Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:112960
MIME-Version: 1.0
References: <766fb62e-9413-4472-89be-6557a45c64c8@Spark> <19bd9792-b966-4b8e-af24-a8e4399f7a2f@www.fastmail.com>
In-Reply-To: <19bd9792-b966-4b8e-af24-a8e4399f7a2f@www.fastmail.com>
Date: Thu, 21 Jan 2021 16:37:53 -0600
Message-ID: <CAESVnVrNOw4jFhV4Ebr6tn=+XsEiVb=9W=KupWFmGnoHSz9NxQ@mail.gmail.com>
To: Larry Garfield <larry@garfieldtech.com>
Cc: php internals <internals@lists.php.net>
Content-Type: multipart/alternative; boundary="000000000000889f1c05b970b9e8"
Subject: Re: [PHP-DEV] [RFC] Null Coalescing Assignment in Method Signature
From: pollita@php.net (Sara Golemon)

--000000000000889f1c05b970b9e8
Content-Type: text/plain; charset="UTF-8"

On Thu, Jan 21, 2021 at 3:29 PM Larry Garfield <larry@garfieldtech.com>
wrote:
>
> I'm unclear why you'd allow null at all then.
>  If you want $bar to be optional, and to be an empty array if not
specified, then just do:
>
> function foo(array $bar = []) { ... }
>
> At that point, the only thing adding ?array does is allow you to
explicitly pass null,
> presumably because it has some meaning to your function.
> If you don't want that, don't allow it.
>

Smells a little like it's verging on the `default` proposal that was
brought up awhile ago...

function foo(int $a, array $b = [], string $c = '') { ... }
foo(123, default, "bar");

In this case, foo() never wants `null` as a valid value, but neither does
the caller actually want anything different from the default.

Allowing a null-coalescish sort of initializer is another potential way to
solve this problem, and I'm not here to say I endorse any of them, but
maybe that's the intent.

-Sara

--000000000000889f1c05b970b9e8--