#!/usr/bin/env python
#
# pnu2rss by Pieter Edelman
# This script converts the Physics News Update (http://www.aip.org/pnu/) to an
# RSS feed. This update appears about once a week, so please choose your
# update interval wisely.
#
# ============================================================================
# WARNING: THIS IS A STAND-ALONE SCRIPT, NOT A FILTER!!
# The reason for this is that the PNU website only mentions when a new issue
# has appeared. For contents an extra link should be followed.
# It works with the PNU issues from 2004 onwards.
# ============================================================================
#
# Copyright (c) 2004 Pieter Edelman
# Released under the terms of the GNU General Public License (GPL) Version 2.
# See http://www.gnu.org/ for details.

import urllib, re, sys

# The matches used in this script:
blockquote = re.compile("</blockquote>");
whitespace = re.compile("^\s*");
eol        = re.compile("\r\n");
paragraph  = re.compile("<p>");
tags       = re.compile("<(.*?)>");
ampersand  = re.compile("&");
headline   = re.compile("p class=\"headline\"><PNUZONE\d>(.*)</PNUZONE\d></p>");

# Print the header
print "<?xml version=\"1.0\" encoding=\"ISO-8859-1\"?>";
print "<!DOCTYPE rss PUBLIC \"-//Netscape Communications//DTD RSS 0.91//EN\"";
print "\"http://my.netscape.com/publish/formats/rss-0.91.dtd\">";
print "<rss version=\"0.91\">";
print "<channel>";
print "  <title>Physics News Update</title>";
print "  <link>http://www.aip.org/pnu/</link>";
print "  <description>The AIP Bulletin of Physics News</description>";
print "  <image>";
print "    <url>http://www.aip.org/pnu/images/pnulogo.gif</url>";
print "  </image>";

# First, get the link to the newest issue from the frontpage
try:
  # Open the channel
  webpage = urllib.urlopen("http://www.aip.org/pnu/");
  
  # Keep reading until the issue is found
  line = webpage.readline();
  while (line):
    match = re.search("td bgcolor=\"#ffff99\" colspan=\"2\"><a href=\"(/pnu/.*html)\"><span class=\"headline\">Update #\d*\s*</span></a>.*</td>", line);
    if (match):
      # If the line is found, construct the link, close the channel, and skip out of the loop
      issue = "http://www.aip.org" + match.groups()[0];
      webpage.close();
      break;
    line = webpage.readline();
except:
  sys.exit();

# Fetch the issue page
try:
  webpage = urllib.urlopen(issue);
except:
  sys.exit();

# Keep track of the number of items within a page
counter = 0;

# Start reading the page
line = webpage.readline();
while (line):
  # Search for headlines
  match = re.search(headline, line);

  if (match):
    # Save the title
    title = match.groups()[0];
    
    # Construct the link to the particular item
    counter = counter + 1;
    link = re.sub("(.*)/(.*)\.html", r"\1/split/\2-" + str(counter) + ".html", issue);
    
    # Skip over the <blockquote> statement
    webpage.readline();
    
    # Append lines to the description until </blockquote> is found
    description = "";
    while (line):
      line = webpage.readline();
      if (re.search(blockquote, line)): break;
      line = re.sub(whitespace, "", line);
      line = re.sub(eol, "", line);
      line = re.sub(paragraph, "      \n\n      <p>", line);
      line = re.sub(ampersand, r"&amp;", line);
      line = re.sub(tags, r"&lt;\1&gt;", line);
      description += line;
      
    # Print out the item
    print "  <item>";
    print "    <title>" + title + "</title>";
    print "    <link>" + link + "</link>";
    print "    <description>" + description;
    print "    </description>";
    print "  </item>";
  
  # Read the next line
  line = webpage.readline();

# Close the channel
webpage.close();

# Print the footer
print "  </channel>";
print "</rss>";