Match non-capturing group multiple times

I tried really hard to make a good title, but I’m not sure if I’m asking this right. Here’s my best attempt:

I’m using Python’s flavor of regex

I need to match numbers using named groups:

15x20x30    ->  'values': [15,20,30]
15bits      ->  'values': [15]
15          ->  'values': [15]
x15         ->  'values': [15]

but should not match:

456.48
888,12
6,4.8,4684.,6

my best attempt so far has been:

((?:[sa-z])(?P<values>d+)(?:[sa-z]))

I’m using [sa-z] instead of a word-boundary because 15x20 are two different values.

But it fails to match both 15 and 20 for the 15x20 case. It does work if I put an extra space as in 15x 20. How do I tell it to “reset” the non-capturing group at the end so it also works for the non-capturing group at the beginning?

Answer

You may use

(?<![^sa-z])d+(?![^sa-z])

Case insensitive version:

(?i)(?<![^sa-z])d+(?![^sa-z])

Or, compile the pattern with the re.I / re.IGNORECASE flags.

See the regex demo

Details

• (?<![^sa-z]) – a negative lookbehind that fails the match if, immediately to the left, there is no whitespace or a lowercase letter (any ASCII letter if (?i) or re.I are used)
• d+ – 1+ digits
• (?![^sa-z]) – a negative lookahead that fails the match if, immediately to the right, there is no whitespace or a lowercase letter (any ASCII letter if (?i) or re.I are used)