Hi,
Yuval Kogman wrote:
> On Sat, Aug 27, 2005 at 14:29:29 +0200, Ingo Blechschmidt wrote:
>> * @foo[$idx] := $var;
>> my @bar = @foo;
>> $var = $new_var;
>> # @foo[$idx] and $var are now $new_var, but @bar is unchanged,
>> # right?
>
> Yes, I agree. But we do need a way in the middle. Right now we have:
>
> @bar := @foo; # array container aliased, so all nested
> containers are shared
>
> @bar = @foo; # array structure duplicated, elem containers
> duplicated
>
> but no way to say all the elements are the same, but the structure
> isn't. Maybe this works, but I don't think so since assignment isn't
> an operator, but a syntactic construct (i think):
>
> @bar >>:=<< @foo;
I like this very much! :)
Additionally, if we allow hyperizing = and :=, we could get rid of the
special lists-as-lvalues magic:
# If you had previously written
($foo, $bar) = ($grtz, $baka);
# you'd now write
($foo, $bar) »=« ($grtz, $baka);
(But, FWIW, I kind of like the special lists-as-lvalues magic.)
--Ingo
--
Linux, the choice of a GNU | In Nature there are neither rewards nor
generation on a dual AMD | punishments, there are consequences.
Athlon! | -- R.G. Ingersoll
