2024-03-21

fn find_sequence_boundaries(file: &File, num_threads: usize) -> io::Result<Vec<(u64, u64)>> {
    let mut reader = BufReader::new(file);
    let mut sequence_boundaries = Vec::new(); // 存储序列边界的向量
    let mut current_position = 0; // 当前文件指针位置
    let mut sequence_start = 0;   // 序列起始位置

    // 读取文件并记录每个序列的起始和结束位置
    for line in reader.lines() {
        let line = line?;
        if line.starts_with('>') {
            if !sequence_boundaries.is_empty() {
                // 设置上一个序列的结束位置
                sequence_boundaries.last_mut().unwrap().1 = current_position;
            }
            // 开始一个新序列，并记录起始位置
            sequence_start = current_position;
            sequence_boundaries.push((sequence_start, 0));
        }
        current_position += line.len() as u64 + 1; // 更新位置（包括换行符）
    }

    // 确保最后一个序列的结束位置也被记录
    if let Some(last) = sequence_boundaries.last_mut() {
        last.1 = current_position;
    }

    // 根据线程数将序列边界分配到不同的块中
    let mut distributed_boundaries = Vec::new();
    let sequences_per_thread = (sequence_boundaries.len() + num_threads - 1) / num_threads;

    for i in 0..num_threads {
        let start_idx = i * sequences_per_thread;
        let end_idx = ((i + 1) * sequences_per_thread).min(sequence_boundaries.len());
        
        // 如果计算结果超出范围，则可能不会为最后一个线程分配任务
        if start_idx < sequence_boundaries.len() {
            distributed_boundaries.push((
                sequence_boundaries[start_idx].0, 
                sequence_boundaries[end_idx - 1].1,
            ));
        }
    }

    Ok(distributed_boundaries)
}