Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:124131
Feedback-ID: ifab94697:Fastmail
User-Agent: Cyrus-JMAP/3.11.0-alpha0-566-g3812ddbbc-fm-20240627.001-g3812ddbb
Precedence: bulk
MIME-Version: 1.0
Message-ID: <a69a759f-f185-42ff-b769-1f85beab673b@app.fastmail.com>
In-Reply-To: <5BA404A4-7281-475F-A1F6-D1252ABB059D@miles.systems>
References: <07A8534A-1B45-4F15-A78B-AA7DDF92B8B6@koalephant.com>
 <5BA404A4-7281-475F-A1F6-D1252ABB059D@miles.systems>
Date: Mon, 01 Jul 2024 08:55:40 +0200
To: "Richard Miles" <richard@miles.systems>,
 "php internals" <internals@lists.php.net>
Cc: "Stephen Reay" <php-lists@koalephant.com>
Subject: Re: [PHP-DEV] [Initial Feedback] Typed Arrays
Content-Type: multipart/alternative;
 boundary=40ed06e018374f6090096fd0a1f9bcaf
From: rob@bottled.codes ("Rob Landers")

--40ed06e018374f6090096fd0a1f9bcaf
Content-Type: text/plain;charset=utf-8
Content-Transfer-Encoding: quoted-printable



On Mon, Jul 1, 2024, at 05:05, Richard Miles wrote:
> Howdy Stephen,=20
>=20
>>>=20
>>>=20
>>> It seems like everyone always gets really hung up on generics! That =
or how to syntactically implement this.
>>=20
>> Hi
>>=20
>> Given other improvements in the language, I don't really see much nee=
d for the ability to specify fixed keys and their types (ie associative =
arrays) (I assume that's what the "['a' =3D> string]" syntax is meant to=
 be?)
>>=20
>> Is there some fundamental aspect of the original arrayof RFC that pre=
vents it from being revived today? The whole "does it include nulls" par=
t is moot now presumably because we have both ? And |null, and the Type[=
] syntax is very common in docblocks.
>=20
> I appreciate your feedback, and I think it=E2=80=99s valid. I try to c=
ompare my expectations to what is possible in typescript.
> This comparison, while not a perfect match, is a great source of learn=
ing and inspiration.=20
>=20
> While it=E2=80=99s true that PHP can take arrays and mash them into ob=
jects, I have to ask myself if this other language has this=20
> feature, why don't we? Is it because we don=E2=80=99t care? Is it a ba=
d feature? I don't think so.=20
> I have a PR for WebSockets that integrates with Apache. Why hasn=E2=80=
=99t it been done before?
> I don=E2=80=99t think this is an oversight on others, but rather, the =
community is what makes PHP better, challenging norms.=20
>=20
> I=E2=80=99m not saying it will outperform what objectiveness currently=
 provides, but I'm also not saying it won=E2=80=99t! I don=E2=80=99t thi=
nk there=E2=80=99s a negative to=20
> me implementing this syntax, and who knows, maybe it will be awesome f=
or performance. I think what we should consider is that other=20
> languages have this, and I feel the community wants this but is hung u=
p on generics, which may have been an oversight.?  I=E2=80=99m pretty ne=
w=20
> here, and I=E2=80=99d love to make a meaningful impact, so the fact th=
at this idea has been brought up before is encouraging.=20
>=20
> Take, for example:
>=20
> interface iArrayA ['a' =3D> string, 'b' =3D> string, 'c' =3D> ?string,=
 =E2=80=98d=E2=80=99 =3D>  SomeClass, =E2=80=98e=E2=80=99=3D>  iArrayA, =
=E2=80=98f=E2=80=99 =3D> mixed ]
>=20
> $a =3D new Dict<string, iArrayA>;=20
>=20
>=20
> Maybe the signature for Dict is:
>=20
> class Dict<R is string | int, S is mixed> {}
>=20
> Everyone harps on Generics, but no one can say how I'm supposed to mod=
el iArrayA if we focus on Generics.=20
> We cannot safely define this now, as iArrayA is not a valid interface.=
 I believe that is what I'm advocating for alone. =20
> Array interfaces.=20
>=20
> Best,=20
> Richard Miles

Hello,

> I appreciate your feedback, and I think it=E2=80=99s valid. I try to c=
ompare my expectations to what is possible in typescript.

I feel like Typescript is the wrong model to follow. The type system in =
Typescript is Turing Complete (https://github.com/microsoft/TypeScript/i=
ssues/14833) and php barely has a type system. It doesn=E2=80=99t even h=
ave consistent type checking, for that matter (properties are checked di=
fferently than arguments which are checked differently than constructor =
arguments =E2=80=94 and the fact they agree with each other at all, is i=
nteresting, because they don=E2=80=99t arrive at the same conclusions in=
 the same way. I=E2=80=99ve been dealing with subtle bugs here for weeks=
 on an unannounced RFC implementation) and this is why generics is a can=
 of worms: implementing it would essentially force a rewrite of the type=
 system into a proper type system.

In my honest opinion, before we can even have this type of conversation,=
 we need to have an RFC about the syntax (there may already be one, I ha=
ven=E2=80=99t checked =E2=80=94 and internet is spotty where I currently=
 am). The pattern matching RFC was proposed as though it would be writte=
n one way, people have thrown out ideas of different syntax, but I don=E2=
=80=99t think there is an =E2=80=9Cofficial=E2=80=9D syntax. A good star=
ting point may be to simply propose an RFC about syntax so future RFCs c=
an be consistent.=20

=E2=80=94 Rob
--40ed06e018374f6090096fd0a1f9bcaf
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><br></div><div>=
<br></div><div>On Mon, Jul 1, 2024, at 05:05, Richard Miles wrote:<br></=
div><blockquote type=3D"cite" id=3D"qt" style=3D"overflow-wrap:break-wor=
d;line-break:after-white-space;"><div>Howdy Stephen,&nbsp;<br></div><div=
><div><br></div><blockquote type=3D"cite"><div><div dir=3D"auto"><div di=
r=3D"ltr"><blockquote type=3D"cite"><div dir=3D"ltr"><div><div style=3D"=
display:block;"><div style=3D"display:block;"><br></div></div></div><div=
><br></div><div>It seems like everyone always gets really hung up on gen=
erics! That or how to syntactically implement this.<br></div></div></blo=
ckquote><div><br></div><div>Hi<br></div><div><br></div><div>Given other =
improvements in the language, I don't really see much need for the abili=
ty to specify fixed keys and their types (ie associative arrays) (I assu=
me that's what the "['a' =3D&gt; string]" syntax is meant to be?)<br></d=
iv><div><br></div><div>Is there some fundamental aspect of the original =
arrayof RFC that prevents it from being revived today? The whole "does i=
t include nulls" part is moot now presumably because we have both ? And =
|null, and the Type[] syntax is very common in docblocks.<br></div></div=
></div></div></blockquote></div><div><br></div><div>I appreciate your fe=
edback, and I think it=E2=80=99s valid. I try to compare my expectations=
 to what is possible in typescript.<br></div><div>This comparison, while=
 not a perfect match, is a great source of learning and inspiration.&nbs=
p;<br></div><div><br></div><div>While it=E2=80=99s true that PHP can tak=
e arrays and mash them into objects, I have to ask myself if this other =
language has this&nbsp;<br></div><div>feature, why don't we? Is it becau=
se we don=E2=80=99t care? Is it a bad feature? I don't think so.&nbsp;<b=
r></div><div>I have a PR for WebSockets that integrates with Apache. Why=
 hasn=E2=80=99t it been done before?<br></div><div>I don=E2=80=99t think=
 this is an oversight on others, but rather, the community is what makes=
 PHP better, challenging norms.&nbsp;<br></div><div><br></div><div>I=E2=80=
=99m not saying it will outperform what objectiveness currently provides=
, but I'm also not saying it won=E2=80=99t! I don=E2=80=99t think there=E2=
=80=99s a negative to&nbsp;<br></div><div>me implementing this syntax, a=
nd who knows, maybe it will be awesome for performance. I think what we =
should consider is that other&nbsp;<br></div><div>languages have this, a=
nd I feel the community wants this but is hung up on generics, which may=
 have been an oversight.? &nbsp;I=E2=80=99m pretty new&nbsp;<br></div><d=
iv>here, and I=E2=80=99d love to make a meaningful impact, so the fact t=
hat this idea has been brought up before is encouraging.&nbsp;<br></div>=
<div><br></div><div>Take, for example:<br></div><div><br></div><div><div=
 style=3D"color:rgb(0, 0, 0);">interface iArrayA ['a' =3D&gt; string, 'b=
' =3D&gt; string, 'c' =3D&gt; ?string, =E2=80=98d=E2=80=99 =3D&gt; &nbsp=
;SomeClass, =E2=80=98e=E2=80=99=3D&gt; &nbsp;iArrayA, =E2=80=98f=E2=80=99=
 =3D&gt; mixed ]<br></div></div><div><br></div><div>$a =3D new Dict&lt;s=
tring,&nbsp;<span style=3D"color:rgb(0, 0, 0);">iArrayA&gt;;&nbsp;</span=
><br></div><div><br></div><div><br></div><div><span style=3D"color:rgb(0=
, 0, 0);">Maybe the signature for Dict is:</span><br></div><div><span st=
yle=3D"color:rgb(0, 0, 0);"></span><br></div><div><span style=3D"color:r=
gb(0, 0, 0);">class Dict&lt;R is string | int, S is mixed&gt; {}</span><=
br></div><div><span style=3D"color:rgb(0, 0, 0);"></span><br></div><div>=
<span style=3D"color:rgb(0, 0, 0);">Everyone harps on Generics, but no o=
ne can say how I'm supposed to model&nbsp;</span><span style=3D"color:rg=
b(0, 0, 0);">iArrayA if we focus on Generics.&nbsp;</span><br></div><div=
><span style=3D"color:rgb(0, 0, 0);">We cannot safely define this now, a=
s&nbsp;</span><span style=3D"color:rgb(0, 0, 0);">iArrayA is not a valid=
 interface. I believe that is what I'm advocating for alone. &nbsp;</spa=
n><br></div><div>Array interfaces.&nbsp;<br></div><div><br></div><div><s=
pan class=3D"color" style=3D"color:rgb(0, 0, 0);"><span style=3D"">Best,=
&nbsp;</span></span><br></div><div><span class=3D"color" style=3D"color:=
rgb(0, 0, 0);"><span style=3D"">Richard Miles</span></span><br></div></b=
lockquote><div><br></div><div>Hello,</div><div><br></div><blockquote typ=
e=3D"cite"><div>I appreciate your feedback, and I think it=E2=80=99s val=
id. I try to compare my expectations to what is possible in typescript.<=
br></div></blockquote><div><br></div><div>I feel like Typescript is the =
wrong model to follow. The type system in Typescript is Turing Complete =
(<a href=3D"https://github.com/microsoft/TypeScript/issues/14833">https:=
//github.com/microsoft/TypeScript/issues/14833</a>) and php barely has a=
 type system. It doesn=E2=80=99t even have consistent type checking, for=
 that matter (properties are checked differently than arguments which ar=
e checked differently than constructor arguments =E2=80=94 and the fact =
they agree with each other at all, is interesting, because they don=E2=80=
=99t arrive at the same conclusions in the same way. I=E2=80=99ve been d=
ealing with subtle bugs here for weeks on an unannounced RFC implementat=
ion) and this is why generics is a can of worms: implementing it would e=
ssentially force a rewrite of the type system into a proper type system.=
<br></div><div><br></div><div>In my honest opinion, before we can even h=
ave this type of conversation, we need to have an RFC about the syntax (=
there may already be one, I haven=E2=80=99t checked =E2=80=94 and intern=
et is spotty where I currently am). The pattern matching RFC was propose=
d as though it would be written one way, people have thrown out ideas of=
 different syntax, but I don=E2=80=99t think there is an =E2=80=9Coffici=
al=E2=80=9D syntax. A good starting point may be to simply propose an RF=
C about syntax so future RFCs can be consistent.&nbsp;</div><div><br></d=
iv><div id=3D"sig121229152">=E2=80=94 Rob<br></div></body></html>
--40ed06e018374f6090096fd0a1f9bcaf--