Python 博客园备份迁移脚本

鉴于有些小伙伴在寻找博客园迁移到个人博客的方案，本人针对博客园实现了一个自动备份脚本，可以快速将博客园中自己的文章备份成Markdown格式的独立文件，备份后的md文件可以直接放入到hexo博客中，快速生成自己的站点，而不需要自己逐篇文章迁移，提高了备份文章的效率。

首先第一步将博客园主题替换为 codinglife默认主题，第二步登录到自己的博客园后台，然后选择 博客备份，备份所有的随笔文章，如下所示：

• https://i1.cnblogs.com/BlogBackup.aspx

备份出来以后将其命名为 backup.xml，然后新建一个 main.py脚本，以及一个 blog目录，代码实现的原理是，解析xml格式并依次提取出文档内容，然后分别保存为markdown文件。

转存文章到MarkDown格式: 写入备份脚本，代码如下所示，运行后即可自动转存文件到 blog目录下，当运行结束后备份也就结束了。

powerby: LyShark
blog: www.cnblogs.com/lyshark
from bs4 import BeautifulSoup
import requests, os,re

header = {"User-Agent": "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) By LyShark CnblogsBlog Backup Script"}

获取文章,并转成markdown
blog: www.lyshark.com
def GetMarkDown(xml_file):
    con = open(xml_file, 'r', encoding='utf8').read()
    # 每篇文章都在  标签里
    items = re.findall(".*?", con, re.I | re.M | re.S)
    ele2 = ['(.+?)', '(.+?)', '(.+?)']
    # md_name = xml_file.split('.xml')[0] + '.md'
    for item in items:
        try:
            title = re.findall(ele2[0], item, re.I | re.S | re.M)[0]
            link = re.findall(ele2[1], item, re.I | re.S | re.M)[0]
            des = re.findall(ele2[2], item, re.I | re.S | re.M)[0]
            des = re.findall('', des, re.I | re.S | re.M)[0]  # CDATA 里面放的是文章的内容
            des = des.replace('~~~', "")
            lines = des.split('\n')
            with open("./blog/" + title.replace("/","") + ".md", mode='w+', encoding='utf8') as f:
                f.write("---\n")
                f.write("title: '{}'\n".format(title.replace("##","").replace("###","").replace("-","").replace("*","").replace("","").replace(":","").replace("：","").replace(" ","").replace(" ","").replace("","")))
                f.write("copyright: true\n")

                setdate = "2018-12-27 00:00:00"
                try:
                    # 读取时间
                    response = requests.get(url=link, headers=header)
                    print("读取状态: {}".format(response.status_code))

                    if response.status_code == 200:
                        bs = BeautifulSoup(response.text, "html.parser")
                        ret = bs.select('span[id="post-date"]')[0]
                        setdate = str(ret.text)
                        pass
                    else:
                        f.write("date: '2018-12-27 00:00:00'\n")
                except Exception:
                    f.write("date: '2018-12-27 00:00:00'\n")
                    pass

                f.write("date: '{}'\n".format(setdate))

                # description检测
                description_check = lines[0].replace("##","").replace("###","").replace("-","").replace("*","").replace("","").replace(":","").replace("：","").replace(" ","").replace(" ","")
                if description_check == "":
                    f.write("description: '{}'\n".format("该文章暂无概述"))
                elif description_check == "C":
                    f.write("description: '{}'\n".format("该文章暂无概述"))
                elif description_check == "":
                    f.write("description: '{}'\n".format("该文章暂无概述"))
                else:
                    f.write("description: '{}'\n".format(description_check))

                print("[*] 时间: {} --> 标题: {}".format(setdate, title))
                f.write("tags: '{}'\n".format("tags10245"))
                f.write("categories: '{}'\n".format("categories10245"))
                f.write("---\n\n")
                f.write('%s' %des)
                f.close()
        except Exception:
            pass

if __name__ == "__main__":
    GetMarkDown("backup.xml")

备份后的效果如下所示：

打开Markdown格式看一下，此处的标签和分类使用了一个别名，在备份下来以后，你可以逐个区域进行替换，将其替换成自己需要的分类类型即可。

转存图片到本地: 接着就是继续循环将博客中所有图片备份下来，同样新建一个 image文件夹，并运行如下代码实现备份。

powerby: LyShark
blog: www.cnblogs.com/lyshark
from bs4 import BeautifulSoup
import requests, os,re

header = {"User-Agent": "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) By LyShark CnblogsBlog Backup Script"}

从备份XML中找到URL
def GetURL(xml_file):
    blog_url = []
    con = open(xml_file, 'r', encoding='utf8').read()
    items = re.findall(".*?", con, re.I | re.M | re.S)
    ele2 = ['(.+?)', '(.+?)', '(.+?)']
    for item in items:
        try:
            title = re.findall(ele2[0], item, re.I | re.S | re.M)[0]
            link = re.findall(ele2[1], item, re.I | re.S | re.M)[0]
            print("标题: {} --> URL: {} ".format(title,link))
            blog_url.append(link)
        except Exception:
            pass

    return blog_url

下载所有图片
blog: www.lyshark.com
def DownloadURLPicture(url):
    params = {"encode": "utf-8"}
    response = requests.get(url=url, params=params, headers=header)
    # print("网页编码方式: {} -> {}".format(response.encoding,response.apparent_encoding))
    context = response.text.encode(response.encoding).decode(response.apparent_encoding, "ignore")
    try:
        bs = BeautifulSoup(context, "html.parser")
        ret = bs.select('div[id="cnblogs_post_body"] p img')
        for item in ret:
            try:
                img_src_path = item.get("src")
                img_src_name = img_src_path.split("/")[-1]
                print("[+] 下载图片: {} ".format(img_src_name))
                img_download = requests.get(url=img_src_path, headers=header, stream=True)
                with open("./image/" + img_src_name, "wb") as fp:
                    for chunk in img_download.iter_content(chunk_size=1024):
                        fp.write(chunk)
            except Exception:
                print("下载图片失败: {}".format(img_src_name))
                pass
    except Exception:
        pass

if __name__ == "__main__":
    url = GetURL("backup.xml")
    for u in url:
        DownloadURLPicture(u)

备份后的效果如下：

替换文章内的图片链接地址，可以使用编辑器，启用正则 https://img2020.cnblogs.com/blog/(.*?)/(.*?)/批量替换。

当把博客备份下来以后你就可以把这些文章拷贝到 hexo博客 _post目录下面，然后 hexo命令快速渲染生成博客园的镜像站点，这样也算是增加双保险了。

对过滤器的进一步优化，使之可以直接过滤出特定的标题，只需要稍微修改一下代码即可，如下代码可实现过滤特定关键词的文章。

powerby: LyShark
blog: www.cnblogs.com/lyshark
from bs4 import BeautifulSoup
import requests, os,re

header = {"User-Agent": "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) By LyShark CnblogsBlog Backup Script"}

获取文章,并转成markdown
blog: www.lyshark.com
def GetMarkDown(xml_file,grep_title,tags):
    con = open(xml_file, 'r', encoding='utf8').read()
    # 每篇文章都在  标签里
    items = re.findall(".*?", con, re.I | re.M | re.S)
    ele2 = ['(.+?)', '(.+?)', '(.+?)']
    # md_name = xml_file.split('.xml')[0] + '.md'
    for item in items:
        try:
            title = re.findall(ele2[0], item, re.I | re.S | re.M)[0]
            link = re.findall(ele2[1], item, re.I | re.S | re.M)[0]
            des = re.findall(ele2[2], item, re.I | re.S | re.M)[0]
            des = re.findall('', des, re.I | re.S | re.M)[0]  # CDATA 里面放的是文章的内容
            des = des.replace('~~~', "")
            lines = des.split('\n')

            # 判断是否存在:或者是空格
            if title.find("：") != -1 or title.find(":") != -1 or title.find(" ") != -1:

                # 过滤特定关键字
                if title.find(grep_title) != -1:
                    local_tags = ""
                    if local_tags == "":
                        local_tags = title.split("：")[0]
                    if local_tags == "":
                        local_tags = title.split(":")[0]
                    if local_tags == "":
                        local_tags = title.split(" ")[0]
                    if local_tags == "":
                        local_tags = title.split(" ")[0]
                    # print("[只过滤] 标签: {}".format(local_tags))

                    # 开始转存
                    with open("./blog/" + title.replace("/", "") + ".md", mode='w+', encoding='utf8') as f:
                        f.write("---\n")
                        f.write("title: '{}'\n".format(title.replace("##", "").replace("###", "").replace("-", "").replace("*", "").replace("","").replace(":", "：").replace("：", "：").replace(" ", "").replace(" ", "").replace("", "")))
                        f.write("copyright: true\n")

                        setdate = "2000-01-01 00:00:00"
                        try:
                            # 读取时间
                            response = requests.get(url=link, headers=header)

                            if response.status_code == 200:
                                bs = BeautifulSoup(response.text, "html.parser")
                                ret = bs.select('span[id="post-date"]')[0]
                                setdate = str(ret.text)
                                pass
                            else:
                                f.write("date: '2000-01-01 00:00:00'\n")
                        except Exception:
                            f.write("date: '2000-01-01 00:00:00'\n")
                            pass

                        f.write("date: '{}'\n".format(setdate))

                        # description 检测描述信息的输出
                        description_check = lines[0].replace("##", "").replace("###", "").replace("-", "").replace("*","").replace("", "").replace(":", "").replace("：", "").replace(" ", "").replace(" ", "")
                        if description_check == "":
                            f.write("description: '{}'\n".format("该文章暂无概述,自动增加注释"))
                        elif description_check == "C":
                            f.write("description: '{}'\n".format("该文章暂无概述,自动增加注释"))
                        elif description_check == "":
                            f.write("description: '{}'\n".format("该文章暂无概述,自动增加注释"))
                        elif description_check == "BASH":
                            f.write("description: '{}'\n".format("该文章暂无概述,自动增加注释"))
                        elif description_check == "":
                            f.write("description: '{}'\n".format("该文章暂无概述,自动增加注释"))
                        else:
                            f.write("description: '{}'\n".format(description_check))

                        print("[*] 状态: {} --> 时间: {} --> 标题: {}".format(response.status_code, setdate, title))

                        # 开始打标签
                        f.write("tags: '{}'\n".format(tags))
                        f.write("categories: '{}'\n".format(tags))
                        # print("[*] 打标签")
                        f.write("---\n\n")
                        f.write('%s' % des)
                        f.close()
        except Exception:
            pass

从备份XML中找到URL
def GetURL(xml_file,grep_title):
    blog_url = []
    con = open(xml_file, 'r', encoding='utf8').read()
    items = re.findall(".*?", con, re.I | re.M | re.S)
    ele2 = ['(.+?)', '(.+?)', '(.+?)']
    for item in items:
        try:
            title = re.findall(ele2[0], item, re.I | re.S | re.M)[0]
            link = re.findall(ele2[1], item, re.I | re.S | re.M)[0]
            # print("标题: {} --> URL: {} ".format(title,link))

            # 判断是否存在:或者是空格
            if title.find("：") != -1 or title.find(":") != -1 or title.find(" ") != -1:

                # 过滤特定关键字
                if title.find(grep_title) != -1:
                    # print("[只过滤] 标签: {}".format(grep_title))
                    blog_url.append([title,link])

        except Exception:
            pass

    return blog_url

下载所有图片
blog: www.lyshark.com
def DownloadURLPicture(title,url):
    params = {"encode": "utf-8"}
    response = requests.get(url=url, params=params, headers=header)
    # print("网页编码方式: {} -> {}".format(response.encoding,response.apparent_encoding))
    context = response.text.encode(response.encoding).decode(response.apparent_encoding, "ignore")
    try:
        print("当前文章: {} | URL: {}".format(title,url))

        bs = BeautifulSoup(context, "html.parser")
        ret = bs.select('div[id="cnblogs_post_body"] p img')
        for item in ret:
            try:
                img_src_path = item.get("src")
                img_src_name = img_src_path.split("/")[-1]
                print("[+] Down --> {} ".format(img_src_name))
                img_download = requests.get(url=img_src_path, headers=header, stream=True)
                with open("./image/" + img_src_name, "wb") as fp:
                    for chunk in img_download.iter_content(chunk_size=1024):
                        fp.write(chunk)
            except Exception:
                print("[-] DownError: {}".format(img_src_name))
                pass
    except Exception:
        pass

if __name__ == "__main__":
    backup_title = "驱动开发："
    tags = "Windows 内核安全编程技术实践"
    backup_file = "./backup.xml"

    GetMarkDown(backup_file,backup_title, tags)

    for u in GetURL(backup_file,backup_title):
        DownloadURLPicture(u[0],u[1])

Original: https://www.cnblogs.com/LyShark/p/16652464.html
Author: lyshark
Title: Python 博客园备份迁移脚本