bbcode.py

import urllib2
import urllib
import json
import memcache
import re

#[url=http://example.com]Example[/url] OR [url]http://example.com[/url]
bbcode_url_re = re.compile('\[url(|\=(?P<url>.*?))\](?P<url_or_title>.*?)\[/url]')
cache = memcache.Client(['127.0.0.1:11211'])
MAX_WIDTH = 600

def get_services():
    result = cache.get('embedly_services')
    if result is None:
        result = urllib2.urlopen('http://api.embed.ly/v1/api/services/python').read()
        cache.set('embedly_services', result, 60*60*24)
    return json.loads(result)

def get_oembed(url):
    params = {'url' : url,
              'maxwidth' : MAX_WIDTH,
              'format' : 'json'}
    fetch_url = 'http://api.embed.ly/v1/api/oembed?%s' % urllib.urlencode(params)
    try:
        result = urllib2.urlopen(fetch_url).read()
    except:
        return None
    return json.loads(result)

def match(url):
    for service in get_services():
        for regex in service['regex']:
            if re.match(regex, url):
                return True
    return False

def replace(matchobj):
    url = matchobj.groupdict().get('url', None)
    title = matchobj.groupdict().get('url_or_title', None)
    if url is None and title:
        url = title
        title is None

    #Not Something Embedly Handles
    if not match(url):
        matchobj.group()
    
    oembed = get_oembed(url)
    #embed was not found or it's a link type
    if oembed is None or oembed['type'] == 'link':
        return matchobj.group()
    elif oembed['type'] in ['video', 'rich']:
        return '<div class="embed">%s</div>' % oembed['html']
    elif oembed['type'] == 'photo':
        return '<div class="embed"><a href="%s" title="%s"><img src="%s"></img></a></div>'% (url,oembed.get('title', ''), oembed['url'])

    #bad type?
    return matchobj.group()

def parse(text):
    """
    Parse a piece of text for bbcodes and insert the correct embed code
    """
    return bbcode_url_re.sub(replace, text, re.M )

if __name__ == '__main__':
    random_text = """ This is some text
    [url]http://soundcloud.com/jameszabiela/blame-edit[/url] replaced
    [url=http://twitpic.com/1sw5l1]Nice Hair[/url] replaced
    [url=http://example.com]Example[/url] not replaced
    """
    print parse(random_text)