#!/usr/bin/env python

# A hard-coded, naive HTML to Atom converter for sil2100.vexillium.org
# What could/should be done is rewriting this code to make it more generic,
# allowing the user to input what webpage should be served etc. But for now,
# everything is hard-coded.
# Atom entries are defined by <div class="entry">.

# Copyright (C) 2011 Lukasz 'sil2100' Zemczak <sil2100@vexillium.org>

import sys
from HTMLParser import HTMLParser

class AtomFeed(HTMLParser):

	def __init__(self):
		HTMLParser.__init__(self)

		# We use all these flags for our convinience
		self.proc_entry = False
		self.proc_title = False
		self.proc_date = False
		self.proc_time = False
		self.proc_intro = False
		self.div_count = 0
		self.proc_div = -1

		self.atom_title = ''
		self.atom_date = ''
		self.atom_time = ''
		self.atom_intro = ''
		self.atom_id = ''

		self.entries = ''
		self.latest_date = None
		self.latest_time = None

	def handle_starttag(self, tag, attrs):
		attrs = dict(attrs)

		if tag == 'div':
			# Count the number of div's we enter, so we know when the 'entry' div ends
			self.div_count += 1

		if self.proc_entry == False and tag == 'div' and (attrs['class'] == 'entry'):
			self.proc_entry = True
			self.proc_div = self.div_count
		elif self.proc_entry == True:
			if tag == 'h4':
				self.proc_title = True
			elif tag == 'a' and 'name' in attrs:
				self.atom_id = str(attrs['name']);
			elif tag == 'div':
				if (attrs['class'] == 'date'):
					self.proc_date = True
				elif (attrs['class'] == 'time'):
					self.proc_time = True
				elif (attrs['class'] == 'intro'):
					self.proc_intro = True

	def handle_data(self, data):
		# Copy the data, if we're in a reading state
		if self.proc_entry == True:
			if self.proc_title == True:
				self.atom_title += str(data)
			elif self.proc_date == True:
				self.atom_date += str(data)
			elif self.proc_time == True:
				self.atom_time += str(data)
			elif self.proc_intro == True:
				self.atom_intro += str(data)

	def handle_endtag(self, tag):
		if self.proc_entry == True and tag == 'div' and self.div_count == self.proc_div:
			self.proc_entry = False
			self.add_entry()
			self.proc_div = -1
			self.atom_title = ''
			self.atom_date = ''
			self.atom_time = ''
			self.atom_intro = ''
		elif self.proc_entry == True:
			if tag == 'h4':
				self.proc_title = False
			elif tag == 'div':
				if self.proc_date == True:
					self.proc_date = False
				elif self.proc_time == True:
					self.proc_time = False
				elif self.proc_intro == True:
					self.proc_intro = False

		if tag == 'div':
			self.div_count -= 1

	def start_feed(self):
		# A hard-coded header, modify if needed
		print '<?xml version="1.0" encoding="utf-8"?>'
		print ''
		print '<feed xmlns="http://www.w3.org/2005/Atom">'
		print ''
		print '	<title>sil2100//vx web-page - Development</title>'
		print '	<id>http://sil2100.vexillium.org:41/</id>'
		print '	<link rel="self" href="http://sil2100.vexillium.org/dev_feed.xml" />'
		print '	<link href="http://sil2100.vexillium.org" />'
		print '	<updated>' + self.latest_date + 'T' + self.latest_time + ':00Z</updated>'
		print '	<author>'
		print '		<name>Lukasz Zemczak</name>'
		print '		<email>sil2100@vexillium.org</email>'
		print '	</author>'
		print ''

	def add_entry(self):
		if len(self.atom_time) == 0:
			self.atom_time = '00:01'

		if self.latest_date is None:
			self.latest_date = self.atom_date
			self.latest_time = self.atom_time

		self.entries += '	<entry>\n'
		self.entries += '		<id>tag:sil2100.vexillium.org,' + self.atom_date + ':/?id=dev/' + self.atom_id + '</id>\n'
		self.entries += '		<title>' + self.atom_title + '</title>\n'
		self.entries += '		<link href="http://sil2100.vexillium.org/?id=dev#' + self.atom_id + '" />\n'
		self.entries += '		<updated>' + self.atom_date + 'T' + self.atom_time + ':00Z</updated>\n'
		self.entries += '		<summary>' + self.atom_intro + '</summary>\n'
		self.entries += '	</entry>\n'
		self.entries += '\n'

	def end_feed(self):
		print '</feed>'

	def generate_feed(self):
		# We print out the Atom feed generated
		self.start_feed()
		print self.entries
		self.end_feed()


def main():
	if len(sys.argv) < 2:
		print 'Usage: atom_script.py page.html'
		return

	f = open(sys.argv[1], 'r')
	feed = AtomFeed()
	feed.feed(f.read())
	feed.generate_feed()

if __name__ == '__main__': main()