ogloszenia.wsgi

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

"""
Wyświetla RSS dla ogłoszeń umieszczonych na stronie wz.uni.lodz.pl.

BY d33tah, LICENSED UNDER WTFPL.

Wymaga RSS2Gen: http://www.dalkescientific.com/Python/PyRSS2Gen.html
"""

db_config = {'user':'some_user','passwd':'some_password','db':'some_db'}

import urllib 
from lxml import html
import PyRSS2Gen
import datetime
import time
import MySQLdb
#from config import db_config


def application(environ, start_response):
  frequency = 3 #for caching purposes
  now = int(time.time())
  url = 'http://zarzadzanie.uni.lodz.pl/Stronag%c5%82%c3%b3wna/' + \
  'Wyszukiwarkaog%c5%82osze%c5%84/tabid/169/language/pl-PL/'   + \
  'Default.aspx?uid='+environ["QUERY_STRING"]
  
  #open the db and re-initialize it if needed
  conn = MySQLdb.connect(**db_config)
  c = conn.cursor()
  #c.execute("set charset latin2;")
  c.execute("create table IF NOT EXISTS `cache`  (url VARCHAR(256) CHARACTER SET utf8 COLLATE utf8_unicode_ci UNIQUE, value BLOB, lasttime BLOB);")
  
  #look for the entry for a given url. check its time, use the data if correct
  query = c.execute("SELECT * FROM cache WHERE url = %s", (url,) )
  entry = c.fetchone()
  if entry:
    if now - int(entry[2]) < frequency:
      page = entry[1]
    else:
      page = urllib.urlopen(url).read()#.decode('iso8859-2')#.encode('utf-8')
      c.execute("UPDATE cache SET lasttime = %s, value = %s" \
        + "WHERE url = %s", (now,page,url))
      conn.commit()
  else:
    page = urllib.urlopen(url).read()#.decode('iso8859-2')#.encode('utf-8')
    c.execute("INSERT INTO cache VALUES (%s,%s,%s)", (url,page,now))
    conn.commit()
  
  #start_response('200 OK', [('Content-type','application/rss+xml')])
  #return page
  page = page.decode('utf8')
  
  tree = html.fromstring(page)
  notices = tree.xpath('//table[@id="%s"]//td[@style="width:300px;"]'
    % 'dnn_ctr558_Search_grvWykladowca' )

  rss = PyRSS2Gen.RSS2(
  #title = 'test',
  title = "WZ UŁ - ogłoszenia".decode('utf-8'),
  link = "http://deetah.jogger.pl".decode('utf-8'),
  #description = 'test'
  description = "Kanał RSS zawiera najnowsze ogłoszenia od wybranych "
        "wykładowców Wydziału Zarządzania Uniwersytetu Łódzkiego.".decode('utf-8'),
  )
  
  for entry in notices:
    publishDate = entry.getprevious().text
    noticeText = (entry.text_content()
              .replace("\r",        '')
              .replace('\n\t'+12*' ',   '') #some HTML align
          )[4:] #apparently, there are more spaces here :P

    if len(noticeText)>15 :
      summary = '[%s] %s (...)' % ( publishDate, 
                    noticeText.replace('\n','')[:15] )

    rss.items.append(PyRSS2Gen.RSSItem(
    title = summary,
    description = noticeText,
    link = url,
    guid = PyRSS2Gen.Guid( publishDate + noticeText ),
    pubDate = datetime.datetime(*(int(x) for x in publishDate.split('-')))
    ))

  start_response('200 OK', [('Content-type','application/rss+xml'), ])
  #start_response('200 OK', [('Content-type','application/rss+xml'), ('Content-encoding','utf-8') ])
  return rss.to_xml(encoding='iso-8859-2')

#print application({"QUERY_STRING":"135"},'')