#!/usr/bin/python
#
# update-planet
#
# Downloads feeds from the URLs specified and generates the XHTML files.
#
# (c) 2004, Davyd Madeley <davyd@ucc.asn.au>
#

import sys, urllib2, codecs
import XMLParse, XMLWriter

# step 1: read in the config and download the feeds
feeds	= []
for feed in open('feedlist').readlines():
	if feed.strip()[0] != '#':
		storage		= feed.strip().split('\t')
		name, feed	= storage[0], storage[-1]
		sys.stdout.write('Downloading feed "%s" from %s... ' % (name, feed))
		try:
			# XXX: might want to consider some good caching code in here
			feeds.append((name, feed, urllib2.urlopen(feed).read()))
			sys.stdout.write('done.\n')
		except:
			sys.stdout.write('failed.\n')

# step 2: process each feed
blogs	= []
for feed in feeds:
	xml	= XMLParse.XMLParse(feed[2]).parse()
	for blog in xml:
		blog.blogTitle	= feed[0]
	blogs	+= xml

# step 3: write feed to disk
try:
	codecs.open('planet.html', 'wb', 'utf-8').write(XMLWriter.XMLWriter(XMLWriter.XHTMLWriter, blogs).write())
except:
	sys.stderr.write('DEBUG: update-planet: could not write planet.html, aborting\n')
	raise