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 import sha
  17 import sys
  18 import os.path
  19 import urllib2
  20 import difflib
  21
  22 os.chdir(os.path.dirname(sys.argv[0]))
  23
  24 if os.path.exists('hooks.py'):
  25     from hooks import filter
  26 else:
  27     filter = lambda x, y: y
  28
  29 for url in (x for x in open('urls.txt').read().splitlines() if not (x.startswith('#') or x.strip()=='')):
  30     filename = sha.new(url).hexdigest()
  31     data = filter(url, urllib2.urlopen(url).read())
  32     if os.path.exists(filename):
  33         old_data = open(filename).read()
  34         diff = ''.join(difflib.unified_diff(old_data.splitlines(1), data.splitlines(1)))
  35         if len(diff) > 0:
  36             print '%s\nCHANGED: %s\n%s\n%s\n%s\n\n' % ('*'*60, url, '*'*60, diff, '*'*60)
  37     open(filename, 'w').write(data)
  38