I hear everybody constantly chanting the use strict and use warnings
mantra in this and every other perl group. No argument there, but what's
the position on prototypes in subs? I prefer to use them but that's
probably because I was taught Pascal way back when and now I
compulsively want to type all my variables and predeclare my
subroutines, unlike my C trained workmates who put their main() at the
top of the code.
One nagging thing with prototypes is when I use them recursively
warnings carps about not being able to test the prototype so I end up
wrapping it in a no warnings (example below). Am I being too pedantic
about my subs and should I just trust myself not to call the subs
incorrectly?
Steve
sub descend($$){ #map out the
endpoints of an unknown record structure
my ($name,$ref) = @_; #accumulated
name of var and scalar or ref to the next level
if(ref($ref) eq ''){ #map scalar
endpoints
print $name.": [$ref]\n";
}
if(ref($ref) eq 'HASH'){ #explore the
keys of a hash
print "$name: [undef]\n" unless(scalar(keys %$ref));
for my $element (keys %$ref){
no warnings; #disable
warning for recursion
descend($name.".$element",$ref->{$element});
}
}
if(ref($ref) eq 'ARRAY'){ #enumerate
the elements of an array
print "$name: [undef]\n" unless(scalar(@$ref));
my $index = 0;
for my $element (@$ref){
no warnings; #disable
warning for recursion
descend($name.".[$index]",$element);
$index ++;
}
}
}
_______________________________________________
ActivePerl mailing list
[email protected]
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
