Python中如何只使用10M内存对10G文件进行排序？

请自行完成课后作业！抖个机灵，楼下回答

用Python在10M内存下排10G文件，得用外部排序。核心思路是把大文件切成小段，每段在内存里排序后存成临时文件，最后用堆合并。

这里的关键是heapq.merge，它能按顺序合并多个已排序的迭代器，且内存占用很小。下面是一个完整示例：

import heapq
import tempfile
import os

def external_sort(input_file, output_file, chunk_size=5*1024*1024):  # 5MB chunks
    temp_files = []
    
    # 1. 分割并排序
    with open(input_file, 'r') as f:
        chunk = []
        size = 0
        for line in f:
            chunk.append(line)
            size += len(line)
            if size >= chunk_size:
                chunk.sort()
                temp_file = tempfile.NamedTemporaryFile(mode='w+', delete=False)
                temp_file.writelines(chunk)
                temp_file.flush()
                temp_files.append(temp_file.name)
                chunk = []
                size = 0
        
        # 处理最后一块
        if chunk:
            chunk.sort()
            temp_file = tempfile.NamedTemporaryFile(mode='w+', delete=False)
            temp_file.writelines(chunk)
            temp_file.flush()
            temp_files.append(temp_file.name)
    
    # 2. 合并
    streams = [open(f, 'r') for f in temp_files]
    with open(output_file, 'w') as out:
        for line in heapq.merge(*streams):
            out.write(line)
    
    # 清理
    for s in streams:
        s.close()
    for f in temp_files:
        os.unlink(f)

# 使用示例
external_sort('bigfile.txt', 'sorted_bigfile.txt')

要点：