Regex - to match number NOT in a sequence of character classes

Question

Sample file contents:

create 664 root utmp

I would like to match any number that is NOT this:

[0-6][0-4][0]

So it would match 664 and 641, but not 440 and 640.

My full regex is this:

^([\s]`*`?)create[\s]+?([0]?[0-6][0-6][0-6])[\s]+?(\w+?)[\s]+?(\w+?)[\s]`*`?$

But I would like to basically say:

match on not this:
```
([0]?[0-6][0-6][0-6])
```

Depending on the tool or language, you can use a negative lookahead `create(?!\s+[0-6][0-4]0)\s+(\d{3})\b` https://regex101.com/r/jalFbw/1 — The fourth bird, Mar 12 '23 at 10:04
Dupe: https://stackoverflow.com/questions/977251/regular-expressions-and-negating-a-whole-character-group — Nasser Kessas, Mar 12 '23 at 10:09
thanks "bird" and Nasser - helpful comments. Also helpful: https://docs.racket-lang.org/guide/Looking_Ahead_and_Behind.html — Dennis, Mar 15 '23 at 19:39

score 1 · Accepted Answer · answered Mar 12 '23 at 10:04

1

I looked at: How can I "inverse match" with regex?

and found: (?![0-6][0-4][0])[0-9]{3} see: https://regex101.com/r/D3JRfu/1

As far as I can see this works.

answered Mar 12 '23 at 10:04

Luuk

12,245
5
22
33

score 0 · Answer 2 · answered Mar 12 '23 at 18:35

Using the regex in your sample [0]?[0-6][0-6][0-6],
it can be done with a couple of negative assertions.
There has to be boundary assertions to force a constraint on the number of
digits as well as values.

(?<!\d)(?![0]?[0-6][0-6][0-6](?!\d))\d+

The first negative look behind can be replaced with a \s when its put in the larger regex.

 (?<! \d )
 (?!
    
    [0]? [0-6] [0-6] [0-6] 
    (?! \d )
    
 )
 \d+

Regex - to match number NOT in a sequence of character classes

2 Answers2