鑒于有些小伙伴在尋找博客園遷移到個(gè)人博客的方案，本人針對(duì)博客園實(shí)現(xiàn)了一個(gè)自動(dòng)備份腳本，可以快速將博客園中自己的文章備份成Markdown格式的獨(dú)立文件，備份后的md文件可以直接放入到hexo博客中，快速生成自己的站點(diǎn)，而不需要自己逐篇文章遷移，提高了備份文章的效率。

首先第一步將博客園主題替換為codinglife默認(rèn)主題，第二步登錄到自己的博客園后臺(tái)，然后選擇博客備份，備份所有的隨筆文章，如下所示：

備份出來(lái)以后將其命名為backup.xml，然后新建一個(gè)main.py腳本，以及一個(gè)blog目錄，代碼實(shí)現(xiàn)的原理是，解析xml格式并依次提取出文檔內(nèi)容，然后分別保存為markdown文件。

轉(zhuǎn)存文章到MD

寫(xiě)入備份腳本，代碼如下所示，運(yùn)行后即可自動(dòng)轉(zhuǎn)存文件到blog目錄下，當(dāng)運(yùn)行結(jié)束后備份也就結(jié)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"}

# 獲取文章,并轉(zhuǎn)成markdown
# blog: www.lyshark.com
def GetMarkDown(xml_file):
    con = open(xml_file, 'r', encoding='utf8').read()
    # 每篇文章都在 <item> 標(biāo)簽里
    items = re.findall("<item>.*?</item>", con, re.I | re.M | re.S)
    ele2 = ['<title>(.+?)</title>', '<link>(.+?)</link>', '<description>(.+?)</description>']
    # 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('<!\[CDATA\[(.+?)\]\]>', des, re.I | re.S | re.M)[0]  # CDATA 里面放的是文章的內(nèi)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("<br>","").replace(":","").replace("：","").replace(" ","").replace(" ","").replace("`","")))
                f.write("copyright: true\n")

                setdate = "2018-12-27 00:00:00"
                try:
                    # 讀取時(shí)間
                    response = requests.get(url=link, headers=header)
                    print("讀取狀態(tài): {}".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檢測(cè)
                description_check = lines[0].replace("##","").replace("###","").replace("-","").replace("*","").replace("<br>","").replace(":","").replace("：","").replace(" ","").replace(" ","")
                if description_check == "":
                    f.write("description: '{}'\n".format("該文章暫無(wú)概述"))
                elif description_check == "```C":
                    f.write("description: '{}'\n".format("該文章暫無(wú)概述"))
                elif description_check == "```Python":
                    f.write("description: '{}'\n".format("該文章暫無(wú)概述"))
                else:
                    f.write("description: '{}'\n".format(description_check))

                print("[*] 時(shí)間: {} --> 標(biāo)題: {}".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")

備份后的效果如下所示：

打開(kāi)Markdown格式看一下，此處的標(biāo)簽和分類使用了一個(gè)別名，在備份下來(lái)以后，你可以逐個(gè)區(qū)域進(jìn)行替換，將其替換成自己需要的分類類型即可。

轉(zhuǎn)存圖片到本地

接著就是繼續(xù)循環(huán)將博客中所有圖片備份下來(lái)，同樣新建一個(gè)image文件夾，并運(yùn)行如下代碼實(shí)現(xiàn)備份。

# 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"}

# 從備份XML中找到URL
# blog: www.cnblogs.com/lyshark
def GetURL(xml_file):
    blog_url = []
    con = open(xml_file, 'r', encoding='utf8').read()
    items = re.findall("<item>.*?</item>", con, re.I | re.M | re.S)
    ele2 = ['<title>(.+?)</title>', '<link>(.+?)</link>', '<description>(.+?)</description>']
    for item in items:
        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("標(biāo)題: {} --> URL: {} ".format(title,link))
        blog_url.append(link)
    return blog_url

# 下載所有圖片
# blog: www.lyshark.com
def DownloadURLPicture(url):
    params = {"encode": "utf-8"}
    response = requests.get(url=url, params=params, headers=header)
    # print("網(wǎng)頁(yè)編碼方式: {} -> {}".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("[+] 下載圖片: {} ".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("下載圖片失敗: {}".format(img_src_name))
                pass
    except Exception:
        pass

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

備份后的效果如下：