in reply to remove repeating characters from strings

If you just want to remove them, look at the /s flag to the transliteration operator (tr//). If you want to replace them, use a regular expression substitution with backreferences. It may be something like: s/(\w)\1+/$1\* /g;. That's untested, but it should get you on your way.

Replies are listed 'Best First'.
Re: remove repeating characters from strings
by Abigail-II (Bishop) on Jul 15, 2002 at 10:02 UTC
    The problem with that approach is that from the string ccabcabcab, it will group the first cc and only the second and third cab, while the question indicated the longest match should be favourited, not the left most.

    Here's a solution that favours the longest match, finds recursive repeatitions, and uses {$n} instead of *. 

        $_ = "bbbccabcabcabcdba";
    my $l = length;
    s{((\w{$l})\2+)}{($2){@{[length($1)/length($2)]}}}g while -- $l;

    [download]
    This will produce: 
        (b){3}c(cab){3}cdba

    [download]
    Changing $_ to bbbccaabcaabcaabcdba, the above code will yield: 
        (b){3}c(c(a){2}b){3}cdba

    [download]

    Abigail

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom