"John W. Krahn" <[EMAIL PROTECTED]> wrote in message
[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
My thoughts on perl references. This code:
> foreach $file(@files) {
>
> open file...
> @$file = <FILEHANDLE>;
> close file...
> }
creates named arrays via symbolic references.
[trwww@devel_rh trwww]$ perl
$symRef = "array";
@$symRef = (1,2,3);
print( join( "\n", @array ), "\n" );
Ctrl-D
1
2
3
The array has a name, we just accessed it at runtime instead of hardcoding
the name in the program.
This is illegal with strict, for good reason.
for instance, in the foreach $file( @files ) { ... code above, imagine what
could happen if one of the files in the directory had the same name as an
array in the program!!! If there was a file named 'files' in the directory,
the program would probably crash!!! Why do that to your self when you dont
need to?
see: http://perl.plover.com/varvarname.html
> > On Fri, Jan 03, 2003 at 03:33:30PM -0800, John W. Krahn wrote:
> > > Paul Kraus wrote:
> > > >
> > > > Ok a couple questions on Ref from pg 251 programming Perl.
> > > >
> > > > push @$arrrayref,$filename);
> > > > $$arrayref[0]="January";
> > > > @$arrayref[4..6]=qw/May June July/;
> > > >
> > > > So this is actually creating an anonymous array that it then
references
> > > > correct?
> > > > so the assignments January ect are being made to an anonymous array.
> > >
> > > No, an anonymous array is delimited by [ and ]. That is creating an
> > > actual array that is only accessible through an array reference.
Anonymous arrays are sometimes created with [ and ]. The above snippet of
code is dereferencing the array so it can access it.
If $arrayref stores a reference ( which stringified looks like
ARRAY(0x80f560c) ) and the array was never given a name, then it is an
anonymous array... if it stores a string, it is a plain 'ol array being
accessed symbollically.
Another way to create an anonymous data structure is by returning a
reference ( using \ ) of a lexically scoped data structure from a
subroutine. This and autovivification are probably two of the coolest
features of perl and are why one can write 10 lines of perl code that would
usually take about 100 in other languages.
This cant be done in C++:
[trwww@devel_rh trwww]$ perl
use strict;
my($var) = getRef();
sub getRef {
my(@var) = (1,2,3);
return( \@var );
}
print $var->[1], "\n";
Ctrl-D
2
you would have to call new(), and then remember to delete() it.
anyways,
> > > my $ref = [ 1, 2, 3, 4, 5 ];
> > > ^^^^^^^^^^^^^^^^^
> > > anonymous array
creates an anonymous array... it has no name.
> >
> > Not worth picking a fight over, to be sure, but I don't see the
> > difference. Assuming $arrayref didn't exist before, then after each of
> > those three examples it will, and it will be a reference to an array
> > with no name. Seems to me that Paul got it spot on.
>
> You are saying that it has no name but it does: arrayref. If it truly
> had "no name" then there would be no way to access it anywhere else in
> the program.
>
The referent that the contents of $arrayref refers to has no name. There is
no place in the program we can access the contents of the memory, at least
directly. We have to dereference the reference stored in $arrayref to access
it.
Todd W.
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
