Is delete idempotent?

Yes for hash elements. It does not even appear to reset the hash's iterator (which I find surprising).

$ perl -E'%h=(a=>1,b=>1,c=>1); while (my ($k) = each(%h)) { say $k; }'
c
a
b

$ perl -E'%h=(a=>1,b=>1,c=>1); while (my ($k) = each(%h)) { say $k; de
+lete $h{d}; }'
c
a
b
[download]

Yes for array elements. It doesn't extend the array up to the element to delete.

$ perl -E'my @a = (4,5); say 0+@a; delete $a[10]; say 0+@a;'
2
2
[download]

(Using delete on array element is documented as being deprecated.)

if I try to delete something that either never existed

Not a problem.

$ perl -E'my %h; delete $h{x}; say "ok";'
ok
[download]

or has already been deleted

Not a problem.

$ perl -E'my %h = (x=>1); delete $h{x}; delete $h{x}; say "ok";'
ok
[download]

It does not even appear to reset the hash's iterator (which I find surprising).

It's not documented in delete, but it is in each:

If you add or delete a hash's elements while iterating over it, entries may be skipped or duplicated--so don't do that. Exception: It is always safe to delete the item most recently returned by each(), so the following code works properly:

while (($key, $value) = each %hash) {
    print $key, "\n";
    delete $hash{$key}; # This is safe
}
[download]

It probably wouldn't hurt if the POD for delete mentioned this as well.




Dave

perl -e 'use Data::Dumper; $x = {a => "B"}; print Dumper $x; delete $x
+->{y}{2}; print Dumper $x;'
$VAR1 = {
          'a' => 'B'
        };
$VAR1 = {
          'y' => {},
          'a' => 'B'
        };
[download]

++

That one has bit me a time or three.   It’s not wrong, per se, but maybe a bit unexpected.

There is no "delete" op or function in Perl. Could you be more specific? I can think of a dozen things you could mean.

uh, delete. Nevermind!