python一键下载与替换hexo博客里的图片地址

1
# -*- coding: utf-8 -*-
2
# @Time: 2021/3/31
3
# @Update: 2024/02/09
4
# @Author: Eritque arcus
5
# @File: downloadImage.py
6
import mimetypes, os, pathlib, re, requests
7

8
if __name__ == '__main__':
9
    p = pathlib.Path("blog/source")
10
    image_p = p / "images"
11
    post_p = p / "_posts"
12
    if not post_p.exists():
13
        print(f"{post_p.absolute()} noy found")
14
        exit(1)
15
    else:
16
        os.chmod(post_p, 0o755)
17
    image_p.mkdir(exist_ok=True)
18
    os.chmod(p, 0o755)
19
    # 循环目录下每一个.md文件
20
    post_p.glob("*.md")
21
    for file in post_p.glob("*.md"):
22
        print(file)
23
        # 以utf8编码读取
24
        f = open(file, "r", encoding='utf8', errors='ignore')
25
        i = 0
26
        content = f.read()
27
        f.close()
28
        # 匹配markdown中的图片 即 ![tag](url)
29
        for text in re.findall(r"\!\[[^\s].*\]\([a-zA-z]+:\/\/[^\s]*\)", content):
30
            print(text)
31
            # 提取tag内容
32
            tag = re.findall(r"\!\[[^\s].*\](?=\([a-zA-z]+:\/\/[^\s]*\))", text)[0]
33
            # 提取url内容
34
            urldata = re.findall(r"(?<=\!\[[^\s].*\])\([a-zA-z]+:\/\/[^\s]*\)", text)[0]
35
            # 去掉括号
36
            u = urldata[1:len(urldata) - 1]
37
            # 提取当前文章的名字加上编号即为图片名字
38
            name = file.name + "-" + str(i)
39
            # 获取图片
40
            response = requests.get(u)
41
            # 取请求头content-type属性获取扩展名如.png/.jpg
42
            content_type = response.headers['content-type']
43
            extension = mimetypes.guess_extension(content_type)
44
            # 请求返回内容
45
            img = response.content
46
            # 写入图片文件
47
            with open(image_p / (name + extension), 'wb') as f:
48
                f.write(img)
49
            f.close()
50
            # 新引用地址
51
            new_u = "/images/" + name + extension
52
            # 替换原文
53
            content = content.replace(text, tag + "(" + new_u + ")")
54
            i += 1
55
        # 写入文章文件
56
        f = open(file, "w", encoding="utf8")
57
        f.write(content)
58
        f.close()