Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:111889
X-MailCleaner-return_path: brendt@stitcher.io
X-MailCleaner-sender_address: brendt@stitcher.io
X-MailCleaner-recipients: larry@garfieldtech.com, internals@lists.php.net
X-MailCleaner-SPF: pass
Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.80.23.2.2\))
In-Reply-To: <4608608e-906c-45d2-9f7e-e7f4213aa994@www.fastmail.com>
Date: Thu, 17 Sep 2020 19:28:27 +0200
Cc: php internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
Message-ID: <D50B046A-FDA9-4117-8C7B-295ABE070CF3@stitcher.io>
References: <09DE673B-71B7-4C8E-A795-900B51F8E476@stitcher.io>
 <CAKuGYLSLm0aFVXdxcywzTAqbaohE--4zaaqFg9HYpDgQv6o=Pg@mail.gmail.com>
 <4608608e-906c-45d2-9f7e-e7f4213aa994@www.fastmail.com>
To: Larry Garfield <larry@garfieldtech.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 Matthew and Larry

> On 17 Sep 2020, at 18:21, Larry Garfield <larry@garfieldtech.com> =
wrote:
>=20
> On Thu, Sep 17, 2020, at 10:21 AM, Matthew Brown wrote:
>> Quick thing before I get into my own reaction:
>>=20
>> Transpiling is normally thought of as the process of converting one
>> language into another. Tools like Babel transpile TypeScript to =
JavaScript.

I looked up the definition [1] before writing my mail and it said=20

> A [=E2=80=A6] transpiler is a type of translator that takes the source =
code of a program written in a programming language as its input and =
produces an equivalent source code in the same or a different =
programming language

But I don't want to get stuck on phrasing, if elidiing is the right term =
as Larry suggests, let's go with it!

>>=20
>> What's being proposed here (AFAICT) is type erasure =E2=80=93 the =
generic type
>> information would be erased during the conversion to opcodes.
>>=20
>> Python (the language) has type erasure for all its types (including =
its
>> generic types), so no popular Python interpreters check that a =
function
>> call's argument types match up with the function signature.

I didn't know about Python's approach before, and I think it's =
validating the idea, it's at least something worth looking into.

>>=20
>> PHP currently does not erase any types so the opcodes generated by =
PHP's
>> interpreter include type checks for all the arguments passed to a =
typed
>> function signature.
>>=20
>> Hack follows PHP's model, but erases generic types by default (though =
has
>> more recently introduced the concept of reified generics) in much the =
same
>> way you're proposing.
>=20
> I concur; this is type elidiing, not transpiling.  Transpiling implies =
a user-triggered pre-run step.  I know Sara has mused about it before =
but it's not happened yet.
>=20
> Basically what is being proposed (for those who need a concrete =
example) is you'd write this:
>=20
> class Collection<T> {
>=20
>  public function add(T $item) { ... }
>=20
> }
> $c =3D new Collection<Product>().
>=20
> And linters would recognize that, and it would parse, but at runtime =
it would compile to:
>=20
> class Collection {
>=20
>  public function add(mixed $item) { ... }
>=20
> }
> $c =3D new Collection().
>=20
>=20
> And that's the opcodes that would be saved.  (I'm hand-waving a lot =
here.)
>=20
> I would be on board with this, although I would ask if it's possible =
to provide some compile time checks; I'm thinking LSP validation, which =
is still relevant for inheritance with generics.  Beyond that, leave the =
runtime to not enforce it.  (I have no idea how viable that is, but Sara =
seemed to think it would work so it probably=20

I agree that's an approach worth looking into. I didn't want to discuss =
specifics yet, and first get a feel of what people think of the general =
idea, but it's good that you're mentioning this, and I'm glad others are =
open to the idea.

>=20
> Another possibility could be to desugar that original to:
>=20
> class Collection {
>  public $__T;
>=20
>  public function add(mixed $item) {
>    if (!$item instanceof $this->__T) {
>      throw new TypeError();
>    }
>  }
> }
> $c =3D new Collection().
> $c->__T =3D=3D 'Product';
>=20
> Which is essentially what you can do in userspace today, but =
automated.  I don't know how viable that is, but it's another thing to =
consider.
>=20
> --Larry Garfield
>=20
> --
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: https://www.php.net/unsub.php
>=20

Kind regards
Brent

[1] https://en.wikipedia.org/wiki/Source-to-source_compiler