#!/usr/bin/python
"""
rss2latex
Author: Nathan Charles <ncharles at gmail dot com>
Version: 0.1
This program renders a pdf of rss feeds

This program has no warranty to the full extent of the law
"""

import urllib2
import sys
import time
from xml.dom import minidom

def convert_to_tex(input):
	"""
	Strip out HTML codes and replace with LaTeX

	"""
	out = input.replace("&#8217;","'")
	out = out.replace("&#8211;","-")
	out = out.replace("&#8212;","--")
	out = out.replace("&#8220;","``")
	out = out.replace("&#8221;","''")
	out = out.replace("\&#8230;","\ldots")
	out = out.replace("<br />","\\\\")
	out = out.replace("<p>","")
	out = out.replace("</p>","\\\\\n")
	out = out.replace("&","\\amper")
	out = out.replace("%","\%")
	out = out.replace("$","\$")
	out = out.replace("_","\_")
	out = out.replace(u"\u2019","?")
	out = out.replace(u'\xa0',"?")
	return out
		

def renderfeed(address):
	texstream = "\\section*{\href{%s}{%s}}\\\\\n" % (address, address)
	file_request = urllib2.Request(address)
	file_opener = urllib2.build_opener()
	file_feed = file_opener.open(file_request).read()
	file_xml = minidom.parseString(file_feed)

	item_node = file_xml.getElementsByTagName("item")
	
	for item in item_node:
		title = item.getElementsByTagName("title")[0].firstChild.data
		
		link = item.getElementsByTagName("link")[0].firstChild.data
		timestamp = item.getElementsByTagName('pubDate')[0].firstChild.data
		entry = item.getElementsByTagName('content:encoded')[0].firstChild.data
		#rtime = time.strptime(timestamp,"%a %d %b %Yi %H:%M:%S %Z")
		ctitle = convert_to_tex(title)
		texstream += "\n\\subsection*{%s}" % ctitle
		texstream += "\n\\noindent \href{%s}{%s}" % (link, link)
		texstream += "\n\\timestamp{%s}\\\\" % timestamp
		print >> sys.stderr, "attempting converstion"
		clean =  convert_to_tex(entry)

		if clean == entry: 
			print >> sys.stderr, "conversion failed"
			texstream += entry
		else:
			print >> sys.stderr, "success" 
			texstream += clean
			
	return texstream


texheader = """
%!TEX TS-program = xelatex
%!TEX encoding = UTF-8 Unicode

\\documentclass[11pt, a4paper]{article}
\\usepackage{fontspec}

\\usepackage{geometry}
\\geometry{a4paper, textwidth=5.5in, textheight=8.5in, marginparsep=7pt, marginparwidth=.6in}
\\setlength\\parindent{0in}

\\defaultfontfeatures{Mapping=tex-text} 
\\setromanfont [Ligatures={Common}, Numbers={OldStyle}]{Hoefler Text}
\\setmonofont[Scale=0.8]{Monaco}
\\setsansfont[Scale=0.9]{Optima Regular}

\\newcommand{\\amper}{{\\fontspec[Scale=.95]{Hoefler Text}\\selectfont\\itshape\\&}}
\\newcommand{\\timestamp}[1]{\\marginpar{\\scriptsize #1}}

\\usepackage{sectsty}
\\usepackage[normalem]{ulem}
\\sectionfont{\\sffamily\\mdseries\\large\\underline}
\\subsectionfont{\\rmfamily\\mdseries\\scshape\\normalsize}
\\subsubsectionfont{\\rmfamily\\bfseries\\upshape\\normalsize}

\\usepackage[dvipdfm, bookmarks, colorlinks, breaklinks, pdftitle={rss2latex}]{hyperref}
\\hypersetup{linkcolor=blue,citecolor=blue,filecolor=black,urlcolor=blue}

\\begin{document}
\\reversemarginpar\\textsf{rss2latex} 
"""

texfooter = """
\\end{document}
"""

def rss2latex(feeds):
	if feeds:
		latexfile = texheader
		for feed in feeds:
			latexfile += renderfeed(feed)
		latexfile += texfooter
		return latexfile

def main ():
	filename = 'rss2latex.tex'
	address = ['http://char1es.net/feed/']
	texsource = rss2latex(address)
	texfile = open(filename, 'w')
	texfile.write(texsource)
	texfile.close()

main()
if __name__ == "__main__":
	    main()