Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:125152
Date: Fri, 23 Aug 2024 13:10:48 -0400
To: Stephen Reay <php-lists@koalephant.com>
Cc: =?utf-8?Q?Rowan_Tommins_=5BIMSoP=5D?= <imsop.php@rwec.co.uk>, 
 "=?utf-8?Q?internals=40lists.php.net?=" <internals@lists.php.net>
Message-ID: <131FC621-823F-4E17-83DB-D8B83F535A2E@getmailspring.com>
In-Reply-To: <828A2555-E7CC-4DD8-AE1B-8D50EB7EDC6F@koalephant.com>
References: <828A2555-E7CC-4DD8-AE1B-8D50EB7EDC6F@koalephant.com>
Subject: Re: [PHP-DEV] [Concept] Flip relative function lookup order
 (global, then local)
Precedence: bulk
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="66c8c298_643c9869_15341"
From: john@coggeshall.org (John Coggeshall)

--66c8c298_643c9869_15341
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline



On Aug 23 2024, at 12:35 pm, Stephen Reay <php-lists@koalephant.com> wrote:
>
> > would find it non-sensical that you must add a backslash for the engine to do the "right thing" (in this case, optimize their code with a security benefit), vs just doing the right thing by default.
>
> What do you mean by this? What is "the right thing"?
I mean this:
<?php
// something.php
namespace App\Models;

function password_hash(string $password, string|int|null $algo, array $options = []): string
{
print("Hello");
return $password;
}

<?php
// my code
namespace App\Models;

include "something.php";
password_hash('foobar', PASSWORD_DEFAULT);
This code IMO shouldn't print "Hello", but it does. The current behavior of looking up the local namespace first for functions, instead of the global namespace first, IMO is the "wrong thing" because it expects developers to fully qualify internal function calls every single time because, under very specific legitimate use cases that are pretty rare in the course of normal development, you actually might want to do that.
> I'm not saying that isn't necessarily how people think, I have literally zero data about this besides my own thoughts, but it seems like a bizarre idea that people would *expect* function name resolution to work completely opposite to how class/class-like name resolution works.
>

At it's core a vast majority of the functionality of the PHP language exists within internally-implemented functions, not classes. So yes, I think it's entirely reasonable that people would expect that internal functions resolve at a higher priority than user-defined functions with the same name, and that if you'd like to reuse a global namespaced function in your local namespace you need to be explicit about that -- not the other way around.

--66c8c298_643c9869_15341
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

<br><br><div class=3D=22gmail=5Fquote=5Fattribution=22>On Aug 23 2024, at=
 12:35 pm, Stephen Reay &lt;php-lists=40koalephant.com&gt; wrote:</div><b=
lockquote><br><blockquote>would find it non-sensical that you must add a =
backslash for the engine to do the =22right thing=22 (in this case, optim=
ize their code with a security benefit), vs just doing the right thing by=
 default.&nbsp;</blockquote><div><br></div><div>What do you mean by this=3F=
 What is =22the right thing=22=3F</div></blockquote><br><div>I mean this:=
</div><br><div>&lt;=3Fphp</div><div>// something.php</div><div>namespace =
App=5CModels;</div><br><div>function password=5Fhash(string =24password, =
string=7Cint=7Cnull =24algo, array =24options =3D =5B=5D): string</div><d=
iv>=7B</div><div>&nbsp;&nbsp; print(=22Hello=22);</div><div>&nbsp;&nbsp; =
return =24password;</div><div>=7D</div><br><div>&lt;=3Fphp</div><div>// m=
y code</div><div>namespace App=5CModels;</div><br><div>include =22somethi=
ng.php=22;</div><br><div>password=5Fhash('foobar', PASSWORD=5FDE=46AULT);=
</div><br><div>This code IMO shouldn't print =22Hello=22, but it does. Th=
e current behavior of looking up the local namespace first for functions,=
 instead of the global namespace first, IMO is the =22wrong thing=22 beca=
use it expects developers to fully qualify internal function calls every =
single time because, under very specific legitimate use cases that are pr=
etty rare in the course of normal development, you actually might want to=
 do that.</div><br><blockquote><div><div>I'm not saying that isn't necess=
arily how people think, I have literally zero data about this besides my =
own thoughts, but it seems like a bizarre idea that people would *expect*=
 function name resolution to work completely opposite to how class/class-=
like name resolution works.</div></div><div><br></div></blockquote><br><d=
iv>At it's core a vast majority of the functionality of the PHP language =
exists within internally-implemented functions, not classes. So yes, I th=
ink it's entirely reasonable that people would expect that internal funct=
ions resolve at a higher priority than user-defined functions with the sa=
me name, and that if you'd like to reuse a global namespaced function in =
your local namespace you need to be explicit about that -- not the other =
way around.</div><br>
--66c8c298_643c9869_15341--