from collections import deque
from typing import Optional
import numpy as np


class BaseDetector:
    """
    检测器基类
    提供通用的帧回溯缓存功能，子类可按需使用
    """

    def __init__(self):
        # 帧回溯缓存（子类需要时调用 init_frame_buffer 初始化）
        self._frame_buffer: Optional[deque] = None

    def init_frame_buffer(self, buffer_seconds: float, fps: float):
        """
        初始化帧回溯缓存队列

        Args:
            buffer_seconds: 需要缓存的时间长度（秒）
            fps: 视频帧率
        """
        maxlen = int(buffer_seconds * fps)
        self._frame_buffer = deque(maxlen=maxlen)

    def append_frame(self, frame: np.ndarray, timestamp: float):
        """
        将当前帧入队缓存

        Args:
            frame: 当前帧图像
            timestamp: 当前帧的时间戳
        """
        if self._frame_buffer is not None:
            self._frame_buffer.append({
                'timestamp': timestamp,
                'frame': frame.copy(),
            })

    def find_target_frame(self, target_time_sec: float) -> Optional[np.ndarray]:
        """
        在帧缓存中找到最接近目标时间的帧

        Args:
            target_time_sec: 目标时间戳

        Returns:
            最接近目标时间的帧图像，缓存为空则返回 None
        """
        if self._frame_buffer is None or len(self._frame_buffer) == 0:
            return None

        target_frame = None
        min_time_diff = float('inf')
        for buffered in self._frame_buffer:
            time_diff = abs(buffered['timestamp'] - target_time_sec)
            if time_diff < min_time_diff:
                min_time_diff = time_diff
                target_frame = buffered['frame']

        return target_frame