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

I need to find two consecutive occurrences of a given character. The following SSCCE works ('cc' matches, 'cz' does not match), but I have to turn off 'uninitialized' warnings prior to the regex to make it work without getting 'uninitialized' warning messages about $1. Do you know a better way to do this, preferrably without the need to suppress warnings?

On a side note, the "start of string" anchor (^) in the regex is fine, but if I add an "end of string" anchor to the regex (either '$' or '\z') the match fails. The match should not fail with the addition of an "end of string" anchor, because each string is exactly 2 characters long.

Where is Jeffrey Friedl when you need him? (ha ha) :-) 

#!/usr/bin/perl
use strict;
use warnings;

my @txt = ('cc','cz');
for (@txt) {
    no warnings 'uninitialized';
    if (m/^(.)$1/) {
        print "True   \$_ = |$_|   \$1 = |$1|\n";
    } else {
        print "False  \$_ = |$_|   \$1 = |$1|\n";
    }
    use warnings;
}

[download]

"It's not how hard you work, it's how much you get done."

Replies are listed 'Best First'.
Re: How To Remove Warnings From Regex
by Athanasius (Archbishop) on Feb 04, 2024 at 04:22 UTC

    Hello roho,

    Change $1 to \1 within the regex:

    use strict;
use warnings;

my @txt = ('cc', 'cz');

for (@txt)
{
    if (m/^(.)\1$/)
    {
        print "True   \$_ = |$_|   \$1 = |$1|\n";
    }
    else
    {
        print "False  \$_ = |$_|   \$1 = |$1|\n";
    }
}

    [download]

    Output:

    14:13 >perl 2092.SoPW.pl
True   $_ = |cc|   $1 = |c|
False  $_ = |cz|   $1 = |c|

14:14 >

    [download]

    As you can see:

    • warnings are gone
    • the closing $ is OK now
    • $1 is no longer the empty string on match fail, but is rather the captured character “c”.

    See e.g. Capture groups.

    Hope that helps,

    Athanasius <°(((>< contra mundum סתם עוד האקר של פרל,

[reply]
[d/l]
[select]
      Thank you Athanasius! That works perfectly. For as long as I can remember, I've avoided the backslash numbers in favor of the dollar-sign numbers, but I see in this case that is exactly what is called for. Thanks again.

      "It's not how hard you work, it's how much you get done."

[reply]
        \1 works inside the regex where the capture group was defined. $1 works everywhere else, i.e. in a replacement of a substitution. $1 in a regex refers to a previously matched regex (which explains the uninitialized warnings).

        map{substr$_->[0],$_->[1]||0,1}[\*||{},3],[[]],[ref qr-1,-,-1],[{}],[sub{}^*ARGV,3]
[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom