At 09:29 AM 5/2/2001 -0700, Larry Wall wrote:
>H.Merijn Brand writes:
>: wanting the side effects of grep/map finishing over all elements (which
>could
>: of course be from a tied array or database connection)
>
>If we do that kind of optimization, then we would certainly provide
>some kind of easy syntax to defeat it.
There are a lot of side-effect things that I'd personally love to be able
to optimize out, and this is one of them.
Generally speaking, I think that operations on or involving a variable
should be up for grabs to the optimizer unless there are some lexically
visible hints that say otherwise. Which means that this:
sub foo {
my @bar;
@bar = grep {/foo/} @_;
return;
}
foo(bar());
would get chewed up by the optimizer such that the foo sub became a noop,
and the resulting code was the same as:
bar();
and that this:
foo(@bar, $baz);
would be a complete noop unless @bar or $baz had some sort of attribute on
them that forced their evaluation.
Same thing with:
foreach (1..100) {
$foo = $_;
}
should get turned into:
$foo = 100;
barring any explicit reason that it shouldn't. (Like the compiler sees that
$foo has the no-short-circuit attribute or something)
I'd really like to get into the details of what is and isn't valid for the
optimizer to do, though I expect it's still a little early in the
Apocalypse season for that.
Dan
--------------------------------------"it's like this"-------------------
Dan Sugalski even samurai
[EMAIL PROTECTED] have teddy bears and even
teddy bears get drunk
