326 lines
13 KiB
Python
326 lines
13 KiB
Python
import os
|
||
import re
|
||
import shutil
|
||
import uuid
|
||
|
||
from dotenv import load_dotenv
|
||
|
||
from transfer_md.upload_img import upload_image
|
||
from transfer_md.download_img import download_image
|
||
import sys
|
||
# 加载 .env 文件中的环境变量
|
||
load_dotenv()
|
||
|
||
def extract_image_paths(content):
|
||
"""
|
||
从 Markdown 内容中提取所有图片路径（支持 Markdown 和 HTML 格式）
|
||
"""
|
||
pattern_md = re.compile(r'!\[.*?\]\((.*?)\)')
|
||
pattern_html = re.compile(r'<img\s+[^>]*src\s*=\s*"(.*?)"')
|
||
return set(pattern_md.findall(content) + pattern_html.findall(content))
|
||
|
||
|
||
def process_local_image_copy(abs_img_path, dest_folder):
|
||
"""
|
||
复制本地图片到目标文件夹，并返回新文件名（使用 UUID 命名，保留扩展名）
|
||
"""
|
||
ext = os.path.splitext(abs_img_path)[1]
|
||
new_filename = f"{uuid.uuid4()}{ext}"
|
||
dest_path = os.path.join(dest_folder, new_filename)
|
||
shutil.copy2(abs_img_path, dest_path)
|
||
return new_filename
|
||
|
||
|
||
def process_md_file_local(md_file, pics_folder):
|
||
"""
|
||
处理一个 Markdown 文件：
|
||
- 提取 Markdown 和 HTML 格式的图片路径
|
||
- 复制本地图片到 output_path，并修改 md 文件中的图片引用路径
|
||
- 下载网络图片到 output_path，并修改 md 文件中的图片引用路径
|
||
- 图片复制时使用 UUID 作为文件名（保留扩展名）
|
||
- 更新后的图片路径为绝对路径
|
||
"""
|
||
with open(md_file, 'r', encoding='utf-8') as f:
|
||
content = f.read()
|
||
|
||
# 使用抽离的函数提取图片路径
|
||
img_paths = extract_image_paths(content)
|
||
|
||
# 获取当前 md 文件所在目录
|
||
md_dir = os.path.dirname(md_file)
|
||
abs_output_path = os.path.abspath(pics_folder)
|
||
|
||
for img_path in img_paths:
|
||
# 判断图片路径是本地路径还是网络 URL
|
||
if img_path.startswith(('http://', 'https://')):
|
||
# 处理网络图片
|
||
new_filename = download_image(img_path, pics_folder)
|
||
if new_filename:
|
||
# 使用绝对路径替换
|
||
new_ref = os.path.join(pics_folder, new_filename).replace('\\', '/')
|
||
content = content.replace(img_path, new_ref)
|
||
else:
|
||
# 处理本地图片
|
||
if os.path.isabs(img_path):
|
||
abs_img_path = img_path
|
||
else:
|
||
abs_img_path = os.path.normpath(os.path.join(md_dir, img_path))
|
||
abs_img_path = os.path.abspath(abs_img_path)
|
||
|
||
# 如果图片已经在 output 目录中，直接跳过复制
|
||
if abs_img_path.startswith(abs_output_path):
|
||
print(f"跳过已存在于 output 文件夹的图片: {abs_img_path}")
|
||
continue
|
||
|
||
if os.path.exists(abs_img_path):
|
||
if os.path.isfile(abs_img_path): # 确保是文件而不是文件夹
|
||
# 使用抽离的复制函数处理图片
|
||
new_filename = process_local_image_copy(abs_img_path, pics_folder)
|
||
dest_path = os.path.join(pics_folder, new_filename)
|
||
print(f"已复制: {abs_img_path} → {dest_path}")
|
||
# 使用绝对路径替换
|
||
new_ref = dest_path.replace('\\', '/')
|
||
content = content.replace(img_path, new_ref)
|
||
else:
|
||
print(f"警告: 跳过文件夹 {abs_img_path}")
|
||
else:
|
||
print(f"警告: 图片文件不存在 {abs_img_path}")
|
||
|
||
# 写回修改后的内容
|
||
with open(md_file, 'w', encoding='utf-8') as f:
|
||
f.write(content)
|
||
print(f"已更新: {md_file}")
|
||
|
||
|
||
def process_md_file_with_assets(md_file, output_path):
|
||
"""
|
||
处理单个 Markdown 文件，将其拷贝到 output_base_path/<md_name>/ 下，
|
||
并在该文件夹中建立 assets 文件夹保存相关图片。
|
||
同时更新 md 文件中图片的引用路径为相对路径 assets/<new_filename>
|
||
"""
|
||
# 创建对应的输出文件夹及 assets 子文件夹
|
||
md_filename = os.path.basename(md_file)
|
||
md_name, _ = os.path.splitext(md_filename)
|
||
target_folder = os.path.join(output_path, md_name)
|
||
assets_folder = os.path.join(target_folder, "assets")
|
||
os.makedirs(assets_folder, exist_ok=True)
|
||
|
||
# 读取 Markdown 文件内容
|
||
with open(md_file, 'r', encoding='utf-8') as f:
|
||
content = f.read()
|
||
|
||
# 使用抽离的函数提取图片路径
|
||
img_paths = extract_image_paths(content)
|
||
|
||
# 获取 md 文件所在目录（用于处理相对路径的本地图片）
|
||
md_dir = os.path.dirname(md_file)
|
||
|
||
# 遍历所有图片路径
|
||
for img_path in img_paths:
|
||
new_filename = None
|
||
if img_path.startswith(('http://', 'https://')):
|
||
# 处理网络图片：下载图片到 assets_folder
|
||
try:
|
||
# 处理网络图片：下载图片到 assets_folder
|
||
new_filename = download_image(img_path, assets_folder)
|
||
except Exception as e:
|
||
print(f"错误: 下载图片 {img_path} 时出错: {e}")
|
||
else:
|
||
# 处理本地图片
|
||
if os.path.isabs(img_path):
|
||
abs_img_path = img_path
|
||
else:
|
||
abs_img_path = os.path.normpath(os.path.join(md_dir, img_path))
|
||
if os.path.exists(abs_img_path) and os.path.isfile(abs_img_path):
|
||
try:
|
||
# 使用抽离的复制函数处理图片
|
||
new_filename = process_local_image_copy(abs_img_path, assets_folder)
|
||
print(f"已复制: {abs_img_path} → {os.path.join(assets_folder, new_filename)}")
|
||
except PermissionError as e:
|
||
print(f"错误: 无法复制文件 {abs_img_path}，权限被拒绝: {e}")
|
||
else:
|
||
print(f"警告: 图片文件不存在或不是文件 {abs_img_path}")
|
||
|
||
# 如果成功处理图片，则替换 md 文件中的引用路径
|
||
if new_filename:
|
||
new_ref = f"assets/{new_filename}"
|
||
content = content.replace(img_path, new_ref)
|
||
|
||
# 将更新后的 md 内容写入目标文件夹中的 md 文件
|
||
target_md_path = os.path.join(target_folder, md_filename)
|
||
with open(target_md_path, 'w', encoding='utf-8') as f:
|
||
f.write(content)
|
||
print(f"已更新: {target_md_path}")
|
||
|
||
|
||
def process_md_file_remote(md_file):
|
||
"""
|
||
处理一个 Markdown 文件：
|
||
- 提取 Markdown 和 HTML 格式的图片路径
|
||
- 对于本地图片，调用 upload_image 上传到 easyimage 图床，
|
||
并替换 md 文件中的图片引用路径为返回的公网地址
|
||
- 对于网络图片，保持不变
|
||
"""
|
||
with open(md_file, 'r', encoding='utf-8') as f:
|
||
content = f.read()
|
||
|
||
# 使用抽离的函数提取图片路径
|
||
img_paths = extract_image_paths(content)
|
||
|
||
# 获取当前 md 文件所在目录
|
||
md_dir = os.path.dirname(md_file)
|
||
|
||
for img_path in img_paths:
|
||
# 判断是否为本地图片（非网络 URL）
|
||
if not img_path.startswith(('http://', 'https://')):
|
||
if os.path.isabs(img_path):
|
||
abs_img_path = img_path
|
||
else:
|
||
abs_img_path = os.path.normpath(os.path.join(md_dir, img_path))
|
||
|
||
if os.path.exists(abs_img_path) and os.path.isfile(abs_img_path):
|
||
try:
|
||
public_url = upload_image(abs_img_path)
|
||
print(f"图片已上传: {abs_img_path} → {public_url}")
|
||
content = content.replace(img_path, public_url)
|
||
except Exception as e:
|
||
print(f"错误: 图片上传失败 {abs_img_path}: {e}")
|
||
else:
|
||
print(f"警告: 图片文件不存在 {abs_img_path}")
|
||
else:
|
||
print(f"跳过网络图片: {img_path}")
|
||
|
||
with open(md_file, 'w', encoding='utf-8') as f:
|
||
f.write(content)
|
||
print(f"已更新: {md_file}")
|
||
|
||
|
||
def format_mdfile(filepath, output_path, language="text"):
|
||
"""
|
||
对代码块进行格式化：若代码块没有指定语言，则添加指定语言。
|
||
同时将修改后的文件保存到 output_path/<category> 下，由于md格式存在不确定性，脚本处理结果不一定符合预期！。
|
||
"""
|
||
with open(filepath, 'r', encoding='utf-8') as f:
|
||
lines = f.readlines()
|
||
|
||
in_code_block = False
|
||
new_lines = []
|
||
# 匹配整行仅包含可选空白、可选列表标记和三个反引号（无其他内容）
|
||
# 这个模式匹配开始的代码块标记，可能包含列表前缀（如- ```）
|
||
start_pattern = re.compile(r'^(\s*(?:[-*+]\s+)?)(`{3})(\s*)$')
|
||
# 匹配已经有语言标记的代码块开始
|
||
lang_pattern = re.compile(r'^(\s*(?:[-*+]\s+)?)(`{3})[a-zA-Z0-9_+-]+')
|
||
# 匹配代码块结束
|
||
end_pattern = re.compile(r'^(\s*)(`{3})(\s*)$')
|
||
|
||
for line in lines:
|
||
# 若不在代码块内，尝试匹配代码块起始行
|
||
if not in_code_block:
|
||
# 检查是否为不带语言的代码块起始
|
||
start_match = start_pattern.match(line)
|
||
# 检查是否为带语言的代码块起始
|
||
lang_match = lang_pattern.match(line)
|
||
|
||
if start_match:
|
||
prefix, backticks, suffix = start_match.groups()
|
||
# 添加语言参数后重构该行（保留原始前缀和尾随空白）
|
||
line = f"{prefix}{backticks}{language}{suffix}\n"
|
||
in_code_block = True
|
||
elif lang_match:
|
||
# 如果已经有语言标记，直接进入代码块模式
|
||
in_code_block = True
|
||
else:
|
||
# 检测代码块结束
|
||
end_match = end_pattern.match(line)
|
||
if end_match:
|
||
in_code_block = False
|
||
|
||
# 同时对每一行内的 $$公式$$ 替换为 $公式$
|
||
line = re.sub(r'\$\$(.+?)\$\$', r'$\1$', line)
|
||
new_lines.append(line)
|
||
|
||
# 计算保存路径：
|
||
# 取原文件所在文件夹的名称作为 category
|
||
category = os.path.basename(os.path.dirname(filepath))
|
||
target_folder = os.path.join(output_path, category)
|
||
os.makedirs(target_folder, exist_ok=True)
|
||
target_md_path = os.path.join(target_folder, os.path.basename(filepath))
|
||
|
||
with open(target_md_path, 'w', encoding='utf-8') as f:
|
||
f.writelines(new_lines)
|
||
print(f"已格式化并保存: {target_md_path}")
|
||
|
||
|
||
def scan_files(base_folder, exclude_folders):
|
||
"""
|
||
扫描 base_folder 目录下所有 Markdown 文件，
|
||
并排除路径中包含 exclude_folders 中任一字符串的目录
|
||
"""
|
||
md_files = []
|
||
for root, dirs, files in os.walk(base_folder):
|
||
# 如果当前目录中包含需要排除的文件夹，则跳过该目录
|
||
if any(exclude in root for exclude in exclude_folders):
|
||
continue
|
||
for file in files:
|
||
if file.lower().endswith('.md'):
|
||
md_files.append(os.path.join(root, file))
|
||
return md_files
|
||
|
||
|
||
def process_md_files(input_path, output_path, type, exclude_folders=None):
|
||
"""
|
||
处理输入目录下所有 Markdown 文件，并将处理后的图片保存到 output_path/pics，
|
||
最后将经过 format_mdfile 格式化后的 Markdown 文件保存到 output_path/updated_files/<category> 下
|
||
type 参数决定了使用哪种图片处理方式：
|
||
type == 1: process_md_file_local
|
||
type == 2: process_md_file_with_assets
|
||
type == 3: process_md_file_remote
|
||
type == 4: 仅执行格式化（format_mdfile）
|
||
"""
|
||
os.makedirs(output_path, exist_ok=True)
|
||
# 确保 pics 文件夹存在
|
||
pics_folder = os.path.join(output_path, "pics")
|
||
assets_type_folder=os.path.join(output_path, "assets_type")
|
||
updated_folder=os.path.join(output_path, "updated_files")
|
||
os.makedirs(pics_folder, exist_ok=True)
|
||
os.makedirs(assets_type_folder, exist_ok=True)
|
||
os.makedirs(updated_folder, exist_ok=True)
|
||
if exclude_folders is None:
|
||
exclude_folders = []
|
||
md_files = scan_files(input_path, exclude_folders)
|
||
|
||
for md_file in md_files:
|
||
# 根据不同类型先进行图片处理（如果需要）
|
||
if type == 1:
|
||
process_md_file_local(md_file, pics_folder)
|
||
elif type == 2:
|
||
process_md_file_with_assets(md_file, assets_type_folder)
|
||
elif type == 3:
|
||
process_md_file_remote(md_file)
|
||
elif type == 4:
|
||
format_mdfile(md_file, updated_folder)
|
||
else:
|
||
print(f"未知的处理类型: {type}")
|
||
|
||
print("该文件夹下的 Markdown 文件已全部处理完成！")
|
||
|
||
|
||
if __name__ == "__main__":
|
||
# 从命令行获取 type 参数，如果未传入则默认使用 1
|
||
if len(sys.argv) > 1:
|
||
try:
|
||
type_value = int(sys.argv[1])
|
||
except ValueError:
|
||
print("第一个参数必须为整数，表示处理类型（1, 2 , 3, 4）")
|
||
sys.exit(1)
|
||
else:
|
||
type_value = 4
|
||
|
||
# 这里的输入输出路径根据实际情况修改
|
||
# input_path = os.getenv('BASE_FOLDER') #docker环境
|
||
input_path=r'D:\folder\study\md_files'
|
||
# output_path = os.getenv('OUTPUT_FOLDER')
|
||
output_path=r'D:\folder\study\md_files\output'
|
||
|
||
process_md_files(input_path, output_path, type_value)
|