#!/usr/bin/env python
"""
Inside - Shows inside a site.
License: GPL 2. Share and enjoy!
Author: Sean B. Palmer, http://purl.org/net/sbp/
"""

import cgitb; cgitb.enable()

import sys, os, itertools, robotparser

# # # # # # # # #
# <Configurable>

indexen = ('index.cgi', 'index.py')
suffixen = ('.cgi', '.py', '.php')
overview = 'overview.html'

# </Configurable>
# # # # # # # # #

scriptname = os.environ.get('SCRIPT_NAME') or '/inside/index.cgi'
base, script = os.path.split(scriptname)
site = '../' * base.count('/')

if not os.path.exists('.htaccess'): 
   f = open('.htaccess', 'w')
   print >> f, 'DirectoryIndex %s' % script
   print >> f, 'Options -MultiViews'
   print >> f, 'RewriteEngine on'
   print >> f, 'RewriteBase %s' % base
   print >> f, 'RewriteRule ^(.+)$ %s [L]' % script
   f.close()

def all(seq, pred=bool): 
   """Returns True if pred(x) is True for every element in the iterable."""
   return False not in itertools.imap(pred, seq)

def any(seq, pred=bool):
   """Returns True if pred(x) is True at least one element in the iterable."""
   return True in itertools.imap(pred, seq)

def serve(status, body, mime=None): 
   mime = mime or 'text/html'
   sys.stdout.write("Status: %s\r\n" % status)
   sys.stdout.write("Content-Type: %s; charset=utf-8\r\n\r\n" % mime)
   sys.stdout.write(body)
   sys.exit()

def homepage(): 
   if os.path.exists(overview): 
      content = open(overview).read()
   else: serve(500, '<p>Error: no %s found!</p>\n' % overview)
   serve(200, content)

def main(args=None): 
   if args is None: 
      args = sys.argv[1:]

   path = os.environ.get('REQUEST_URI') or (base + '/')
   path = path[len(base):]

   # Directories to ignore--parse robots.txt
   robotstxt = os.path.join(site, 'robots.txt')
   if os.path.exists(robotstxt): 
      rp = robotparser.RobotFileParser()
      rp.set_url(robotstxt)
      try: 
         rp.read() # @@ this reads borked files too
         if not rp.can_fetch('*', path): 
            serve(403, "<p>Forbidden by robots.txt</p>\n")
      except IOError: serve(500, "<p>Couldn't parse robots.txt</p>\n")

   if path == '/': homepage()
   elif path == '/misc/feedback': 
      serve(404, "<p>Not found!</p>")
   else: path = os.path.join(site, path[1:])

   # If it's a directory, try to serve the directory index
   if os.path.isdir(path): 
      for fn in indexen: 
         filename = os.path.join(path, fn)
         if os.path.isfile(filename): 
            serve(200, open(filename).read(), 'text/plain')
            break # @@ not strictly necessary
      else: serve(404, "<p>Not Found (no CGI in directory)</p>\n")

   # Check that it's a .cgi or .py file
   if any(suffixen, lambda ext: path.endswith(ext)): 
      if os.path.isfile(path): 
         serve(200, open(path).read(), 'text/plain')
      else: serve(404, "<p>Not Found (%s: not a file)</p>\n" % path)

   # See if we can get a valid CGI by adding an extension
   for ext in suffixen: 
      if os.path.isfile(path + ext): 
         serve(200, open(path + ext).read(), 'text/plain')

   # See if we can get a valid CGI by trimming a trailing slash
   # @@ could hook this up with metagen to deliver a canonical URI
   if path.endswith('/'): 
      for ext in suffixen: 
         if os.path.isfile(path.rstrip('/') + ext): 
            serve(200, open(path.rstrip('/') + ext).read(), 'text/plain')

   # # Check to see if it's a .htaccess file
   # # @@ Better to allow them to be served normally?
   # if path.endswith('/htaccess'): 
   #    htaccess = path[:-8] + '.htaccess'
   #    if os.path.isfile(htaccess): 
   #       serve(200, open(htaccess).read(), 'text/plain')

   # Otherwise, we can't find anything we want to serve
   serve(404, "<p>Not Found</p>\n")

if __name__=="__main__": 
   main()