Re: Regex curly bracket ambiguity

To indicate where a variable ends in a regex, you surround the variable with {}: $string =~ m/${match}{$repeat}/;.

That keeps it from thinking you mean %match, but gives you m/Hello{2}/ (match "Hell", followed by two "o"s). To make the repeat apply to the whole $match, you need to group it: m/(?:${match}){$repeat}/, which makes the added {} unnecessary, leaving you with m/(?:$match){$repeat}/.

Comment on Re: Regex curly bracket ambiguity Select or Download Code

Replies are listed 'Best First'.
Re^2: Regex curly bracket ambiguity by Anonymous Monk on Jun 21, 2007 at 18:08 UTC
Thank you that was incredibly helpful. so now that I have regular brackets around (?:$match); what do I do if I want to capture the whole thing? eg with: m/((?:$match){$repeat})/ will $1 give me everything or just $match back?	[reply]
Re^3: Regex curly bracket ambiguity by Roy Johnson (Monsignor) on Jun 21, 2007 at 18:11 UTC
`$1` will give you the whole match. `(?:)` is a non-capturing grouping, so it does not affect `$1`, etc. If you used regular parens around `$match`, it would be returned in `$2`. Caution: Contents may have been coded under pressure.	[reply] [d/l] [select]
Re^4: Regex curly bracket ambiguity by ysth (Canon) on Jun 21, 2007 at 18:23 UTC
If you used regular parens around $match, it would be returned in $2. That is, the last repetition would be returned in $2: `$ perl -w $match = "ab+"; $string = "fooababbabbbar"; $string =~ /(($match){3})/ && print "\$1: $1 \$2: $2\n"; __END__ $1: ababbabbb $2: abbb` [download]	[reply] [d/l]