perl search and replace

LeukvoorJ has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: perl search and replace by japhy (Canon) on Jun 06, 2006 at 16:33 UTC
Yes it can, but your s/// fails on both accounts: it doesn't match duplicated letters, and it doesn't keep track of what character that was. Here is code that matches duplicated letters and replaces the second one with a '2': `$string =~ s/([a-zA-Z])\1/${1}2/g;` [download] Note these things: `[a-zA-Z]` matches a lowercase or uppercase English letter `([a-zA-Z])` captures that letter to the backreference \1 (and later, to $1) `\1` matches whatever is stored in backreference \1 (the specific letter matched) `${1}2` is the variable $1 (how you retrieve backreference \1 outside a regex) followed by the number 2 -- the reason I have $1 in braces (${1}) is so that Perl doesn't see $12 as the scalar variable $12 If you don't care about case, you can use the /i modifier, like so: `$string =~ s/([a-z])\1/${1}2/ig;` [download] Please read perlretut for more. Jeff `japhy` Pinyan, P.L., P.M., P.O.D, X.S.: Perl, regex, and `perl` hacker How can we ever be the sold short or the cheated, we who for every service have long ago been overpaid? ~~ Meister Eckhart	[reply] [d/l] [select]
Re: perl search and replace by thundergnat (Deacon) on Jun 06, 2006 at 17:13 UTC
You can generalize it for any number of repeated characters too. `while ( my $string = <DATA> ) { $string =~ s/(.)(\1+)/$1.length($1.$2)/ge; print $string; } __DATA__ Balloon bookkeeper Wheeeee!!!` [download] Output: `Bal2o2n bo2k2e2per Whe5!3` [download]	[reply] [d/l] [select]
Re: perl search and replace by vkon (Curate) on Jun 06, 2006 at 16:36 UTC
`$string =~ s/(.)\1/${1}2/gi;` [download]	[reply] [d/l]