watch.py

   1 #!/usr/bin/python
   2 # Minimalistic Python URL watcher
   3 # 2008-03-04 Thomas Perl <thpinfo.com>
   4 # http://thpinfo.com/2008/urlwatch
   5
   6 # 1. Create an "urls.txt" file and add one URL per
   7 #    line that you want to watch.
   8 # 2. Add watch.py as a cronjob or run it manually.
   9 # 3. If something changed, you'll get a diff output
  10 #    to stdout. If nothing changed, no output.
  11 # 4. If you want to filter the web pages, because
  12 #    there is some dynamic content that _always_
  13 #    changes, create a "hooks.py" file that has a
  14 #    filter(url, data) -> filtered_data function
  15
  16 # Configuration section
  17 display_errors = False
  18
  19 # Code section
  20
  21 import sha
  22 import sys
  23 import os.path
  24 import urllib2
  25 import difflib
  26
  27 os.chdir(os.path.dirname(sys.argv[0]))
  28
  29 if os.path.exists('hooks.py'):
  30     from hooks import filter
  31 else:
  32     filter = lambda x, y: y
  33
  34 for url in (x for x in open('urls.txt').read().splitlines() if not (x.startswith('#') or x.strip()=='')):
  35     filename = sha.new(url).hexdigest()
  36     try:
  37         data = filter(url, urllib2.urlopen(url).read())
  38         if os.path.exists(filename):
  39             old_data = open(filename).read()
  40             diff = ''.join(difflib.unified_diff(old_data.splitlines(1), data.splitlines(1)))
  41             if len(diff) > 0:
  42                 print '%s\nCHANGED: %s\n%s\n%s\n%s\n\n' % ('*'*60, url, '*'*60, diff, '*'*60)
  43         else:
  44             print '%s\nNEW: %s\n%s\n\n' % ('*'*60, url, '*'*60)
  45         open(filename, 'w').write(data)
  46     except urllib2.HTTPError, error:
  47         if display_errors:
  48             print '%s\nERROR: %s\n%s\n%s\n%s\n\n' % ('*'*60, url, '*'*60, error, '*'*60)
  49     except urllib2.URLError, error:
  50         if display_errors:
  51             print '%s\nERROR: %s\n%s\n%s\n%s\n\n' % ('*'*60, url, '*'*60, error, '*'*60)
  52