From mikeflan at att.net  Wed Jun 17 16:04:10 2020
From: mikeflan at att.net (Mike Flannigan)
Date: Wed, 17 Jun 2020 18:04:10 -0500
Subject: [pm-h] Perl Code
References: <fae1d71a-5616-4ed0-6f56-67d245047173.ref@att.net>
Message-ID: <fae1d71a-5616-4ed0-6f56-67d245047173@att.net>


Recently I found this bit of code that I wrote
long ago.? I am a little surprised this works
($1 and $2 get transferred to the scalars).
I don't write that way anymore, but I am
leaving it that way in this program.


use strict;
use warnings;

my $test = 'abcdefgh';
my ($latm, $longm) = $test =~ /(ab).*(fg)/i;
print "$latm, $longm\n";


Prints this:
ab, fg



Mike

From uri at stemsystems.com  Wed Jun 17 23:13:00 2020
From: uri at stemsystems.com (Uri Guttman)
Date: Thu, 18 Jun 2020 02:13:00 -0400
Subject: [pm-h] Perl Code
In-Reply-To: <fae1d71a-5616-4ed0-6f56-67d245047173@att.net>
References: <fae1d71a-5616-4ed0-6f56-67d245047173.ref@att.net>
 <fae1d71a-5616-4ed0-6f56-67d245047173@att.net>
Message-ID: <0adb5f2f-bd00-73da-4001-d7b182dfbad5@stemsystems.com>

On 6/17/20 7:04 PM, Mike Flannigan wrote:
>
> Recently I found this bit of code that I wrote
> long ago.? I am a little surprised this works
> ($1 and $2 get transferred to the scalars).
> I don't write that way anymore, but I am
> leaving it that way in this program.
>
>
> use strict;
> use warnings;
>
> my $test = 'abcdefgh';
> my ($latm, $longm) = $test =~ /(ab).*(fg)/i;
> print "$latm, $longm\n";
>
>
> Prints this:
> ab, fg
>
that is perfectly normal perl behavior. grabs in a regex return those 
grabs when in a list context. the two scalars will get the grabbed 
strings. that is a very common style and i am not sure why you would 
want to change it. the only comment i could make is that checking if a 
regex worked is usually important too. but you could check the 
definedness of one of the scalars for that.

uri

From jellyson at gmail.com  Thu Jun 18 22:19:19 2020
From: jellyson at gmail.com (John Ellyson)
Date: Fri, 19 Jun 2020 00:19:19 -0500
Subject: [pm-h] Perl Code
In-Reply-To: <fae1d71a-5616-4ed0-6f56-67d245047173@att.net>
References: <fae1d71a-5616-4ed0-6f56-67d245047173.ref@att.net>
 <fae1d71a-5616-4ed0-6f56-67d245047173@att.net>
Message-ID: <CAHe5CJsiOEiQcZ4GK=EfUHs7P2SPh8+OsHP3nqFuM7K8vLJg0w@mail.gmail.com>

Mike,

The behavior that you are describing is covered in Perl regular expressions
tutorial at
https://metacpan.org/pod/distribution/perl/pod/perlretut.pod#Extracting-matches
.

I personally leverage this a lot in the code that I write - especially to
capture all occurrences of something into an array.  In fact, I have gotten
into the habit of trying to write my code in such a way to try to avoid
using the regex variables like $1 and $2.

And Uri has a very good point about needing to include some way to verify
that the regex actually matched and returned values.  Not having that has
caused me headaches with trying to figure out what went wrong with my code.

John Ellyson

On Wed, Jun 17, 2020 at 6:04 PM Mike Flannigan <mikeflan at att.net> wrote:

>
> Recently I found this bit of code that I wrote
> long ago.  I am a little surprised this works
> ($1 and $2 get transferred to the scalars).
> I don't write that way anymore, but I am
> leaving it that way in this program.
>
>
> use strict;
> use warnings;
>
> my $test = 'abcdefgh';
> my ($latm, $longm) = $test =~ /(ab).*(fg)/i;
> print "$latm, $longm\n";
>
>
> Prints this:
> ab, fg
>
>
>
> Mike
> _______________________________________________
> Houston mailing list
> Houston at pm.org
> https://mail.pm.org/mailman/listinfo/houston
> Website: http://houston.pm.org/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/houston/attachments/20200619/ac3b9dc5/attachment.html>

From mrdvt92 at yahoo.com  Fri Jun 19 15:34:43 2020
From: mrdvt92 at yahoo.com (Michael R. Davis)
Date: Fri, 19 Jun 2020 22:34:43 +0000 (UTC)
Subject: [pm-h] Perl Code
In-Reply-To: <CAHe5CJsiOEiQcZ4GK=EfUHs7P2SPh8+OsHP3nqFuM7K8vLJg0w@mail.gmail.com>
References: <fae1d71a-5616-4ed0-6f56-67d245047173.ref@att.net>
 <fae1d71a-5616-4ed0-6f56-67d245047173@att.net>
 <CAHe5CJsiOEiQcZ4GK=EfUHs7P2SPh8+OsHP3nqFuM7K8vLJg0w@mail.gmail.com>
Message-ID: <1746711845.789281.1592606083994@mail.yahoo.com>

PM,>And Uri has a very good point about needing to include some way to verify that the regex actually matched and returned values.
I always? - put the regexp in a condition? - use the optional "m"? - use m{} if the regexp has a "/" in it.? - use \A for ^? - use \Z for $? - use?(?:)? when group not needed in $N
? - and assign $N to well named scoped variables before anything else.
if ($string =~ m/\A(.)(?:.)(.)\Z/) { ? my $char1 = $1;? my $char3 = $2;
? dosomething($char1, $char3);} else {
? die("handle string not what expected");}
Michael R. Davis 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/houston/attachments/20200619/12254471/attachment.html>