#!/usr/bin/env python

# Script that reads through all of the *.ht files, determines their
# dates, and outputs a tab-separated file of news items to standard output.

import glob, re, sys
import rfc822

pat = re.compile('\s*(\d{4}-\d{1,2}-\d{1,2}).*(\d{4}-\d{1,2}-\d{1,2})')

def main ():
    L = []
    for filename in glob.glob('20*.ht'):
        input = open(filename, 'r')
        msg = rfc822.Message(input)
        title = msg.getheader('title')
        if title is None:
            print >>sys.stderr,  'File %s lacks title -- skipping' % filename
        input.close()

        m = pat.search(title)
        if m is None:
            print >>sys.stderr, ('No start/end dates found in title of %s\n%r'
                                 % (filename, title))
            continue
        
        start, end = m.group(1,2)
        url = 'http://www.python.org/dev/summary/%sml' % filename
        L.append((end, title, url))

    L.sort() ; L.reverse()
    for date, title, url in L:
        print date + '\t' + title + '\t' + url
            
if __name__ == '__main__':
    main()