Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:114998
To: "internals@lists.php.net" <internals@lists.php.net>
Thread-Topic: [PHP-DEV] [PROPOSAL] Bare name array literals (again)
Thread-Index: AQHXZryCJm/mj5rNokqkQmATCwMJ5qseudGAgABma00=
Date: Tue, 22 Jun 2021 00:03:56 +0000
Message-ID:
 <DM6PR07MB66180AC527E2461D163F8D6DF90A9@DM6PR07MB6618.namprd07.prod.outlook.com>
References:
 <DM6PR07MB66186AD1AF40B256C5C3ABA0F90A9@DM6PR07MB6618.namprd07.prod.outlook.com>,<7f2da982-e29b-ccca-bddc-24ae7f4b0390@gmail.com>
In-Reply-To: <7f2da982-e29b-ccca-bddc-24ae7f4b0390@gmail.com>
Accept-Language: en-CA, en-US
Content-Language: en-CA
x-ms-exchange-messagesentrepresentingtype: 1
x-tmn: [4q6znVcOQeGLpYK8v90kzcPovVcesXSwXLBebulPYmzfImUO6/SAx/kJlnIJ08iS]
x-ms-publictraffictype: Email
x-incomingheadercount: 44
x-eopattributedmessage: 0
x-ms-office365-filtering-correlation-id: afeecdff-f476-46cc-b77b-08d935113ab9
x-ms-traffictypediagnostic: DM6NAM11HT165:
x-microsoft-antispam: BCL:0;
x-microsoft-antispam-message-info:
 T0ht9gGqxqFh9ZRL/vZnIF8xC9MNw/bj7PIoq5gp0me+D/bXyMk5U1tCyP+x/hTf1+eBbsS3jjDNSng58qoVEMS2+vnC+o+O5ikFSrf9FNMF+7Y+lHCyhbwgZE9X9NamRf5y2zq256wjg9PSBeSJg3NWcOpOlngEfUP7Rfelvdr2j39RqdQbLw2rOqP4OKKZltVSTMRH4LzcfGLPuA994jauysP+8796xVBDJnqlsbR4uvoqgoswZYjNIgG9XOCWClwo3PfKB017dcE1XFbJCj/g8cylrQYbAaqr+qAhyQmw28EOJ/Lsmk+68Pnp8u3q1oZjyAn/27/G02H1utvWEiW7oMbABgXqgPOlUMbw7np2K094Y1pExCIc7gnE3GslNXI8Cjc9/CcdjBPS90vuyX9V+DYQI0M5sdSI5kjOPqF4wbu2Z24Wg/akDWWs9p9ln++K2gRn2CFCgBeqJErXXzZsLXOoK0gvp0eWxSRfir4=
x-ms-exchange-antispam-messagedata:
 jSaRcEzGdROGJ6+rLBtScCKhrhdUK5+HHy/J465voTuISivbPE8Kh/q69T3Hp3m/Ktd6QvmBGRGipqWC4W3AtelzakKWvaBMKZVw0TvFSmnFQro+anFzu1PhXxsIewMVm9AeaBaRUdbcR4UYGtrr5sA1y9k0nOpnm6aIcAnaG+dXOJlebBNjPV90coWmunzq+tohxQtfMAmyE3TGEim4uw==
x-ms-exchange-transport-forked: True
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: hotmail.com
X-MS-Exchange-CrossTenant-AuthAs: Anonymous
X-MS-Exchange-CrossTenant-AuthSource: DM6NAM11FT060.eop-nam11.prod.protection.outlook.com
X-MS-Exchange-CrossTenant-RMS-PersistedConsumerOrg: 00000000-0000-0000-0000-000000000000
X-MS-Exchange-CrossTenant-Network-Message-Id: afeecdff-f476-46cc-b77b-08d935113ab9
X-MS-Exchange-CrossTenant-originalarrivaltime: 22 Jun 2021 00:03:56.4812
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Internet
X-MS-Exchange-CrossTenant-id: 84df9e7f-e9f6-40af-b435-aaaaaaaaaaaa
X-MS-Exchange-CrossTenant-rms-persistedconsumerorg: 00000000-0000-0000-0000-000000000000
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM6NAM11HT165
Subject: Re: [PHP-DEV] [PROPOSAL] Bare name array literals (again)
From: tysonandre775@hotmail.com (tyson andre)

Hi Rowan Tommins,=0A=
=0A=
> This is an immediate "no" from me: it multiplies the ways to write the=0A=
> same thing from 2 to 4, in order to save a few bytes, in a few instances.=
=0A=
=0A=
It's from 4 to 6 if you include single quoted strings vs double quoted stri=
ngs.=0A=
If linters and automatic fixers were set up for a project to enforce it (e.=
g. phpcbf),=0A=
there would only be one way that literals would be used in a project either=
 way.=0A=
=0A=
> I think this is something that Douglas Crockford got absolutely right=0A=
> when he simplified JavaScript object syntax to formalise JSON: every=0A=
> valid key can be represented as a quoted string, so if the quotes are=0A=
> always there, you don't need to remember a list of rules about reserved=
=0A=
> words, allowed characters, etc.=0A=
=0A=
It's the same rules as parts of identifiers or variable names.=0A=
I don't think there are any reserved words for named params or this proposa=
l(`default`, etc. are allowed despite being keywords)=0A=
=0A=
What makes sense for a serialization format which will have hundreds of enc=
oders/decoders may not make sense for a programming language.=0A=
=0A=
> > This is useful for shortening long arrays where the keys are known lite=
rals,=0A=
> > e.g.=0A=
> > =0A=
> > return [success: true, data: $data, cursor: $cursor];=0A=
> > // is equivalent to the following, but shorter:=0A=
> > return ['success' =3D> true, 'data' =3D> $data, 'cursor' =3D> $cursor];=
=0A=
> =0A=
> Although common, this is not a good use of arrays; if your keys are=0A=
> "known literals", they should be fields of some object:=0A=
> =0A=
> return new Result(success: true, data: $data, cursor: $cursor);=0A=
=0A=
If there is only a single place where an array with those keys is returned,=
=0A=
or if dynamic properties get added later,=0A=
then creating a class may not be worth it.=0A=
=0A=
Developers would have to switch between files and keep track of which class=
es were =0A=
ordinary data storage and which had side effects or transformed parameters.=
=0A=
=0A=
> If you don't want to declare a class (yet), you can use an anonymous=0A=
> object. Rather than yet another way to write arrays, it would be great=0A=
> to have some more powerful syntax for those; currently you'd have=0A=
> something like this:=0A=
> =0A=
> return new class(success: true, data: $data, cursor: $cursor) { public=0A=
> function __construct(public bool $success, public array $data, public=0A=
> CursorInterface $cursor) {} };=0A=
> Brainstorming, we could perhaps extend property promotion into the "new=
=0A=
> class" clause, so that you could write this:=0A=
> =0A=
> return new class(public bool success: true, public array data: $data,=0A=
> public CursorInterface cursor: $cursor) {};=0A=
=0A=
I think anonymous objects would benefit some use cases but would not be use=
ful for every use case (e.g. short single-file shell scripts),=0A=
though the typed properties and constructor property promotion are definite=
ly convenient.=0A=
=0A=
Also, with no common interface between those anonymous classes,=0A=
using just anonymous classes would be writing functions that accept any obj=
ect or return any object, =0A=
which would be error prone and hard to analyze.=0A=
=0A=
Those anonymous classes wouldn't have any ancestors in common for `instance=
of` checks.=0A=
Also, if there were optional parameters, that can be represented in non-sta=
ndard JSDoc supported by static analyzers=0A=
(`success: bool, data?: array, cursor?: CursorInterface, errorMessage?: str=
ing`),=0A=
but that wouldn't be represented in an anonymous class=0A=
(setting a property to null is not the same as omitting it).=0A=
=0A=
This reminds me of https://docs.python.org/3/library/collections.html#colle=
ctions.namedtuple=0A=
but I don't plan to propose that (something similar could be done by valida=
ting all property names are valid, =0A=
getting the sha1 of the ordered list of property names to choose a class na=
me, =0A=
and optionally setting properties to `readonly` (if the RFC passes) and for=
bidding dynamic properties,=0A=
and extending some common interface).=0A=
=0A=
Thanks,=0A=
Tyson=0A=