Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:111890
X-MailCleaner-return_path: brendt@stitcher.io
X-MailCleaner-sender_address: brendt@stitcher.io
X-MailCleaner-recipients: tovilo.ilija@gmail.com, internals@lists.php.net
X-MailCleaner-SPF: pass
Content-Type: text/plain;
	charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.80.23.2.2\))
In-Reply-To: <CAPyj-LDLGeo7NHBqbVeA4CXTRhhNMS4VQa-oQxrK-JFCQT3DPQ@mail.gmail.com>
Date: Thu, 17 Sep 2020 19:30:45 +0200
Cc: PHP internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
Message-ID: <736887E5-1015-4909-9087-274E69C75960@stitcher.io>
References: <09DE673B-71B7-4C8E-A795-900B51F8E476@stitcher.io>
 <CAPyj-LDLGeo7NHBqbVeA4CXTRhhNMS4VQa-oQxrK-JFCQT3DPQ@mail.gmail.com>
To: Ilija Tovilo <tovilo.ilija@gmail.com>
X-Mailer: Apple Mail (2.3608.80.23.2.2)
X-MailCleaner-TrustedIPs: Ok
Subject: Re: [PHP-DEV] The case for transpiled generics
From: brendt@stitcher.io (Brent Roose)

Hi Ilija

> On 17 Sep 2020, at 19:01, Ilija Tovilo <tovilo.ilija@gmail.com> wrote:
>=20
> Hi Brent
>=20
>> Today I'd like to hear your thoughts on what might be a controversial =
topic, though I think it's worth having this discussion. I want to make =
the case for adding generic syntax, without actually enforing any =
additional type checks at runtime. Please hear me out.
>=20
> I think having the option to omit runtime type checks could be
> valuable even for non-generic code. If your own internal classes all
> use static analysis the chance of passing a wrong type is pretty low.
> This would probably have to be on a per class or even per function
> basis as you can't reason about other people's code.

I agree, if there's enough interest in this, we might need to consider =
the scope of one or several RFCs.

>=20
> I do think we need some form of generic runtime check. If we don't,
> how generics work is completely open to interpretation. Static
> analysis tools copy the behavior of PHP. If PHP has no behavior in
> this regard tools will deviate and generics won't have any real
> meaning at all. There are many nuances to generics (ask Matthew who
> wrote Psalm), we shouldn't underestimate that. With a hybrid approach
> we could also do full type checking in testing environments and
> disable it in more performance critical situations with more
> confidence.

Larry also shared this concern, and I agree that it's important to at =
least have an official spec from the PHP project. Whether part of it =
should be enforced at runtime is up for debate, but it's definitely =
important.

>=20
> Ilija
>=20
> --
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: https://www.php.net/unsub.php
>=20

Kind regards
Brent