name == "file_B" else 2
    print(f"  [{file_name}] 开始抽取，预计耗时 {sleep_time}s")
    for _ in range(sleep_time * 10):
        time.sleep(0.1)
        if cancel_event.is_set():
            print(f"  [{file_name}] 收到取消信号，退出不写 JSON")
            return {"file": file_name, "status": "timeout"}
    print(f"  [{file_name}] 抽取完成，写 JSON")
    return {"file": file_name, "status": "ok"}


def run():
    files = ["file_A", "file_B", "file_C"]
    results = []
    t_start = time.time()

    with ThreadPoolExecutor(max_workers=2) as pool:
        future_meta = []
        for f in files:
            evt = threading.Event()
            fut = pool.submit(fake_extract, f, evt)
            future_meta.append((fut, f, evt))

        for fut, f, evt in future_meta:
            try:
                result = fut.result(timeout=EXTRACT_TIMEOUT)
                results.append(result)
            except concurrent.futures.TimeoutError:
                evt.set()
                print(f"  [{f}] 超时，跳过，记录 fault 文件")
                results.append({"file": f, "status": "timeout"})

    total = time.time() - t_start
    print(f"\n总耗时: {total:.1f}s")
    print("结果汇总:")
    for r in results:
        print(f"  {r}")


if __name__ == "__main__":
    run()