#!/usr/bin/env python
# -*- coding: iso-8859-1 -*-

# rss_parser.cgi  www.braggtown.com  1/2006
# XML/RSS parser to grab and display feeds on web page
# To add or remove feeds add a unique name and the rss url to the url_dict in the global class. 
# Don't forget to also add an option with that same feed name to your selector form.
# The feed display works be using Python's string.replace().  The cgi replaces a variable, $FEEDS, with the gathered feed information.
# Make sure that you place $FEEDS in your web page at the location you'd like the rss feeds displayed.

# Thanks to GazaM and Van_Gogh from ubuntuforums.org for their contributions.


### Import Python modules
import cgi, cgitb; cgitb.enable()
from urllib import urlopen
from xml.dom import minidom


# Global.  Class to hold global variables.  Alleviates having to pass references between functions.  
class Global:
	feeds = []
	super_item_object_list = []
	feed_list = []
	feed_urls = []	
	url_dict = {"slashdot": "http://rss.slashdot.org/Slashdot/slashdot", "osnews": "http://osnews.com/files/recent.rdf", \
		   "distorwatch": "http://distrowatch.com/news/dw.xml", "zdnet": "http://news.zdnet.com/2509-1_22-0-20.xml", \
		   "bbc": "http://fxfeeds.mozilla.org/rss20.xml", "thinkprogress": "http://thinkprogress.org/feed/", \
		   "asiatimes": "http://www.atimes.com/atimes/atol.xml", "ucimc": "http://indymedia.us/main-features-content.rss", \
		   "crooks": "http://www.crooksandliars.com/rss.xml", "guardian": "http://www.guardian.co.uk/rssfeed/0,,12,00.xml", \
		   "npr": "http://www.npr.org/rss/rss.php?id=1012", "bbctech": "http://newsrss.bbc.co.uk/rss/newsonline_world_edition/technology/rss.xml#", \
		   "google": "http://news.google.com/?output=rss", "freepress": "http://www.freepress.net/news/rss/news_all.rss"}


# populateFeeds.  Parse out variables passed from form
def populateFeeds(form):
	feed = form['feed']							# Set feed var to object containing all feeds selected
	if type(feed) is type([]):						# Feed var is list. Multiple RSS feeds selected		
		for item in feed: 				
			Global.feeds.append(item.value)	
	else:									# Feed var is string.  Single RSS feed 						
		Global.feeds.append(feed.value)

# printForm.  Print selector form
def printForm(form_template):
	form_template = open(form_template, 'r')
	print "Content-Type: text/html\n"
	print form_template.read()
	form_template.close()

# nameToUrl.  Take feed name, return feed url
def nameToUrl():
	for name in Global.feeds:
		Global.feed_urls.append(Global.url_dict[name])

# getAnchors.  Get list of names for named anchors for page navigation
def getAnchors():
	anchor_str = "| "
	for feed in Global.feed_list:
		anchor_str += """  <a href="#%s">%s</a> | """ % (feed[0], feed[0])	#(name, name)
	return anchor_str
	
# getFeeds. parser logic
def getFeeds(rss_icon, display_template):
	for url in Global.feed_urls:
		try:			
			content = urlopen(url).read()
			rss_xml = minidom.parseString(content)
			rss_xml.normalize()
			getFeedTitle(rss_xml)
			item_list = rss_xml.getElementsByTagName('item')
			getChildNodes(item_list)		
		except IOError:
			print " No valid feed at ", url
			pass	
	display_template = open(display_template, 'r')
	display_template_str = display_template.read()
	display_template.close()	
	anchor_str = getAnchors()				
	i = 0
	feed_output = """ <a name="top"> <p align="center">%s</p>""" % (anchor_str) + """<br /><br />"""
	for feed in Global.feed_list:										#(title, link, title)
		feed_output += """<br /><h1 align="center"><a name="%s"><a href="%s">%s</a></a></h1>""" % (feed[0], feed[1], feed[0])  
		for story in Global.super_item_object_list[i]:
			try:								
				if rss_icon:
					feed_output += """<img src="%s" border="0" />&nbsp; """ % (rss_icon)# name, url, name
				feed_output += """<a href="%s">%s</a><br />%s<br /> <br />""" % (story[1], story[0], story[2])  
	        	except:
				pass
		i += 1
		feed_output += """<p align="center">""" + """<a href="#top">Back to top</a>""" + "</p>" + "<hr />"
	print "Content-Type: text/html\n"
	print display_template_str.replace('$FEEDS', feed_output)

# getFeedTitle  Aquire the title of each rss feed
def getFeedTitle(rss_xml):
	channel_element = rss_xml.getElementsByTagName('channel')
        feed_title = (channel_element[0].getElementsByTagName('title')[0]).childNodes[0].data.encode('iso-8859-1')
        feed_link = (channel_element[0].getElementsByTagName('link')[0]).childNodes[0].data.encode('iso-8859-1')
        feed_list_object = [feed_title, feed_link]
	Global.feed_list.append(feed_list_object)
		
# getChildNodes takes list of items and append child nodes of each item to list
def getChildNodes(item_list):
	item_object_list = []
        for item in item_list: 
        	try:
        		title = (item.getElementsByTagName('title')[0]).childNodes[0].data.encode('iso-8859-1')
        		link = (item.getElementsByTagName('link')[0]).childNodes[0].data.encode('iso-8859-1')
			description = (item.getElementsByTagName('description')[0]).childNodes[0].data.encode('iso-8859-1')
			item_object = [title, link, description]
			item_object_list.append(item_object)
		except:
			pass
	Global.super_item_object_list.append(item_object_list)

# main application logic
def main():	
	### Set the following variables
	form_template =	"../includes/select_form.html"		# Set to location of your selection form
	rss_icon = "../images/rss.png"	# Set to location of icon displayed next to titles. Change to False (no quotes) if no icon desired.	
	display_template = "../includes/display_feeds.html"	# Set to location of page feeds displayed in
	###		
	form = cgi.FieldStorage()
	if form.has_key('feed'):				# Selector has been called previously.  Show feeds.
		populateFeeds(form)#ok
		nameToUrl()#ok
		getFeeds(rss_icon, display_template)
	else:							# Selector has not been called.  Print form.
		printForm(form_template)
	
### Call main function				
if (__name__ == "__main__"):
   main()