#!/usr/bin/python
#
# File: RandomPageFromList.py
# Author: Angus McIntyre
# Date: 07.05.2001
# Updated: 07.05.2001
#
# Script to send the user off to a random URL, chosen from a list of URLs
# in a file. The file to use can be specified as an argument to the script.
#
# ---------------------------------------------------------------------------
# REVISION HISTORY
#
# 07.05.2001 SLAM First implementation.
#
# ---------------------------------------------------------------------------
# LEGAL NOTICE
#
# This script may be freely copied, distributed and modified. Use of the
# script is at the risk of the user. The script is presented "as-is" without
# any warranty, and the author is not liable for any loss or damages arising
# out of the use of or failure to use this script. This notice must appear
# in any modified copy of the script in which the name of the original
# author also appears.
# ---------------------------------------------------------------------------
# ---------------------------------------------------------------------------
# IMPORTS
# ---------------------------------------------------------------------------
import cgi
import re
import os
import sys
import traceback
import string
import random
# ---------------------------------------------------------------------------
# CONSTANTS
# ---------------------------------------------------------------------------
# default_url_file: path to default file containing a list of URLs
default_url_file = "RandomURLs.txt"
# path_separator: path separator for your operating system, e.g. '/'
# for UNIX, '\' for DOS/Windows, ':' for MacOS etc.
path_separator = "/"
# full_path_pattern: a pattern that recognises whether a pathname passed
# represents a complete pathname. For UNIX (as shown below) this is fairly
# simple. For MacOS and DOS/Windows you'll need to be more ingenious.
full_path_pattern = re.compile("^/")
# allowed_url_pattern: a pattern designed to check that the line returned
# as a URL *is* in fact a plausible URL. If it isn't, the script will die
# with an error message. This is partly a security feature, and partly a
# convenience feature. It's a security feature because it should help to
# stop malicious people from getting anything of interest from system
# files such as the password file; it's a convenience feature because we
# don't want to serve the user anything that isn't a valid URL anyway.
allowed_url_pattern = re.compile("^http://|^ftp://|^gopher://")
# allowed_path_pattern: a pattern designed to check that the final
# pathname to the file isn't something it shouldn't be. I set the pattern
# to match 'URLs.txt' and then make all my URL lists end with that string.
# An alternative approach would be to specify the path to your HTML
# directory so that, whatever happens, the script won't be allowed to get
# any file that isn't in your area.
allowed_path_pattern = re.compile("^.*URLs.txt$")
# listfile_base: if you want to specify the directory containing the
# list files, you can use this variable. Otherwise the files will be
# assumed to be relative to the directory containing the script to be
# executed. The string should end with a path separator.
listfile_base = "";
# ---------------------------------------------------------------------------
# GLOBALS
# ---------------------------------------------------------------------------
# urls: list of URLs recovered from a URL file
urls = []
# ---------------------------------------------------------------------------
# GLOBALS
# ---------------------------------------------------------------------------
# report_error
#
# A general-purpose function for printing out a slightly friendly error
# message as an HTML page.
def report_error(message):
print 'Content-type: text/html\n\n\
Random URL Error\
Random URL Error
\
No random URL could be returned because the error:
\
'
print message;
print '
occurred. Sorry.';
# read_url_file
#
# Given a filename (passed as an argument to the script), work out exactly
# which file we should be reading, and then read its contents into the
# global list 'urls'
def read_url_file(filename):
# See if the filename we've been passed is a complete filepath. If it
# is, we can just use it as is.
if full_path_pattern.match(filename):
filepath = filename
# If it isn't, but a base filepath has been defined, interpret it as
# a partial path relative to that base.
elif listfile_base:
filepath = listfile_base + filename
# Otherwise interpret it as a partial path relative to the location
# of the script file itself.
else:
script_pathname = os.environ['SCRIPT_FILENAME']
script_pathname_elements = string.split(script_pathname,path_separator)
script_pathname_elements.pop()
filepath = string.join(script_pathname_elements,path_separator) + path_separator + filename
# Now check to see if the path is a permissible path, using the
# predefined criteria for checking the names of files.
if allowed_path_pattern.match(filepath):
# If the file passes the test, open it up for reading, and read
# its contents line by line
handle = open(filepath,"r")
while 1:
# Read a line from the file
line = handle.readline()
# Exit the loop on end of file
if not line: break
# Any line in the file that looks like it might be
# a URL is added to the list of URLs
if allowed_url_pattern.match(line):
urls.append(string.strip(line))
handle.close
else:
raise "Illegal URL file", filepath
# perform_redirect
#
# Send a redirect instruction to the browser. The redirect is actually
# contained in the 'Location:' header, but for the benefit of older browsers
# that don't handle this properly, we include some HTML text to give the user
# the chance to go to the selected URL manually.
def perform_redirect(url):
print 'Content-type: text/html\nLocation: ' + url + '\n\n'
print 'Redirection\
Redirection
\
It appears that your browser cannot handle redirection \
automatically. Please proceed to the randomly-selected page \
by clicking here.
'
# ---------------------------------------------------------------------------
# MAIN ROUTINE
# ---------------------------------------------------------------------------
try:
# Try to get the name of the URL file to use from the arguments
# passed to the script.
try:
input = cgi.SvFormContentDict()
file = input['file']
# If that didn't work, fall back on a default filename
except:
file = default_url_file
# Read the file that contains the URLs. Throw an exception if no
# valid URLs could be found in the file. Otherwise, pick one at
# random and send the user to it.
read_url_file(file)
if len(urls) == 0:
raise "No valid URLs found in file", file
url = random.choice(urls)
perform_redirect(url)
# If anything went wrong, issue an error message
except:
type, value = sys.exc_info()[:2]
report_error(traceback.format_exception_only(type,value))