Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:126034
Feedback-ID: ifab94697:Fastmail
Precedence: bulk
MIME-Version: 1.0
Date: Sat, 23 Nov 2024 16:43:31 +0100
To: "Eugene Sidelnyk" <zsidelnik@gmail.com>
Cc: internals@lists.php.net
Message-ID: <b8f62fe1-402c-404e-8079-eb8ea29a07db@app.fastmail.com>
In-Reply-To: 
 <CAM0Ccxh8Hb3+JCwhkDR_85gnx7ZECpQBWhrTW_z7EopniHW+sg@mail.gmail.com>
References: <18b85ba5-5f1c-489c-9096-3ae203977fbe@app.fastmail.com>
 <CAM0Ccxh8Hb3+JCwhkDR_85gnx7ZECpQBWhrTW_z7EopniHW+sg@mail.gmail.com>
Subject: Re: [PHP-DEV] [RFC] Data Classes
Content-Type: multipart/alternative;
 boundary=12fcd0f9302b42cfb5b3d02857d281dd
From: rob@bottled.codes ("Rob Landers")

--12fcd0f9302b42cfb5b3d02857d281dd
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

On Sat, Nov 23, 2024, at 16:02, Eugene Sidelnyk wrote:
> Hello
>=20
> If I remember correctly, the whole concept of "value" is fully describ=
ed in DDD book by Eric Evans. If that's the point of the RFC, I wonder o=
f there's any point in not making such classes immutable by default, and=
 to keep only one instance of value object unique per given set of prope=
rties in memory, thereby eliminating cloning altogether and optimizing t=
he memory usage.
>=20
>=20
> On Sat, Nov 23, 2024, 3:12 PM Rob Landers <rob@bottled.codes> wrote:
>> __
>> Hello internals,
>>=20
>> Born from the Records RFC (https://wiki.php.net/rfc/records) discussi=
on, I would like to introduce to you a competing RFC: Data Classes (http=
s://wiki.php.net/rfc/dataclass).=20
>>=20
>> This adds a new class modifier: data. This modifier drastically chang=
es how classes work, making them comparable by value instead of referenc=
e, and any mutations behave more like arrays than objects (by vale). If =
desired, it can be combined with other modifiers, such as readonly, to e=
nforce immutability.
>>=20
>> I've been playing with this feature for a few days now, and it is sur=
prisingly intuitive to use. There is a (mostly) working implementation a=
vailable on GitHub (https://github.com/php/php-src/pull/16904) if you wa=
nt to have a go at it.
>>=20
>> Example:
>>=20
>> data class UserId { public function __construct(public int $id) {} }
>>=20
>> $user =3D new UserId(12);
>> // later
>> $admin =3D new UserId(12);
>> if ($admin =3D=3D=3D $user) { // do something } // true
>>=20
>> Data classes are true value objects, with full copy-on-write optimiza=
tions:
>>=20
>> data class Point {
>>   public function __construct(public int $x, public int $y) {}
>>   public function add(Point $other): Point {
>>     // illustrating value semantics, no copy yet
>>     $previous =3D $this;
>>     // a copy happens on the next line
>>     $this->x =3D $this->x + $other->x;
>>     $this->y =3D $this->y + $other->y;
>>     assert($this !=3D=3D $previous); // passes
>>     return $this;
>>   }
>> }
>>=20
>> I think this would be an amazing addition to PHP.=20
>>=20
>> Sincerely,
>>=20
>> =E2=80=94 Rob

Hello!

Don't forget to bottom-post!

> I wonder of there's any point in not making such classes immutable by =
default, and to keep only one instance of value object unique per given =
set of properties in memory, thereby eliminating cloning altogether and =
optimizing the memory usage.

This was the entire point of the records RFC ;) it was immutable by defa=
ult, but people were wondering what it would look like if it were more '=
composable' vs. 'dedicated syntax'.

=E2=80=94 Rob
--12fcd0f9302b42cfb5b3d02857d281dd
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 Sat, Nov 23,=
 2024, at 16:02, Eugene Sidelnyk wrote:<br></div><blockquote type=3D"cit=
e" id=3D"qt" style=3D""><p dir=3D"ltr">Hello<br></p><p dir=3D"ltr">If I =
remember correctly, the whole concept of "value" is fully described in D=
DD book by Eric Evans. If that's the point of the RFC, I wonder of there=
's any point in not making such classes immutable by default, and to kee=
p only one instance of value object unique per given set of properties i=
n memory, thereby eliminating cloning altogether and optimizing the memo=
ry usage.<br></p><div><br></div><div class=3D"qt-gmail_quote"><div dir=3D=
"ltr" class=3D"qt-gmail_attr">On Sat, Nov 23, 2024, 3:12 PM Rob Landers =
&lt;rob@bottled.codes&gt; wrote:<br></div><blockquote class=3D"qt-gmail_=
quote" style=3D"margin-top:0px;margin-right:0px;margin-bottom:0px;margin=
-left:0.8ex;border-left-width:1px;border-left-style:solid;border-left-co=
lor:rgb(204, 204, 204);padding-left:1ex;"><div><u></u><br></div><div><di=
v>Hello internals,<br></div><div><br></div><div>Born from the Records RF=
C (<a href=3D"https://wiki.php.net/rfc/records" target=3D"_blank" rel=3D=
"noreferrer">https://wiki.php.net/rfc/records</a>) discussion, I would l=
ike to introduce to you a competing RFC: Data Classes (<a href=3D"https:=
//wiki.php.net/rfc/dataclass" target=3D"_blank" rel=3D"noreferrer">https=
://wiki.php.net/rfc/dataclass</a>).&nbsp;<br></div><div><br></div><div>T=
his adds a new class modifier: data. This modifier drastically changes h=
ow classes work, making them comparable by value instead of reference, a=
nd any mutations behave more like arrays than objects (by vale). If desi=
red, it can be combined with other modifiers, such as readonly, to enfor=
ce immutability.<br></div><div><br></div><div>I've been playing with thi=
s feature for a few days now, and it is surprisingly intuitive to use. T=
here is a (mostly) working implementation available on GitHub (<a href=3D=
"https://github.com/php/php-src/pull/16904" target=3D"_blank" rel=3D"nor=
eferrer">https://github.com/php/php-src/pull/16904</a>) if you want to h=
ave a go at it.<br></div><div><br></div><div>Example:<br></div><div><br>=
</div><div>data class UserId { public function __construct(public int $i=
d) {} }<br></div><div><br></div><div>$user =3D new UserId(12);<br></div>=
<div>// later<br></div><div>$admin =3D new UserId(12);<br></div><div>if =
($admin =3D=3D=3D $user) { // do something } // true<br></div><div><br><=
/div><div>Data classes are true value objects, with full copy-on-write o=
ptimizations:<br></div><div><br></div><div>data class Point {<br></div><=
div>&nbsp; public function __construct(public int $x, public int $y) {}<=
br></div><div>&nbsp; public function add(Point $other): Point {<br></div=
><div>&nbsp; &nbsp; // illustrating value semantics, no copy yet<br></di=
v><div>&nbsp; &nbsp; $previous =3D $this;<br></div><div>&nbsp; &nbsp; //=
 a copy happens on the next line<br></div><div>&nbsp; &nbsp; $this-&gt;x=
 =3D $this-&gt;x + $other-&gt;x;<br></div><div>&nbsp; &nbsp; $this-&gt;y=
 =3D $this-&gt;y + $other-&gt;y;<br></div><div>&nbsp; &nbsp; assert($thi=
s !=3D=3D $previous); // passes<br></div><div>&nbsp; &nbsp; return $this=
;<br></div><div>&nbsp; }<br></div><div>}<br></div><div><br></div><div>I =
think this would be an amazing addition to PHP.&nbsp;<br></div><div><br>=
</div><div>Sincerely,<br></div><div><br></div><div id=3D"qt-m_-319920756=
4271075714sig121229152">=E2=80=94 Rob<br></div></div></blockquote></div>=
</blockquote><div><br></div><div>Hello!</div><div><br></div><div>Don't f=
orget to bottom-post!<br></div><div><br></div><blockquote type=3D"cite">=
<div>I wonder of there's any point in not making such classes immutable =
by default, and to keep only one instance of value object unique per giv=
en set of properties in memory, thereby eliminating cloning altogether a=
nd optimizing the memory usage.<br></div></blockquote><div><br></div><di=
v>This was the entire point of the records RFC ;) it was immutable by de=
fault, but people were wondering what it would look like if it were more=
 'composable' vs. 'dedicated syntax'.</div><div><br></div><div id=3D"sig=
121229152">=E2=80=94 Rob<br></div></body></html>
--12fcd0f9302b42cfb5b3d02857d281dd--