Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:123927
Feedback-ID: ifab94697:Fastmail
User-Agent: Cyrus-JMAP/3.11.0-alpha0-538-g1508afaa2-fm-20240616.001-g1508afaa
Precedence: bulk
MIME-Version: 1.0
Message-ID: <0802b731-cbe6-4353-b554-b57b9f74a2c5@app.fastmail.com>
In-Reply-To: 
 <CAPyj-LDD01-u9TT6CNvvH1-9qnxM9WxMqHzWFmY11Ys6SxF2Aw@mail.gmail.com>
References: <2a6b92eb-d5e9-4a1a-9548-a068ac42ebd2@app.fastmail.com>
 <CAG4FTXz6cH26C5H51VmTGnpFK5XxROFbW4yDzwAw2v1gcQWGgA@mail.gmail.com>
 <CAPzBOBOTSCCKDR4pDocr4xW+oEXJ27YizgPLkn1PGR9nyPU7ww@mail.gmail.com>
 <1E295280-619B-4490-B53C-0899B64F9215@chaz.works>
 <CAPzBOBPSSEmEbPj1zLcfGOOG7voo7YMUO16NjXzpTy7Wfx4ycw@mail.gmail.com>
 <CAPyj-LAM7yPZpcArjHYEG8HhNRR-hA8sgzAjXT7Wf8X4mVgmQA@mail.gmail.com>
 <97a93ae2-5202-47eb-bf51-ec1e976ea765@app.fastmail.com>
 <CAPyj-LC49o=RVXigTrHte8EWSmwxPKqT4tyRGWpGf022PDCk8Q@mail.gmail.com>
 <c72033aa-7d55-4560-95ac-6c81c21cdc43@app.fastmail.com>
 <78D0FFF7-3867-4D71-B0C0-FA23E5121C39@rwec.co.uk>
 <10390963-0c2c-441a-bcce-20a0433948c5@app.fastmail.com>
 <d9711524-6818-4148-873a-4f27c32a270e@varteg.nz>
 <8498ecf7-f431-4d68-aaf1-dc4cfb00241b@app.fastmail.com>
 <CAPyj-LDD01-u9TT6CNvvH1-9qnxM9WxMqHzWFmY11Ys6SxF2Aw@mail.gmail.com>
Date: Thu, 27 Jun 2024 08:57:22 +0200
To: internals@lists.php.net
Subject: Re: [PHP-DEV] [Early Feedback] Pattern matching
Content-Type: multipart/alternative;
 boundary=8bb83fd6aaeb44a88e5cd970954c4a46
From: rob@bottled.codes ("Rob Landers")

--8bb83fd6aaeb44a88e5cd970954c4a46
Content-Type: text/plain;charset=utf-8
Content-Transfer-Encoding: quoted-printable

On Wed, Jun 26, 2024, at 23:42, Ilija Tovilo wrote:
>=20
> function test($value) {
>     if ($value is ['foo' =3D> ?string]) {
>         $foo =3D $value['foo'];
>     }
> }
> test([]);

(Scroll to the end if you don=E2=80=99t care about my rebuttal)

I mean, there=E2=80=99s nothing wrong with this code, $foo will be null =
at the end.

>=20
> With your approach, the example above would emit a warning, even
> though the context within test() doesn't look like it should.=20

That happens every time I get one of these warnings. It=E2=80=99s nice t=
o get the heads up that I didn=E2=80=99t think through something and nee=
d to either A) figure out how it got there, or B) provide a default =E2=80=
=94 even if it is null =E2=80=94 to make the warning go away.=20

> If ?string means that the index might or might not exist, all code that
> accesses them must check for existence, even when not needing to
> handle null itself. That doesn't seem desirable to me.

This is true already. since this warning and null-coalescence, I can pro=
bably count on one hand the number of times I=E2=80=99ve *not* written s=
omething like ?? null beside an array access (though I prefer throwing a=
n exception if appropriate).=20

>=20
> I also think the issue goes further. If anything|null means that the
> offset might not exist, does that include mixed? That makes ['foo' =3D>
> mixed] essentially useless.

In thinking about it some more, for lists, it would be nice if there wer=
e two modes:
 1. regular mode: $a is [1,2,3] where it must match exactly.=20
 2. set mode: $a is set([1,2,3]) where order doesn=E2=80=99t matter, onl=
y that $a contains at least one of every given element.=20
=E2=80=94=E2=80=94=E2=80=94=E2=80=94=E2=80=94=E2=80=94

In other news, I finally found a realistic example where there is a diff=
erence between non-existence and null.

The scenario is this:

Imagine you accept a callback and you prepare an array to pass as parame=
ters. In this case, you call the callback like so:

$callback(=E2=80=A6$args);

Today, you either have to yolo it and hope everything is perfect or meti=
culously check the args to make sure the types match your documented cal=
lback signature.

If a key in $args doesn=E2=80=99t exist, it might be an error (or defaul=
t value provided by the callback implementation) but if it is null, it m=
ight also be disastrously incorrect because the default value won=E2=80=99=
t be used.=20

By using pattern matching, we can check the structure of the array for n=
on-existence or if exists, the documented type. This is where matching a=
 list exactly is important (if using positional args).

So, now that there is a use case for it, we can work out expected behavi=
ors of lists and associative arrays and nullability for both.

Anyway, does anyone have a suggestion for what kind of hat I should have=
 for dinner?

=E2=80=94 Rob
--8bb83fd6aaeb44a88e5cd970954c4a46
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 Wed, Jun 26,=
 2024, at 23:42, Ilija Tovilo wrote:<br></div><blockquote type=3D"cite" =
id=3D"qt" style=3D""><div><br></div><div>function test($value) {<br></di=
v><div>&nbsp;&nbsp;&nbsp; if ($value is ['foo' =3D&gt; ?string]) {<br></=
div><div>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; $foo =3D $value['foo=
'];<br></div><div>&nbsp;&nbsp;&nbsp; }<br></div><div>}<br></div><div>tes=
t([]);<br></div></blockquote><div><br></div><div>(Scroll to the end if y=
ou don=E2=80=99t care about my rebuttal)</div><div><br></div><div>I mean=
, there=E2=80=99s nothing wrong with this code, $foo will be null at the=
 end.<br></div><div><br></div><blockquote type=3D"cite" id=3D"qt" style=3D=
""><div><br></div><div>With your approach, the example above would emit =
a warning, even<br></div><div>though the context within test() doesn't l=
ook like it should. <br></div></blockquote><div><br></div><div>That happ=
ens every time I get one of these warnings. It=E2=80=99s nice to get the=
 heads up that I didn=E2=80=99t think through something and need to eith=
er A) figure out how it got there, or B) provide a default =E2=80=94 eve=
n if it is null =E2=80=94 to make the warning go away.&nbsp;</div><div><=
br></div><blockquote type=3D"cite" id=3D"qt" style=3D""><div>If ?string =
means that the index might or might not exist, all code that<br></div><d=
iv>accesses them must check for existence, even when not needing to<br><=
/div><div>handle null itself. That doesn't seem desirable to me.<br></di=
v></blockquote><div><br></div><div>This is true already. since this warn=
ing and null-coalescence, I can probably count on one hand the number of=
 times I=E2=80=99ve&nbsp;<b>not</b> written something like ?? null besid=
e an array access (though I prefer throwing an exception if appropriate)=
.&nbsp;</div><div><br></div><blockquote type=3D"cite" id=3D"qt" style=3D=
""><div><br></div><div>I also think the issue goes further. If anything|=
null means that the<br></div><div>offset might not exist, does that incl=
ude mixed? That makes ['foo' =3D&gt;<br></div><div>mixed] essentially us=
eless.<br></div></blockquote><div><br></div><div>In thinking about it so=
me more, for lists, it would be nice if there were two modes:<br></div><=
ol><li>regular mode: $a is [1,2,3] where it must match exactly.&nbsp;<br=
></li><li>set mode: $a is set([1,2,3]) where order doesn=E2=80=99t matte=
r, only that $a contains at least one of every given element.&nbsp;<br><=
/li></ol><div>=E2=80=94=E2=80=94=E2=80=94=E2=80=94=E2=80=94=E2=80=94<br>=
</div><div><br></div><div>In other news, I finally found a realistic exa=
mple where there is a difference between non-existence and null.<br></di=
v><div><br></div><div>The scenario is this:<br></div><div><br></div><div=
>Imagine you accept a callback and you prepare an array to pass as param=
eters. In this case, you call the callback like so:<br></div><div><br></=
div><div>$callback(=E2=80=A6$args);<br></div><div><br></div><div>Today, =
you either have to yolo it and hope everything is perfect or meticulousl=
y check the args to make sure the types match your documented callback s=
ignature.<br></div><div><br></div><div>If a key in $args doesn=E2=80=99t=
 exist, it might be an error (or default value provided by the callback =
implementation) but if it is null, it might also be disastrously incorre=
ct because the default value won=E2=80=99t be used.&nbsp;<br></div><div>=
<br></div><div>By using pattern matching, we can check the structure of =
the array for non-existence or if exists, the documented type. This is w=
here matching a list exactly is important (if using positional args).<br=
></div><div><br></div><div>So, now that there is a use case for it, we c=
an work out expected behaviors of lists and associative arrays and nulla=
bility for both.<br></div><div><br></div><div>Anyway, does anyone have a=
 suggestion for what kind of hat I should have for dinner?<br></div><div=
><br></div><div id=3D"sig121229152">=E2=80=94 Rob<br></div></body></html=
>
--8bb83fd6aaeb44a88e5cd970954c4a46--