from ConfigParser import ConfigParser
import urllib, re, sys
from xml.sax.saxutils import unescape

def clean(data):
    try:
        data.decode('utf-8')
    except:
        data = data.decode('iso-8859-1').encode('utf-8')
    return unescape(data)

def ibm2config(html, config=None):

    if hasattr(html, 'read'):
        html = html.read()

    if not config:
        config = ConfigParser()

    urls  = dict(re.findall('var url(\d+)="(.*?)";',  html))
    feeds = dict(re.findall('var feed(\d+)="(.*?)";', html))

    name_re = re.compile("<b>([^>]*?)</b></td><td valign='top' " + 
        "align='right'><span id=\"feedtype(\d+)")
    names = dict([(name,value) for value,name in name_re.findall(html)])

    for i, url in urls.items():
        if not feeds[i]: continue

        if url.startswith('feed:'): url=url[5:]
        url = clean(url)

        if config.has_section(url):
            config.set(url, 'name', clean(feeds[i]))
        else:
            config.add_section(url)
            config.set(url, 'name', clean(names.get(i,feeds[i])))

    return config

if __name__ == "__main__":
    ibm2config(sys.stdin).write(sys.stdout)