Convert exported file from wordpress.com for importing posterous.

convert.py

# /usr/bin/env python
# -*- coding: utf-8 -*-

import sys
import re
from xml.dom.minidom import parse

def markdown_nize(html):
    ret = ''
    html = html.encode('utf-8')
    html = codeblock2md(html)
    html = link2md(html)
    html = h22md(html)
    html = h32md(html)
    html = ul2md(html)
    html = ol2md(html)

    ret += '<markdown>\n'
    ret += html
    ret += '</markdown>\n'

    return ret

def codeblock2md(html):
    ret = ''
    in_codeblock = False
    source_header = re.compile('\[sourcecode language=[\'"](\w+)[\'"].*?\]')
    source_footer = re.compile('\[/sourcecode\]')

    for line in html.split("\n"):
        source_header_match = source_header.search(line)
        source_footer_match = source_footer.search(line)

        if not in_codeblock and source_header_match:
            in_codeblock = True
            lang = source_header_match.groups()[0]
            if lang == 'cpp':
                ret += '    #!cpp\n'
        elif not in_codeblock:
            ret += line + "\n"
        elif in_codeblock and source_footer_match:
            in_codeblock = False
        elif in_codeblock:
            ret += '    ' + line + "\n"

    return ret

def link2md(html):
    ret = ''
    link = re.compile('<a.*?href="(.+?)".*?>(.+?)</a>')
    link_pos = re.compile('<a.*?>.*?</a>')

    for line in html.split("\n"):
        if link.search(line):
            pat = link.findall(line)
            pos = link_pos.findall(line)
            for i in xrange(len(pat)):
                md = "[{0}]({1})".format(pat[i][1], pat[i][0])
                line = line.replace(pos[i], md, 1)
        ret += line + "\n"

    return ret

def h22md(html):
    ret = ''
    h3 = re.compile('<h2.*?>(.+)</h2>')

    for line in html.split("\n"):
        res = h3.search(line)
        if res:
            md = "# {0} #".format(res.groups()[0])
            line = re.sub('<h2.*?>.*?</h2>', md, line)
        ret += line + "\n"

    return ret

def h32md(html):
    ret = ''
    h3 = re.compile('<h3.*?>(.+)</h3>')

    for line in html.split("\n"):
        res = h3.search(line)
        if res:
            md = "## {0} ##".format(res.groups()[0])
            line = re.sub('<h3.*?>.*?</h3>', md, line)
        ret += line + "\n"

    return ret

def ul2md(html):
    ret = ''
    in_list = False
    header = re.compile('<ul>')
    footer = re.compile('</ul>')
    list = re.compile('<li.*?>(.+)</li>')

    for line in html.split("\n"):
        header_match = header.search(line)
        footer_match = footer.search(line)

        if not in_list and header_match:
            in_list = True
        elif not in_list:
            ret += line + "\n"
        elif in_list and footer_match:
            in_list = False
        elif in_list:
            res = list.search(line)
            if res:
                md = "* {0}".format(res.groups()[0])
                line = re.sub('<li.*?>.*?</li>', md, line)
            ret += line + "\n"

    return ret

def ol2md(html):
    ret = ''
    in_list = False
    counter = 0
    header = re.compile('<ol>')
    footer = re.compile('</ol>')
    list = re.compile('<li.*?>(.+)</li>')

    for line in html.split("\n"):
        header_match = header.search(line)
        footer_match = footer.search(line)

        if not in_list and header_match:
            in_list = True
            counter = 1
        elif not in_list:
            ret += line + "\n"
        elif in_list and footer_match:
            in_list = False
        elif in_list:
            res = list.search(line)
            if res:
                md = "{0}. {1}".format(counter, res.groups()[0])
                line = re.sub('<li.*?>.*?</li>', md, line)
            ret += line + "\n"
            counter += 1

    return ret

def main():
    dom = parse(sys.argv[1])

    for element in dom.getElementsByTagName('content:encoded'):
        if element.childNodes:
            element.childNodes[0].nodeValue = markdown_nize(element.childNodes[0].nodeValue).decode('utf-8')

    print dom.toxml('utf-8')

if __name__ == '__main__':
    main()