#!/usr/bin/env python
"""
siteupdates.py - Prints out the most recently modified files in a site
License: GPL 2; share and enjoy!
Author: Sean B. Palmer, inamidst.com

Help and usage notes: 
   ./siteupdates.py --help
"""

import sys, os, time, robotparser
from optparse import OptionParser

def siteupdates(rootpath, t=None, n=None, strip=None, output=None): 
   if t is None: 
      t = (3600 * 24 * 7) # a week
   if output is None: 
      output = sys.stdout
   elif isinstance(output, basestring): 
      from StringIO import StringIO
      output = StringIO(output)

   now = time.time()
   then = now - t

   if rootpath.endswith('/'): 
      prefix = '/'
   else: prefix = ''

   robotstxt = os.path.join(rootpath, 'robots.txt')
   if os.path.exists(robotstxt): 
      rp = robotparser.RobotFileParser()
      rp.set_url(robotstxt)
      rp.read() # @@ this reads borked files too
   else: rp = None

   result = []
   for (dirpath, dirnames, filenames) in os.walk(rootpath): 
      for name in filenames: 
         fn = os.path.join(dirpath, name)
         mtime = os.stat(fn).st_mtime
         if mtime > then: 
            datetime = time.strftime('%Y-%m-%d %H:%M:%S', time.gmtime(mtime))
            filename = prefix + fn[len(rootpath):]
            if filename.endswith('.pyc'): continue
            if rp is not None: 
               if rp.can_fetch('*', filename): 
                  result.append((datetime, filename))
            else: result.append((datetime, filename))
   result.sort()
   result.reverse()
   if n is not None: 
      result = result[:n]

   day = None
   strext = ('index.html', '.html', 'index.cgi', 
             '.cgi', '.txt', '.jpg', '.png', '.php', '.rss')
   for (datetime, filename) in result: 
      if strip: 
         for ext in strext: 
            if filename.endswith(ext): 
               filename = filename[:-len(ext)]
               break

      if day is None: 
         day = datetime[:10]
         output.write('<h2 id="T%s">%s</h2>\n' % (day, day))
         output.write('<ul>\n')
      elif datetime[:10] != day: 
         day = datetime[:10]
         output.write('</ul>\n\n')
         output.write('<h2 id="T%s">%s</h2>\n' % (day, day))
         output.write('<ul>\n')
      line = '<li><a href="%s">%s</a> (%s)</li>\n'
      output.write(line % (filename, filename, datetime[-8:]))
   output.write('</ul>\n')
   return output

def main(argv=None): 
   parser = OptionParser(usage='%prog [options] <rootpath>')
   parser.add_option("-t", "--time", dest="time", default=False, 
                     help="cut off last modification date for inclusion")
   parser.add_option("-n", "--number", dest="number", default=False, 
                     help="maximum number of results to be output")

   options, args = parser.parse_args(argv)

   if len(args) == 1: 
      if options.time: 
         t = int(options.time)
      else: t = None
      if options.number: 
         n = int(options.number)
      else: n = None
      siteupdates(args[0], t, n)
   else: parser.error("Incorrect number of arguments.")

if __name__=="__main__": 
   main()