Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:130111
Precedence: list
MIME-Version: 1.0
References: <CAO3baw=Qe-w4UOQzc4_NO=b41N-DJtf8U9Q6HybMCcCtZoEqgA@mail.gmail.com>
 <3ACF2D3E-0376-4033-8CAA-84FAD83E2A69@gmail.com>
In-Reply-To: <3ACF2D3E-0376-4033-8CAA-84FAD83E2A69@gmail.com>
Date: Fri, 20 Feb 2026 23:07:38 +0530
Message-ID: <CAO3baw=A5hnGCWbY3_4LE=nYm8+=R1kcAvfjKQWLS-ODyXx0Kw@mail.gmail.com>
Subject: =?UTF-8?Q?Re=3A_=5BPHP=2DDEV=5D_=5BRFC=5D_array=5Fonly=28=29_and_array=5Fexcept=28=29_?=
	=?UTF-8?Q?=E2=80=93_Utility_functions_for_array_key_filtering?=
To: Claude Pache <claude.pache@gmail.com>
Cc: internals@lists.php.net
Content-Type: multipart/alternative; boundary="0000000000006755ac064b44e17f"
From: arshidkv12@gmail.com (Muhammed Arshid KV)

--0000000000006755ac064b44e17f
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi,

Thanks for the feedback.

1. Yes, *array_intersect_key() *and *array_diff_key() *can be used today,
but the RFC proposes dedicated functions for better readability, less
boilerplate, and potentially better performance and lower memory usage by
avoiding temporary arrays.

2. Good point about naming. Making the key-filtering behavior explicit
(e.g., *array_only_keys* and *array_except_keys*) improves clarity and
avoids confusion.

On Fri, 20 Feb 2026, 3:38=E2=80=AFpm Claude Pache, <claude.pache@gmail.com>=
 wrote:

>
>
> Le 20 f=C3=A9vr. 2026 =C3=A0 10:25, Muhammed Arshid KV <arshidkv12@gmail.=
com> a
> =C3=A9crit :
>
> Hello Internals,
>
> I would like to propose a new RFC titled *array_only() and array_except()=
*
> .
>
> RFC: https://wiki.php.net/rfc/array_only_except
> <https://wiki.php.net/rfc/array_only_except?utm_source=3Dchatgpt.com>
>
> This RFC proposes adding two utility functions to the PHP core:
>
> =E2=80=A2 array_only(array $array, array $keys): array
> =E2=80=A2 array_except(array $array, array $keys): array
>
> These functions allow selecting or excluding specific keys from an array.
>
> Similar helpers are commonly used in frameworks, and adding native suppor=
t
> would improve readability and consistency in userland code.
>
> Summary:
>
> =E2=80=A2 array_only() returns a new array containing only the specified =
keys.
> =E2=80=A2 array_except() returns a new array excluding the specified keys=
.
> =E2=80=A2 Original array is not modified.
> =E2=80=A2 Works with associative and numeric keys.
> =E2=80=A2 No impact on SAPIs or on OPcache.
>
> I would appreciate feedback and suggestions from the community.
> If there is general agreement, I plan to move the RFC forward.
>
> Thanks,
> Muhammed Arshid KV
>
>
> Hi,
>
> The principle of the RFC is interesting to me, as it is a situation I
> encounter from time to time. Two remarks:
>
> 1. It may be interesting to compare with currently available solutions.
> Currently, I use  `array_intersect_key` and `array_diff_key` when I need
> the semantics of proposed `array_only` and `array_except`:
>
> ```php
> array_intersect_key($data, ['id' =3D> null, 'email' =3D> null]);
> ```
>
> 2. I think that the name of the new functions ought to make evident that
> we are filtering on keys and not on something else. I suggest:
> `array_only_keys` and `array_except_keys`.
>
> =E2=80=94Claude
>
>

--0000000000006755ac064b44e17f
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"auto"><div>Hi,</div><div dir=3D"auto"><br></div><div dir=3D"aut=
o">Thanks for the feedback.</div><div dir=3D"auto"><br></div><div dir=3D"au=
to">1. Yes, <b>array_intersect_key() </b>and <b>array_diff_key() </b>can be=
 used today, but the RFC proposes dedicated functions for better readabilit=
y, less boilerplate, and potentially better performance and lower memory us=
age by avoiding temporary arrays.</div><div dir=3D"auto"><br></div><div dir=
=3D"auto">2. Good point about naming. Making the key-filtering behavior exp=
licit (e.g., <b>array_only_keys</b> and <b>array_except_keys</b>) improves =
clarity and avoids confusion.<br><br><div class=3D"gmail_quote gmail_quote_=
container" dir=3D"auto"><div dir=3D"ltr" class=3D"gmail_attr">On Fri, 20 Fe=
b 2026, 3:38=E2=80=AFpm Claude Pache, &lt;<a href=3D"mailto:claude.pache@gm=
ail.com">claude.pache@gmail.com</a>&gt; wrote:<br></div><blockquote class=
=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd=
ing-left:1ex"><div style=3D"line-break:after-white-space"><br id=3D"m_-2083=
200897647273772lineBreakAtBeginningOfMessage"><div><br><blockquote type=3D"=
cite"><div>Le 20 f=C3=A9vr. 2026 =C3=A0 10:25, Muhammed Arshid KV &lt;<a hr=
ef=3D"mailto:arshidkv12@gmail.com" target=3D"_blank" rel=3D"noreferrer">ars=
hidkv12@gmail.com</a>&gt; a =C3=A9crit :</div><br><div><div dir=3D"ltr"><p>=
Hello Internals,</p><p>I would like to propose a new RFC titled <strong>arr=
ay_only() and array_except()</strong>.</p><p>RFC: <a rel=3D"noopener norefe=
rrer" href=3D"https://wiki.php.net/rfc/array_only_except?utm_source=3Dchatg=
pt.com" target=3D"_blank">https://wiki.php.net/rfc/array_only_except<span a=
ria-hidden=3D"true"></span></a></p><p>This RFC proposes adding two utility =
functions to the PHP core:</p><p>=E2=80=A2 array_only(array $array, array $=
keys): array<br>
=E2=80=A2 array_except(array $array, array $keys): array</p><p>These functi=
ons allow selecting or excluding specific keys from an array.=C2=A0</p><p>S=
imilar helpers are commonly used in frameworks, and adding native support w=
ould improve readability and consistency in userland code.</p><p>Summary:</=
p><p>=E2=80=A2 array_only() returns a new array containing only the specifi=
ed keys.<br>
=E2=80=A2 array_except() returns a new array excluding the specified keys.<=
br>
=E2=80=A2 Original array is not modified.<br>
=E2=80=A2 Works with associative and numeric keys.<br>
=E2=80=A2 No impact on SAPIs or on <span><span>OPcache</span></span>.</p><p=
>I would appreciate feedback and suggestions from the community.<br>
If there is general agreement, I plan to move the RFC forward.</p><p>Thanks=
,<br>
Muhammed Arshid KV</p></div>
</div></blockquote><br></div><div>Hi,</div><div><br></div><div>The principl=
e of the RFC is interesting to me, as it is a situation I encounter from ti=
me to time. Two remarks:</div><div><br></div><div>1. It may be interesting =
to compare with currently available solutions. Currently, I use =C2=A0`arra=
y_intersect_key` and `array_diff_key` when I need the semantics of proposed=
 `array_only` and `array_except`:</div><div><br></div><div>```php</div><div=
>array_intersect_key($data, [&#39;id&#39; =3D&gt; null, &#39;email&#39; =3D=
&gt; null]);</div><div>```</div><div><br></div><div>2. I think that the nam=
e of the new functions ought to make evident that we are filtering on keys =
and not on something else. I suggest: `array_only_keys` and `array_except_k=
eys`.</div><div><br></div><div>=E2=80=94Claude</div><div><br></div></div></=
blockquote></div></div></div>

--0000000000006755ac064b44e17f--