Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:130477
Precedence: list
MIME-Version: 1.0
References: <tencent_D07F06F5B18B1450B922355C06A3ED7E4107@qq.com>
 <e3ea7fd0-3fa1-4f6f-a3a6-a7cc692d9d5f@alec.pl> <CAMXcktOAwUvQ0sBs9vpbG-L2AHqbaoy=fwKfKibVdSxtJW1ygA@mail.gmail.com>
 <5c6d66c4-4c17-444e-bb0a-2b8c4759d2db@app.fastmail.com> <48BA5AE0-FADB-4675-B2FC-D56AB7A793CA@rwec.co.uk>
 <ec381b0a-20e0-4009-9346-38bb2cce72e9@varteg.nz> <CAMXcktO1fjrf98YwiGqV7aOUgC62WOJ3kKSxhW4SYH_d8_jkHA@mail.gmail.com>
 <06a4003b-b74f-4fba-8f96-75fa395ff798@gmail.com> <759fb4e0-4c89-4065-b6dd-abbf297213ac@rwec.co.uk>
In-Reply-To: <759fb4e0-4c89-4065-b6dd-abbf297213ac@rwec.co.uk>
Date: Sat, 28 Mar 2026 15:27:44 -0400
Message-ID: <CAMXcktMgvKexo--rur=cYVBs089-VgCe06L8bjuso6UNhnpZUA@mail.gmail.com>
Subject: Re: [PHP-DEV] [RFC] Four Pragmatic Directions for PHP: Simplicity,
 Arrays, Performance, Concurrency
To: "Rowan Tommins [IMSoP]" <imsop.php@rwec.co.uk>
Cc: internals@lists.php.net
Content-Type: multipart/alternative; boundary="0000000000004d2bda064e1a9d10"
From: dliebner@gmail.com (Dan Liebner)

--0000000000004d2bda064e1a9d10
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Yeah, adding `?` after expressions gets weird because of ternary operators.

I do think the long-form `$foo[?'one'][?'two'][?'three']` is acceptable in
terms of developer experience considering it's not a super common scenario.
I'm not sure if that would be annoying to implement on the C side of things=
.

- Dan

On Sat, Mar 28, 2026 at 2:33=E2=80=AFPM Rowan Tommins [IMSoP] <imsop.php@rw=
ec.co.uk>
wrote:

> On 27/03/2026 15:19, Alex Rock wrote:
> >
> > In that case, maybe reuse something similar to the null-safe operator ?
> >
> > Something like `$foo['bar']?`
> >
>
> I think even if you could make it work in the parser, it would be hard
> to make that read well with the ternary operator:
>
> $foo['a'] ? ['b'] ? ['c'] : ['d'];
>
>
> >     return $foo['one']?['two'];
>
> Looking at that, it's not obvious to me whether it's 'one' or 'two'
> which is optional.
>
> return $foo['one']  ?['two'];
> return $foo['one']?  ['two'];
>
>
> > And here it gets interesting:
> >
> >     return $foo['one']['two']?['three'];
> >     // is roughly equivalent to:
> >     if (!isset($foo['one']) {
> >     trigger_error('Undefined key "one" in ...', ...);
> >     }
> >     return $foo['one']['two']['three'] ?? null;
> >
>
> You can have the same ability with a ? inside the brackets, and without
> the ambiguity:
>
> // all three keys required to exist (as of PHP 9):
> return $foo['one']['two']['three'];
>
> // 'one' and 'two' required, 'three' optional:
> return $foo['one']['two'][?'three'];
>
> // both 'two' and 'three' optional:
> return $foo['one'][?'two'][?'three'];
>
> // optional all the way:
> return $foo[?'one'][?'two'][?'three'];
>
>
> This would maybe have similar short-cut semantics to ?->
>
> $foo =3D [];
> return $foo[?'one']['two'];
> // OK: the ? neutralises access to the missing section of the array,
> short-cutting to NULL
> // Similar to $foo?->bar->baz when $foo is NULL
>
> $foo =3D ['one' =3D> []];
> return $foo[?'one']['two'];
> // Error: the ? had no effect, so we're trying to access non-existent
> key 'two' in the empty array $foo['one']
> // Similar to $foo?->bar->baz when $foo->bar exists
>
>
> But an RFC could pin down those details.
>
>
> --
> Rowan Tommins
> [IMSoP]
>

--0000000000004d2bda064e1a9d10
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Yeah, adding `?` after expressions gets weird because=
 of ternary operators.</div><div><br></div><div>I do think the long-form `$=
foo[?&#39;one&#39;][?&#39;two&#39;][?&#39;three&#39;]` is acceptable in ter=
ms of developer experience considering it&#39;s not a super common scenario=
. I&#39;m not sure if that would be annoying to implement on the C side of =
things.</div><div><br></div><div>- Dan</div></div><br><div class=3D"gmail_q=
uote gmail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Sat, M=
ar 28, 2026 at 2:33=E2=80=AFPM Rowan Tommins [IMSoP] &lt;<a href=3D"mailto:=
imsop.php@rwec.co.uk">imsop.php@rwec.co.uk</a>&gt; wrote:<br></div><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px=
 solid rgb(204,204,204);padding-left:1ex">On 27/03/2026 15:19, Alex Rock wr=
ote:<br>
&gt;<br>
&gt; In that case, maybe reuse something similar to the null-safe operator =
?<br>
&gt;<br>
&gt; Something like `$foo[&#39;bar&#39;]?`<br>
&gt;<br>
<br>
I think even if you could make it work in the parser, it would be hard <br>
to make that read well with the ternary operator:<br>
<br>
$foo[&#39;a&#39;] ? [&#39;b&#39;] ? [&#39;c&#39;] : [&#39;d&#39;];<br>
<br>
<br>
&gt; =C2=A0 =C2=A0 return $foo[&#39;one&#39;]?[&#39;two&#39;];<br>
<br>
Looking at that, it&#39;s not obvious to me whether it&#39;s &#39;one&#39; =
or &#39;two&#39; <br>
which is optional.<br>
<br>
return $foo[&#39;one&#39;]=C2=A0 ?[&#39;two&#39;];<br>
return $foo[&#39;one&#39;]?=C2=A0 [&#39;two&#39;];<br>
<br>
<br>
&gt; And here it gets interesting:<br>
&gt;<br>
&gt; =C2=A0 =C2=A0 return $foo[&#39;one&#39;][&#39;two&#39;]?[&#39;three=
9;];<br>
&gt; =C2=A0 =C2=A0 // is roughly equivalent to:<br>
&gt; =C2=A0 =C2=A0 if (!isset($foo[&#39;one&#39;]) {<br>
&gt; =C2=A0 =C2=A0 trigger_error(&#39;Undefined key &quot;one&quot; in ...&=
#39;, ...);<br>
&gt; =C2=A0 =C2=A0 }<br>
&gt; =C2=A0 =C2=A0 return $foo[&#39;one&#39;][&#39;two&#39;][&#39;three&#39=
;] ?? null;<br>
&gt;<br>
<br>
You can have the same ability with a ? inside the brackets, and without <br=
>
the ambiguity:<br>
<br>
// all three keys required to exist (as of PHP 9):<br>
return $foo[&#39;one&#39;][&#39;two&#39;][&#39;three&#39;];<br>
<br>
// &#39;one&#39; and &#39;two&#39; required, &#39;three&#39; optional:<br>
return $foo[&#39;one&#39;][&#39;two&#39;][?&#39;three&#39;];<br>
<br>
// both &#39;two&#39; and &#39;three&#39; optional:<br>
return $foo[&#39;one&#39;][?&#39;two&#39;][?&#39;three&#39;];<br>
<br>
// optional all the way:<br>
return $foo[?&#39;one&#39;][?&#39;two&#39;][?&#39;three&#39;];<br>
<br>
<br>
This would maybe have similar short-cut semantics to ?-&gt;<br>
<br>
$foo =3D [];<br>
return $foo[?&#39;one&#39;][&#39;two&#39;];<br>
// OK: the ? neutralises access to the missing section of the array, <br>
short-cutting to NULL<br>
// Similar to $foo?-&gt;bar-&gt;baz when $foo is NULL<br>
<br>
$foo =3D [&#39;one&#39; =3D&gt; []];<br>
return $foo[?&#39;one&#39;][&#39;two&#39;];<br>
// Error: the ? had no effect, so we&#39;re trying to access non-existent <=
br>
key &#39;two&#39; in the empty array $foo[&#39;one&#39;]<br>
// Similar to $foo?-&gt;bar-&gt;baz when $foo-&gt;bar exists<br>
<br>
<br>
But an RFC could pin down those details.<br>
<br>
<br>
-- <br>
Rowan Tommins<br>
[IMSoP]<br>
</blockquote></div>

--0000000000004d2bda064e1a9d10--