repository/line_break_replacement.py

import os

source_dir = r"D:\code\repository\RAG资料库—开发"
target_dir = r"D:\code\repository\RAG资料库—上线"
file_extension = 'md'
paragraph_delimiter = '\n'  # 即将被替换的段落分割符
replacement = '。。'  # 替换为该字符

# 获取待处理的所有文件
all_files = []
for root, dirs, files in os.walk(source_dir):
    for file in files:
        if file.split('.')[-1] == file_extension:
            all_files.append(os.path.join(root, file))
# 开始处理
for file in all_files:
    # 文件保存路径
    save_path = file.replace(source_dir, target_dir)
    file_dir, filename = os.path.split(save_path)
    # 创建保存路径，确保存在
    if not os.path.exists(file_dir):
        os.makedirs(file_dir)

    # 读取文件
    with open(file, "r", encoding="utf-8") as f:
        content = f.readlines()
    # 处理孤立段落
    new_content = []
    for l in content:
        if l == paragraph_delimiter:
            continue
        # 末尾有分隔符
        if l[-1] == paragraph_delimiter:
            l = l[:-1]
        new_content.append(l)
    del content
    # 保存文件
    new_content = f"{replacement}".join(new_content)

    with open(save_path, "w", encoding="utf-8") as f:
        f.write(new_content)
    pass