How to match same length digits in a string using regex - regex

Question

I want to find a text (for example: stack) in a String that contains digits and chars (for example: s123t123a123c123k). The only rule is that between every character of the search key there should be the same amount of digits, so all of this should match:
search key: stack
Strings:
stack //0 digit between chars of stack
s7t3a9c0k //1 digit between chars of stack
s27t33a49c50k //2 digit between chars of stack
s127t312a229c330k //3 digit between chars of stack and so on for 4,5,6 digits...
If I could match same length digits then I can write something like: s[]*t[]*a[]*c[]*k if the regex for same length digit is [].
How to match same length digits in a string using regex?

Answer 1

In case you want to do that with Perl, you may use
m/s(\d+)t(??{ "\\d{".length($^N)."}" })a(??{ "\\d{".length($^N)."}" })c(??{ "\\d{".length($^N)."}" })k/
Or, if you want to match digits with [0-9]:
m/s([0-9]+)t(??{ "[0-9]{".length($^N)."}" })a(??{ "[0-9]{".length($^N)."}" })c(??{ "[0-9]{".length($^N)."}" })k/
In the Perl code, you will most likely want to build the pattern dynamically. See the full Perl code demo:
#!/usr/bin/perl
use warnings;
use strict;
use re 'eval'; # stackoverflow.com/a/16320570/3832970
my #input = split /\n/, <<"END";
s7t3a9c0k
s27t33a49c50k
s127t312a229c330k
s1t312a22c3300k
END
my $keyword = "stack";
my $pattern = substr($keyword, 0, 1) . '([0-9]+)' . join( '(??{ "[0-9]{".length($^N)."}" })', split("", substr($keyword, 1)) );
#my $pattern = substr($keyword, 0, 1) . '(\d+)' . join( '(??{ "\\\\d{".length($^N)."}" })', split("", substr($keyword, 1)) );
for my $input ( #input ) {
if ($input =~ m/$pattern/) {
print $input . ": PASS!\n";
} else {
print $input . ": FAIL!\n"
}
}
Output:
s7t3a9c0k: PASS!
s27t33a49c50k: PASS!
s127t312a229c330k: PASS!
s1t312a22c3300k: FAIL!
The $pattern is built dynamically: substr($keyword, 0, 1) gets the first char, then ([0-9]+) is added, then join( '(??{ "[0-9]{".length($^N)."}" })', split("", substr($keyword, 1)) adds the following: it inserts (??{ "[0-9]{".length($^N)."}" }) in between each char of the $keyword substring from the second char. The (??{ "[0-9]{".length($^N)."}" }) part acts as a \d{X} pattern where X is the length of the most recent captured substring (it was ([0-9]+)).
The use re 'eval'; is necessary to build the pattern dynamically. As per this answer, it will only affect the regular expressions in the file or in the curlies where it is used.