<yours>
However, since the whole idea behind using local or my with variables
is to prevent a variable name from writing over the same variable
name elsewhere in the script, it makes even *more* sense to me to
use local with Perl's special variables; in that they are used frequently
throughout one's script, (with different values/purposes).
</yours>
If that's your goal in using local variables, be very careful with 'local'.
You probably want to use 'my' instead. Try this as an example of why:
$var = 42;
dosomething($var);
sub dosomething {
local ($var) = @_;
&dosomethingelse;
print "\$var is now $var\n";
}
sub dosomethingelse {
$var++;
}
Using local, you will print 43. If you change local to my, you will get 42.
local makes a variable visible to all subroutines called from the scope
where the variable was 'local' ized. 'my' does not behave like that.
I will say, I use local variables for convenience, and for making my
subroutines reusable much more often than for not altering a global
variable. The why is probably not a beginners topic, but suffice it to say
if I pass in an array reference of parameters to use in a SQL query, and a
global hash reference, I probably have every intention of dereferencing that
array ref and changing data in a global scope, then I can call the same
subroutine with a different array ref, and a different hash ref, and get
another hash populated entirely differently.
The last paragraph like I said, is probably not something for the beginners
board, but my point is that there is more than one reason to localize.
Steve H.
-----Original Message-----
From: Dave Benware [mailto:[EMAIL PROTECTED]]
Sent: Monday, February 04, 2002 9:00 PM
To: Beginners perl
Subject: Why localize $_, was-> Re: why shift @_ ?
Timothy Johnson wrote:
>
> Ok, I figured that much, but I guess my question is this: Is there a
> pressing need to scope a predefined variable like $_? How could this
> adversely affect a program? (I'm not trying to be a smart aleck, I really
> want to know)
>
Hi,
Sorry for such a short answer in my first post, but actually,
scoping is brand new to me, so I was shy of going into a lot
of details on the matter.
However, since the whole idea behind using local or my with variables
is to prevent a variable name from writing over the same variable
name elsewhere in the script, it makes even *more* sense to me to
use local with Perl's special variables; in that they are used frequently
throughout one's script, (with different values/purposes).
If $_ were a constant value like pi throughout a script, that would be
different. Then there would be no purpose in scoping it.
Since I couldn't understand Jeff's code, (no offense, I'm new), I
wrote a script to demonstrate how the value of $_ could get changed
unexpectedly. (It took me an hour, but I learned from it, heh).
# THE MAIN PROGRAM PRINTS AN ARRAY. HOWEVER IT ALSO CALLS
# A SUBROUTINE THAT CHECKS EACH ELEMENT FOR "b". IF "b" IS
# FOUND, IT SAVES IT IN A FILE. THE SUBROUTINE SHOULD BE
# HARMLESS, RIGHT?
use strict;
our @list = qw(a b c d);
foreach (@list) {
&check_b;
print $_, "\n";
}
sub check_b {
foreach (@list) {
#local $_;
$_ =~ s/b/bb/;
# SAVE TO FILE
}
}
As is, @list is printed incorrectly, but if you uncomment #local $_;,
it'll print as expected.
Bompa
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
