Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:125889
Feedback-ID: ifab94697:Fastmail
Precedence: bulk
MIME-Version: 1.0
Date: Fri, 01 Nov 2024 17:08:52 +0100
To: "Gina P. Banyard" <internals@gpb.moe>
Cc: internals@lists.php.net
Message-ID: <e1d61615-3413-4ff6-ba10-bd76e7e3c1a2@app.fastmail.com>
In-Reply-To: 
 <PEwLeSeAMbMpMjdoXLSNgqroEdeX7yJfLEv1q9cKwaW1nUkIGqU_6MIdgbH-tIZpihEL3i38aJd0mgVhBX8adOvNlzL9P1agUDv01bAwMQs=@gpb.moe>
References: <15da4c13445d7e9c9d768c60c19768d4@bastelstu.be>
 <3b458165-406c-4b70-97bc-6e98d6c44c72@app.fastmail.com>
 <6f39dce9e6b0579baa51bc84cb8140b9@bastelstu.be>
 <fe2230a0-6d1c-4256-8846-ee6de3589451@app.fastmail.com>
 <c2c5b3d2-46bd-4625-a5dd-13c58c865da8@app.fastmail.com>
 <PEwLeSeAMbMpMjdoXLSNgqroEdeX7yJfLEv1q9cKwaW1nUkIGqU_6MIdgbH-tIZpihEL3i38aJd0mgVhBX8adOvNlzL9P1agUDv01bAwMQs=@gpb.moe>
Subject: Re: [PHP-DEV] RFC: Support Closures in constant expressions
Content-Type: multipart/alternative;
 boundary=03950c0b44894ffd895157f3104a8b88
From: rob@bottled.codes ("Rob Landers")

--03950c0b44894ffd895157f3104a8b88
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

On Thu, Oct 31, 2024, at 15:12, Gina P. Banyard wrote:
> On Thursday, 31 October 2024 at 07:22, Rob Landers <rob@bottled.codes>=
 wrote:
>=20
>> To be honest, I thought it was a rhetorical question since the exampl=
e is a runtime error (passing string to a function that takes an array),=
 but on this note, we already know how it should behave: https://3v4l.or=
g/RC6b3 because, as you said, it is captured by value. In theory, to sup=
port this, we would probably need =E2=80=9Clate binding constants=E2=80=9D=
 or constants that are executed just before any userland code is run. Si=
milar to just emulating it yourself: https://3v4l.org/PMY4W
>>=20
>> IMHO, the rules around constant expressions are not sound and feel ar=
bitrary when you run into them. They are too easy to work around (though=
 cumbersome) and the code still works, which seems to prove their arbitr=
ary nature.
>=20
> define() does not produce "true" constants and has always allowed to d=
efine constants at run-time where there is more information available.
>=20
> const expressions have *always* been about determining and defining co=
nstants at compile time so they can be put in SHM.
> Which is why the const keyword exist.
> (and yes I know define() expression that are const expression are "tru=
e compile time constants")
>=20
> Best regards,
>=20
> Gina P. Banyard

Hey Gina,

I was mostly referring to the comment that we can't define the expected =
behavior, not necessarily what is a "true constant" or not.

=E2=80=94 Rob
--03950c0b44894ffd895157f3104a8b88
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html><html><head><title></title><style type=3D"text/css">p.Mso=
Normal,p.MsoNoSpacing{margin:0}</style></head><body><div>On Thu, Oct 31,=
 2024, at 15:12, Gina P. Banyard wrote:<br></div><blockquote type=3D"cit=
e" id=3D"qt" style=3D""><div class=3D"qt-protonmail_quote"><div>On Thurs=
day, 31 October 2024 at 07:22, Rob Landers &lt;rob@bottled.codes&gt; wro=
te:<br></div><div><br></div><blockquote class=3D"qt-protonmail_quote" ty=
pe=3D"cite"><div>To be honest, I thought it was a rhetorical question si=
nce the example is a runtime error (passing string to a function that ta=
kes an array), but on this note, we already know how it should behave:&n=
bsp;<a href=3D"https://3v4l.org/RC6b3" target=3D"_blank" rel=3D"noreferr=
er nofollow noopener">https://3v4l.org/RC6b3</a>&nbsp;because, as you sa=
id, it is captured by value. In theory, to support this, we would probab=
ly need =E2=80=9Clate binding constants=E2=80=9D or constants that are e=
xecuted just before any userland code is run. Similar to just emulating =
it yourself:&nbsp;<a href=3D"https://3v4l.org/PMY4W" target=3D"_blank" r=
el=3D"noreferrer nofollow noopener">https://3v4l.org/PMY4W</a><br></div>=
<div><br></div><div>IMHO, the rules around constant expressions are not =
sound and feel arbitrary when you run into them. They are too easy to wo=
rk around (though cumbersome) and the code still works, which seems to p=
rove their arbitrary nature.<br></div></blockquote><div style=3D"font-fa=
mily:Arial, sans-serif;font-size:14px;color:rgb(0, 0, 0);background-colo=
r:rgb(255, 255, 255);"><br></div><div style=3D"font-family:Arial, sans-s=
erif;font-size:14px;color:rgb(0, 0, 0);background-color:rgb(255, 255, 25=
5);"><div><span>define()=20=0Adoes not produce "true" constants and has =
always allowed to define=20=0Aconstants at run-time where there is more =
information available.</span><br></div><div><br></div><div>const express=
ions have *always* been about determining and defining constants at comp=
ile time so they can be put in SHM.<br></div><div>Which is why the const=
 keyword exist.<br></div><div>(and yes I know define() expression that a=
re const expression are "true compile time constants")<br></div><div><br=
></div><div><div><div style=3D"background-color:rgb(255, 255, 255);"><sp=
an class=3D"font" style=3D"font-family:Verdana, sans-serif;">Best regard=
s,</span><br></div><div style=3D"background-color:rgb(255, 255, 255);"><=
br></div><div><span style=3D"background-color:rgb(255, 255, 255);"><span=
 class=3D"font" style=3D"font-family:Verdana, sans-serif;">Gina P. Banya=
rd</span></span><br></div></div></div></div></div></blockquote><div><br>=
</div><div>Hey Gina,<br></div><div><br></div><div>I was mostly referring=
 to the comment that we can't define the expected behavior, not necessar=
ily what is a "true constant" or not.</div><div><br></div><div id=3D"sig=
121229152">=E2=80=94 Rob<br></div></body></html>
--03950c0b44894ffd895157f3104a8b88--