Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:119900
Date: Tue, 11 Apr 2023 09:48:20 +0100
To: internals@lists.php.net
User-Agent: K-9 Mail for Android
In-Reply-To: <CAFPFaMLghRYP4ioCkfa7WKhTAz+1gWgcZ5mWYAunkb9f3g=fdA@mail.gmail.com>
References: <CAFPFaMLghRYP4ioCkfa7WKhTAz+1gWgcZ5mWYAunkb9f3g=fdA@mail.gmail.com>
Message-ID: <6FA5AA43-9738-4400-8D83-38FC2D464B82@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Subject: =?US-ASCII?Q?Re=3A_=5BPHP-DEV=5D_=5BRFC=5D_New_core_autoloading_mec?= =?US-ASCII?Q?hanism_with_support_for_function_autoloading?=
From: rowan.collins@gmail.com (Rowan Tommins)

On 10 April 2023 13:17:04 BST, "G=2E P=2E B=2E" <george=2Ebanyard@gmail=2Ec=
om> wrote:
>Hello Internals,
>
>Dan and I would like to propose a new core autoloading mechanism that fix=
es
>some minor design issues with the current class autoloading mechanism and
>introduce a brand-new function autoloading mechanism:
>https://wiki=2Ephp=2Enet/rfc/core-autoloading

Hi!=20

Thanks both for tackling this, I know function autoloading has long been o=
n a lot of people's wish lists=2E

A few thoughts:



> As a result of the pinning, function_exists() will return true for funct=
ions in namespaces that have been pinned to a global function=2E=20

The lookup pinning seems like the right way to optimise the implementation=
, but I don't think it should be visible to user code like this - a lookup =
for "strlen" in namespace "Foo" is not the same as defining the function "F=
oo\strlen"=2E Consider this code:

namespace Bar {
    if ( function_exists('Foo\strlen') ) {
        \Foo\strlen('hello');
    }
}
namespace Foo {
     strlen('hello'); // triggers name pinning
}
namespace Bar {
    if ( function_exists('Foo\strlen') ) {
        \Foo\strlen('hello');
    }
}

If I'm reading the RFC correctly, the second function_exists will return t=
rue=2E I'm less clear if the call to \Foo\strlen will actually succeed - if=
 it gives "undefined function", then function_exists is clearly broken; if =
it calls the global strlen(), that's a very surprising side effect=2E For b=
onus points, the call to strlen that triggers pinning could be inside an au=
toloader, making even the first function_exists call return true=2E

Similarly, I think it should be possible to "unpin" a function lookup with=
 a later definition, even if no autoloading would be triggered=2E That is, =
this should not be a duplicate definition error:

namespace Foo;
if ( strlen('magic') !=3D 42 ) {
    function strlen($string) { /* =2E=2E=2E */ }
}




> The use of the word class in the API is currently accurate

This isn't actually true: classes, interfaces, traits, and enums all share=
 a symbol table, and thus an autoloader=2E I don't know of a good name for =
this symbol table, though=2E




Regarding the API, would it be possible to take advantage of nearly all au=
toloaders only being interested in particular namespace prefixes?=20

Currently, every registered autoloader is run for every lookup, and most i=
mmediately check the input for one or two prefixes, and return early if not=
 matched=2E I suspect this design is largely because autoloading came befor=
e namespaces, so the definition of "prefix" wasn't well-defined, but going =
in and out of userland callbacks like this is presumably rather inefficient=
=2E

Perhaps the "register" functions should take an optional list of namespace=
 prefixes, so that the core implementation can do the string comparison, an=
d only despatch to the userland code if the requested class/function name m=
atches=2E



Thanks again for working on this!

Regards,

--=20
Rowan Tommins
[IMSoP]