On Dec 10, 2014, at 12:30 PM, Robert Stoll <p...@tutteli.ch> wrote:
> I stick with it, evaluating it does not make sense IMO. If I want to execute
> it in any case then I would do something
> like this currently
>
> $g = g();
> $h = h();
> if($x !== null){
> $x->foo($g, $h)->bar(baz());
> }
>
> And with the ?-> operator:
>
> $g = g();
> $h = h();
> $x?->foo($g, $h)?->bar(baz());
>
> Where I do not expect that baz() is called.
You can make exactly the opposite argument too, though. I *do* expect that
baz() would be called, and I can always explicitly write it the other way if
that happens to not be what I want at some point.
> You wrote it will be confusing for the user if the arguments would not be
> evaluated. I do not think so, telling them it is a short form for ($x !==
> null ? $x->foo() : null) is very easy and
> straight forward for the user to pick up.
I think it's confusing the other way, and "instead of raising a fatal for
calling a method on null, it returns null" is just as simple of an explanation.
> Therefore, please dig up a better example.
Yeah, gonna go do this now, had a busy morning :)
All of that said: it's not clear to me that this actually matters that much. I
wasn't able to quickly find an example because the vast majority of the Hack
code I have that uses this operator isn't actually affected one way or the
other.
Josh
--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
