实现将csdn博客批量迁移

主要参考这篇文章，实现将csdn博客批量迁移（即文章批量导出为md文件）

https://blog.csdn.net/pang787559613/article/details/105444286

文中有些细节没说，这里补充一下

实现原理

利用python 爬虫完成对页面的访问

流程

https://blog.csdn.net/你的博客名/article/list/页码

1.通过这个api访问所有页，拿到所有文章。

https://blog.csdn.net/你的博客名/article/details/105360586

2.每一篇文章有唯一的id，解析页面获得文章id以及发布时间等信息。

https://blog-console-api.csdn.net/v1/editor/getArticle?id=文章id

3.你拿到文章id后，上面那个api访问就可以得到一个包含文章所有信息的json，解析拿到markdowncontent、文章标题、tags、key等信息。

4.保存为hexo格式的markdown。

两个问题

1.原文中这句

reply = requests.get(url)
parse = BeautifulSoup(reply.content, "lxml")

运行报错。
解决：

 headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.67 Safari/537.36 Edg/87.0.664.47'}
    reply = requests.get(url, headers=headers)
    parse = BeautifulSoup(reply.content, "html.parser")

2.原文这句

 headers = {
        "cookie":"uuid_tt_dd=10_20621362900-1586421086666-163599; dc_session_id=10_1586421086666.420623; Hm_lvt_6bcd52f51e9b3dce32bec4a3997715ac=1586421618; dc_sid=d4ceee41911ac755c162110ff811aee3; __gads=ID=608336bee91baf3d:T=1586421689:S=ALNI_MZozulzITWLw3Hxzo3jrnu5fmz8CA; c_ref=https%3A//blog.csdn.net/pang787559613/article/list/2; c-toolbar-writeguide=1; SESSION=3b5e7c88-b27d-4fcc-a2d5-4e97c1438a3c; UN=pang787559613; Hm_ct_6bcd52f51e9b3dce32bec4a3997715ac=6525*1*10_20621362900-1586421086666-163599!5744*1*pang787559613; announcement=%257B%2522isLogin%2522%253Atrue%252C%2522announcementUrl%2522%253A%2522https%253A%252F%252Fblog.csdn.net%252Fblogdevteam%252Farticle%252Fdetails%252F105203745%2522%252C%2522announcementCount%2522%253A0%252C%2522announcementExpire%2522%253A3600000%257D; UserName=pang787559613; UserInfo=604f13922dc04f2d8071fe0834e95db3; UserToken=604f13922dc04f2d8071fe0834e95db3; UserNick=%E7%AC%91%E8%83%96%E4%BB%94; AU=5FE; BT=1586422795507; p_uid=U000000; Hm_lpvt_6bcd52f51e9b3dce32bec4a3997715ac=1586422842; TY_SESSION_ID=ffc735f4-f5ae-40ed-9b98-ff01e337bf76; dc_tos=q8ijsv",
        "user-agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.120 Safari/537.36"
    }

一开始我没在意这句，后面运行出错，找了一下午，最终结论就是，cookie很重要，如果没cookie，访问页面的时候就会报错，叫你先登录再操作。这个cookie就是保存登录记录，python爬取的时候就不用再登录了。

解决：

查看自己的cookie

我用的谷歌浏览器。

1. 到csdn的首页，f12进去
   
   如果是空白那就刷新一下

2. 往下划一滑就可以找到cookie
   
   把这个cookie复制过去，就可以不用登陆访问自己的博客页面了

代码

import json
import uuid
import time
import requests
import datetime
from bs4 import BeautifulSoup

def request_blog_list(page):
    """获取博客列表
    主要包括博客的id以及发表时间等
    """
    url = f'https://blog.csdn.net/你的博客名/article/list/{page}'
    headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.67 Safari/537.36 Edg/87.0.664.47'}
    reply = requests.get(url, headers=headers)
    parse = BeautifulSoup(reply.content, "html.parser")
    spans = parse.find_all('div', attrs={'class':'article-item-box csdn-tracking-statistics'})
    blogs = []
    for span in spans[:40]:
        try:
            href = span.find('a', attrs={'target':'_blank'})['href']
            # read_num = span.find('span', attrs={'class':'read_num'}).get_text()
            date = span.find('span', attrs={'class':'date'}).get_text()
            blog_id = href.split("/")[-1]
            blogs.append([blog_id, date])
        except:
            print('Wrong, ' + href)
    return blogs

def request_md(blog_id, date):
   
    """获取博客包含markdown文本的json数据"""
    url = f"https://blog-console-api.csdn.net/v1/editor/getArticle?id={blog_id}"

    # headers = {
    #     "cookie":"",
    #     "user-agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.120 Safari/537.36"
    # }
    headers = {
        'Cookie':'',
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.82 Safari/537.36'}
    data = {"id": blog_id}
    reply = requests.get(url, headers=headers, data=data)
    try:
        write_hexo_md(reply.json(), date)
    except Exception as e:
        print("***********************************")
        print(e)
        print(url)
        # print(reply.json())

def write_hexo_md(data, date):
    """将获取的json数据解析为hexo的markdown格式"""
    title = data["data"]["title"]
    title = title.replace("[", "【")
    title = title.replace("]", "】")
    tags = data["data"]["tags"]
    # # 页面唯一标识符，用于统计系统和评论系统
    # key = "key" + str(uuid.uuid4())
    time = f"{date[0]}-{date[1]}-{date[2]}"
    tag = "tags:\n    - " + "\n    - ".join(tags.split(","))
    #hexo内容顶部拼接：title、tag、key
    header = "---\n" + f"title: {title}\n" + f"date: {time}\n"+ tag + "\n" + "---\n\n"
    # header = "---\n" + f"title: {title}\n" + tag + "\n" + "---\n\n"
    #hexo内容顶部拼接：仅title
    # header = "---\n" + f"title: {title}\n" + "---\n\n"

    content = data["data"]["markdowncontent"].replace("@[toc]", "")

    # hexo格式markdown
    with open(f"blogs/{title}.md", "w", encoding="utf-8") as f:
        f.write(header + content)
    print(f"写入 {title}")


    # # 用来博客迁移。遂仅保留内容
    # with open(f"blogs/{title}.md", "w", encoding="utf-8") as f:
    #     f.write(content)
    # print(f"写入 {title}")


def main(total_pages=2):
    """
    获取博客列表，包括id，时间
    获取博客markdown数据
    保存hexo格式markdown
    """
    blogs = []
    for page in range(1, total_pages + 1):
        blogs.extend(request_blog_list(page))
    for blog in blogs:
        blog_id = blog[0]
        date = blog[1].split()[0].split("-")
        request_md(blog_id, date)
        time.sleep(1)