11

I'm using re.findall() to extract some version numbers from an HTML file:

>>> import re
>>> text = "<table><td><a href=\"url\">Test0.2.1.zip</a></td><td>Test0.2.1</td></table> Test0.2.1"
>>> re.findall("Test([\.0-9]*)", text)
['0.2.1.', '0.2.1', '0.2.1']

but I would like to only get the ones that do not end in a dot. The filename might not always be .zip so I can't just stick .zip in the regex.

I wanna end up with:

['0.2.1', '0.2.1']

Can anyone suggest a better regex to use? :)

Ashy
  • 2,014
  • 5
  • 21
  • 25

1 Answers1

27
re.findall(r"Test([0-9.]*[0-9]+)", text)

or, a bit shorter:

re.findall(r"Test([\d.]*\d+)", text)

By the way - you do not need to escape the dot in a character class. Inside [] the . has no special meaning, it just matches a literal dot. Escaping it has no effect.

Tomalak
  • 332,285
  • 67
  • 532
  • 628
  • It should probably be \d+ if numbers can be greater than 9 –  Dec 10 '08 at 15:46
  • 1
    It should be r"Test([\d.]*\d+)" -- \d doesn't mean anything in the string so it works, but generally it's good practice not to rely on that. You could do r"Test(\d*(?:\.\d+))" if you want to be slightly more restrictive (rejecting 1..2, for instance) – ianb Dec 11 '08 at 00:31