trans all blog entities from phpweblog.net to markdown format. each entity a file. include all files and images.

phpweblog2markdown.py

#!/usr/bin/env python
#coding=utf-8

from xml.dom import minidom as md
from datetime import datetime
import re
from urllib2 import urlopen
from os.path import basename
from socket import setdefaulttimeout


xml = md.parse('MyBlogData.xml')
path = 'posts/'
media_path = 'media/'
files_path = media_path + 'files/'
images_path = media_path + 'images/'
content = """---
date: %s
layout: post
title: '%s'
permalink: '/%s/%s.html'
---

%s
"""

setdefaulttimeout(3)

for e in xml.getElementsByTagName('blog_Content'):
    blogID = int(e.getElementsByTagName('ID')[0].childNodes[0].nodeValue)
    blogType = int(e.getElementsByTagName('PostType')[0].childNodes[0].nodeValue)
    blogTitle = e.getElementsByTagName('Title')[0].childNodes[0].nodeValue.replace('  ', ' ')
    blogCreated = datetime.strptime(e.getElementsByTagName('DateAdded')[0].childNodes[0].nodeValue, '%Y-%m-%dT%H:%M:%S')
    blogAuthor = e.getElementsByTagName('Author')[0].childNodes[0].nodeValue
    if e.getElementsByTagName('Email'):
        blogEmail = e.getElementsByTagName('Email')[0].childNodes[0].nodeValue
    if e.getElementsByTagName('Description'):
        blogDescription = e.getElementsByTagName('Description')[0].childNodes[0].nodeValue
    blogUpdated = datetime.strptime(e.getElementsByTagName('DateUpdated')[0].childNodes[0].nodeValue, '%Y-%m-%dT%H:%M:%S')
    blogText = e.getElementsByTagName('Text')[0].childNodes[0].nodeValue
    blogText = blogText.replace("\"BACKGROUND-COLOR: #ffffff\"", "")
    blogText = blogText.replace("background-color: #ffffff;", "")
    blogText = blogText.replace(" ", " ")
    blogText = blogText.replace("<!--<br><br>Code highlighting produced by Actipro CodeHighlighter (freeware)<br>http://www.CodeHighlighter.com/<br><br>-->", "")
    blogText = blogText.replace("<!--<br /><br />Code highlighting produced by Actipro CodeHighlighter (freeware)<br />http://www.CodeHighlighter.com/<br /><br />-->", "")
    blogText = blogText.replace("<img alt=\"\" src=\"http://www.phpweblog.net/Images/dot.gif\" />", "...")
    blogText = blogText.replace("<img alt=\"\" src=\"/Images/dot.gif\" />", "...")

    blogImages = re.findall('src="(.+?)"', blogText)
    blogFiles  = re.findall('href="(\/Files.+?)"', blogText)
    for url in blogImages:
        try:
            if url.startswith('/'):
                img = urlopen('http://www.phpweblog.net' + url).read()
            else:
                img = urlopen(url).read()
        except Exception, e:
            print url, e

        out = open(images_path + basename(url), 'w')
        out.write(img)
        out.close()
        blogText = re.sub('src="'+ url +'"', 'src="/images/' + basename(url) + '"', blogText)
    for url in blogFiles:
        try:
            if url.startswith('/'):
                f = urlopen('http://www.phpweblog.net' + url).read()
            else:
                f = urlopen(url).read()
        except Exception, e:
            print url, e

        out = open(files_path + basename(url), 'w')
        out.write(f)
        out.close()
        blogText = re.sub('src="'+ url +'"', 'src="/files/' + basename(url) + '"', blogText)

    if blogType == 1:
        print blogTitle, blogCreated
        f = file(path+str(blogID)+'.md', 'w')
        f.write(content % (blogCreated.strftime('%Y-%m-%d %H:%M:%S'), blogTitle.encode('utf-8'), blogCreated.strftime('%Y-%m-%d'), blogTitle.encode('utf-8').replace(' ', '-'), blogText.encode('utf-8')))
        f.close()
        print 'Done.'