"my" declaration problem

[Hosen1989]

Dear ALL,

I was debug some old dirty script of mine (which without strict nor warnings ^_^), and faced this issue:

As you can see in case(1), variable $me had been declared more than once, and to output of this script is: []

#----------[ CASE(1) ]----------#
my $v_place = 'home';
my $me = 'moving';
my $me = 'at home' if($v_place =~ m/^home/);
my $me = 'at work' if($v_place =~ m/^work/);

print "[$me]";
__END__
output:
[]

#-------------------------------#

But after correct the declaration problem as in case(2), we got the correct output: [at home].

#----------[ CASE(2) ]----------#
my $v_place = 'home';
my $me = 'moving';
$me = 'at home' if($v_place =~ m/^home/);
$me = 'at work' if($v_place =~ m/^work/);

print "[$me]";
__END__
output:
[at home]
#-------------------------------#

Now,can any monk explain me just what happen here?

BR

Hosen

[Eily]

The note on statement modifiers in the doc states:

NOTE: The behaviour of a my, state, or our modified with a statement modifier conditional or loop construct (for example, my $x if ... ) is undefined. The value of the my variable may be undef, any previously assigned value, or possibly anything else. Don't rely on it. Future versions of perl might do something different from the version of perl you try it out on. Here be dragons.

So even if there only was the last of the three declarations, it would still be wrong.

[davies]

Warnings would tell you that your repeated "my" masked a previous declaration. When you use "my", a new variable is created. If several of the same name are created in the same scope, only the last to be declared is accessible. Therefore, $me is "at work" if $v_place is "work" (simplified, I know), which it isn't. Therefore $me exists but has no value, which is what you are seeing.

Regards,

John Davies

[talexb]

As has already been pointed out, just get into the habit of always using strict and warnings on all your code. It lets Perl remind you about anything that's a bit dodgy.

Second, it's a bad idea to mix a declaration with anything more than a straight initialization. To me, that has a bad 'code smell'. And Perl also thinks it's a bad idea.

[karlgoethebier]

"...explain..."

#!/usr/bin/env perl
use strict;
use warnings;
use feature qw(say);

my $immortal = q(Connor MacLeod);
my $immortal = q(The Kurgan);

say qq("There can be only one");

__END__

karls-mac-mini:monks karl$ ./my.pl

"my" variable $immortal masks earlier declaration in same scope at ./m
+y.pl line 8.
"There can be only one"

Regards, Karl

[Anonymous Monk]

Turn on warnings and read the warning generated

[LanX]

> Now,can any monk explain me just what happen here?

because ...

> without strict nor warnings ^_^

Or differently phrased:

Repeated declarations of the same variable in the same scope make no sense.

[Anonymous Monk]

Repeated declarations of the same variable in the same scope make no sense.

In bad style and obfuscated, sure, but is it nonsensical? Redeclaration and reassignment are not the same when destructors or side-effects are involved. I can imagine this kind of situation arising with "templated" code where sections are copy-paste or somesuch. Job security FTW!

[LanX]

OK ... Please provide an example where the code is not broken.