#!/usr/bin/python

'''rsspubblica.py
Screen scraping di repubblica.it, ne ottengo un feed RSS
'''

print "Content-type: application/xml\n\n";

import re, urllib

uri = "http://www.repubblica.it"
doc = urllib.urlopen(uri).read()

rt = re.compile('<!--titolo ([0-9]*)-->(.*?)<!--/titolo ([0-9]*)-->', re.S)
t = rt.findall(doc)

rs = re.compile('<!--sommario ([0-9]*)-->(.*?)<!--/sommario ([0-9]*)-->', re.S)
s = rs.findall(doc)

rf = re.compile('/sommario 02-->(.*?)<td width', re.S)
f = rf.findall(doc)

ra = re.compile('<span class="txt14">(.*?)(<img|</td>)', re.S)
a = ra.findall(f[0])

ri = re.compile('<a href="(.*?)">', re.S)

rlink = re.compile('^(.*?)</span>', re.S)

def clean(text):
    text = text.replace('  ','')
    text = text.replace('\r\n','')
    text = text.replace('&agrave;','a\'')
    text = text.replace('&egrave;','e\'')
    text = text.replace('&eacute;','e\'')
    text = text.replace('&igrave;','i\'')
    text = text.replace('&ograve;','o\'')
    text = text.replace('&ugrave;','u\'')
    r = re.compile('<(.*?)>')
    text = r.sub(' ', text)
    return text
    
def cleanDescription(text):
    text = text.replace('href="/', 'href="' + uri + '/')
    text = text.replace('<', '&lt;')
    return text

print """
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:sy="http://purl.org/rss/1.0/modules/syndication/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns="http://purl.org/rss/1.0/">
<channel rdf:about="http://cavedoni.com/varie/python/rsspubblica">
  <title>La Repubblica</title>
  <link>http://cavedoni.com/varie/python/rsspubblica</link>
  <description>I titoli della home page di repubblica.it</description>
  <language>it</language>
  <items><rdf:Seq>"""
for item in range(len(t)):
    iuri = ri.findall(t[item][1])
    zuri = uri + iuri[0]
    print '    <rdf:li rdf:resource="' + zuri + '" />'

for item in range(len(a)):
    iuri = ri.findall(a[item][0])
    zuri = uri + iuri[0]    
    print '    <rdf:li rdf:resource="' + zuri + '" />'
    
print """  </rdf:Seq></items>
</channel>"""

for item in range(len(s)):
    iuri = ri.findall(t[item][1])
    zuri = uri + iuri[0]
    print '<item rdf:about="' + zuri + '">'
    print "  <link>"+zuri+"</link>"
    print "  <title>" + clean(t[item][1]) + "</title>"
    print "  <description>" + clean(s[item][1]) + "</description>"
    print "</item>"

for item in range(len(a)):
    iuri = ri.findall(a[item][0])
    title = rlink.findall(a[item][0])
    zuri = uri + iuri[0]
    print '<item rdf:about="' + zuri + '">'
    print '  <title>' + clean(title[0]) + '</title>'
    print "  <link>"+zuri+"</link>"
    print "  <description>" + cleanDescription(a[item][0]) + "</description>"
    print "</item>"    

print "</rdf:RDF>"