Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:129533
Feedback-ID: ifab94697:Fastmail
Precedence: list
MIME-Version: 1.0
Date: Wed, 03 Dec 2025 22:49:08 +0100
To: "Paul M. Jones" <pmjones@pmjones.io>
Cc: internals@lists.php.net
Message-ID: <06fcc9ac-80e6-4046-8067-b26c0a489843@app.fastmail.com>
In-Reply-To: <D3A55D97-285D-4C92-B8EB-CD9565DC044C@pmjones.io>
References: <87e9d1bf-e407-45c1-9fad-d8759405ab8b@app.fastmail.com>
 <D3A55D97-285D-4C92-B8EB-CD9565DC044C@pmjones.io>
Subject: Re: [PHP-DEV] [RFC] Type Aliases
Content-Type: multipart/alternative;
 boundary=d9da8edf21c2493f8c22ae8f8c23974d
From: rob@bottled.codes ("Rob Landers")

--d9da8edf21c2493f8c22ae8f8c23974d
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

On Wed, Dec 3, 2025, at 22:27, Paul M. Jones wrote:
> Hey Rob & all,
>=20
> One objection, one question.
>=20
> The objection is to `include types 'types.php'`. You've clarified why =
types cannot be autoloaded (thanks again!) and so this is a way to let t=
he compiler know where the types are. But I have to say I've spent a hug=
e part of my career removing include/require when refactoring legacy cod=
e, and introducing it here causes me great heartache.

A new way of autoloading is out of scope... :D It doesn't preclude it fr=
om being autoloaded in the future, but currently, the only other reason =
to refactor autoloading would be to get function autoloading. That in it=
self would be nice, having another reason wouldn't necessarily be a bad =
thing either.

Further up-thread, I mentioned there were two ways to do this: this way,=
 and making aliases "global". For example, a "use global type ... as ...=
" or something like that. I'm not opposed to this, per se, and if that=E2=
=80=99s the consensus, I'm fine with it.

> The question is a little more complicated, and also involves autoloadi=
ng. Given a class Number that can be autoloaded ...
>=20
> namespace Foo;
>=20
> class Number { /* ... */ }
>=20
> ... and *another* class in the same namespace, that creates a "number"=
 type alias...
>=20
> namespace Foo;
>=20
> use type int|float as Number;
>=20
> class Bar { public function add(Number $a, Number $b) : Number { /* ..=
. */ }
>=20
> ... which one "wins" -- the runtime autoloaded class, or the compile-t=
ime type alias?
>=20
> (Apologies if this is covered in the RFC and I missed it.)
>=20

It works just like it works today if you were to do this:

namespace Foo {
 class Number {}
}

namespace Foo {
  use GMP as Number;

  class Bar { public function add Number $a, \Foo\Number $b): Number {}
}

In other words, the alias shadows the original and you have to use the F=
QN to resolve the real one.

=E2=80=94 Rob
--d9da8edf21c2493f8c22ae8f8c23974d
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html><html><head><title></title></head><body><div>On Wed, Dec =
3, 2025, at 22:27, Paul M. Jones wrote:</div><blockquote type=3D"cite" i=
d=3D"qt" style=3D""><div>Hey Rob &amp; all,</div><div><br></div><div>One=
 objection, one question.</div><div><br></div><div>The objection is to `=
include types '<a href=3D"http://types.php">types.php</a>'`. You've clar=
ified why types cannot be autoloaded (thanks again!) and so this is a wa=
y to let the compiler know where the types are. But I have to say I've s=
pent a huge part of my career removing include/require when refactoring =
legacy code, and introducing it here causes me great heartache.</div></b=
lockquote><div><br></div><div>A new way of autoloading is out of scope..=
. :D It doesn't preclude it from being autoloaded in the future, but cur=
rently, the only other reason to refactor autoloading would be to get fu=
nction autoloading. That in itself would be nice, having another reason =
wouldn't necessarily be a bad thing either.</div><div><br></div><div>Fur=
ther up-thread, I mentioned there were two ways to do this: this way, an=
d making aliases "global". For example, a "use global type ... as ..." o=
r something like that. I'm not opposed to this, per se, and if that=E2=80=
=99s the consensus, I'm fine with it.<br></div><div><br></div><blockquot=
e type=3D"cite" id=3D"qt" style=3D""><div>The question is a little more =
complicated, and also involves autoloading. Given a class Number that ca=
n be autoloaded ...</div><div><br></div><div>namespace Foo;</div><div><b=
r></div><div>class Number { /* ... */ }</div><div><br></div><div>... and=
 *another* class in the same namespace, that creates a "number" type ali=
as...</div><div><br></div><div>namespace Foo;</div><div><br></div><div>u=
se type int|float as Number;</div><div><br></div><div>class Bar { public=
 function add(Number $a, Number $b) : Number { /* ... */ }</div><div><br=
></div><div>... which one "wins" -- the runtime autoloaded class, or the=
 compile-time type alias?</div><div><br></div><div>(Apologies if this is=
 covered in the RFC and I missed it.)</div><div><br></div></blockquote><=
div><br></div><div>It works just like it works today if you were to do t=
his:</div><div><br></div><pre style=3D"border-top-width:1px;border-right=
-width:1px;border-bottom-width:1px;border-left-width:1px;border-top-styl=
e:solid;border-right-style:solid;border-bottom-style:solid;border-left-s=
tyle:solid;border-top-color:rgb(204, 204, 204);border-right-color:rgb(20=
4, 204, 204);border-bottom-color:rgb(204, 204, 204);border-left-color:rg=
b(204, 204, 204);border-image-source:initial;border-image-slice:initial;=
border-image-width:initial;border-image-outset:initial;border-image-repe=
at:initial;border-top-left-radius:3px;border-top-right-radius:3px;border=
-bottom-right-radius:3px;border-bottom-left-radius:3px;background-image:=
initial;background-position-x:initial;background-position-y:initial;back=
ground-size:initial;background-repeat:initial;background-attachment:init=
ial;background-origin:initial;background-clip:initial;background-color:r=
gb(246, 246, 246);font-family:menlo, consolas, monospace;font-size:90%;m=
argin-top:7px;margin-right:0px;margin-bottom:7px;margin-left:0px;padding=
-top:7px;padding-right:10px;padding-bottom:7px;padding-left:10px;white-s=
pace-collapse:preserve;text-wrap-mode:wrap;overflow-wrap:break-word;">na=
mespace Foo {=0A class Number {}=0A}=0A=0Anamespace Foo {=0A  use GMP as=
 Number;=0A=0A  class Bar { public function add Number $a, \Foo\Number $=
b): Number {}=0A}<br></pre><div><br></div><div>In other words, the alias=
 shadows the original and you have to use the FQN to resolve the real on=
e.</div><div><br></div><div id=3D"sig121229152">=E2=80=94 Rob</div></bod=
y></html>
--d9da8edf21c2493f8c22ae8f8c23974d--