Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:115450
Message-ID: <96778CB3-C753-45B9-971E-7D84E6F81381@newclarity.net>
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_DC8F3ED2-39F5-4267-8C7A-E6CC4BC8C97A"
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.120.23.2.7\))
Date: Sat, 17 Jul 2021 05:04:37 -0400
In-Reply-To: <CAM0Ccxij4xLPu4Mcd7-XPA92t5H=-PeOqUhzMsF3=APTRiB=PQ@mail.gmail.com>
Cc: PHP Internals <internals@lists.php.net>
To: Eugene Sidelnyk <zsidelnik@gmail.com>
References: <CAM0CcxjJYYRrdzNjm285OQ6zDS8J7tqi0kLBeTGAP-2JXyqrDw@mail.gmail.com>
 <CA+hqhpeWACWYp=VNj0=oWkpPa++uiRF+=VmN34ZVBUu1dQOXPw@mail.gmail.com>
 <E2359773-7E55-4A87-9136-D43AA3A0E52F@newclarity.net>
 <CAM0Ccxij4xLPu4Mcd7-XPA92t5H=-PeOqUhzMsF3=APTRiB=PQ@mail.gmail.com>
Subject: Re: [PHP-DEV] Readonly properties - immutability by default
From: mike@newclarity.net (Mike Schinkel)

--Apple-Mail=_DC8F3ED2-39F5-4267-8C7A-E6CC4BC8C97A
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

> On Jul 17, 2021, at 1:40 AM, Eugene Sidelnyk <zsidelnik@gmail.com> =
wrote:
>=20
> > Having a "readonly" class where the properties are read-only by =
default makes sense to me, but maybe the better way to do it is with an =
attribute?
>=20
> We already have such an attribute provided. It is called =
[`#[Immutable]`](https://blog.jetbrains.com/phpstorm/2020/10/phpstorm-2020=
-3-eap-4/#immutable =
<https://blog.jetbrains.com/phpstorm/2020/10/phpstorm-2020-3-eap-4/#immuta=
ble>).

Actually, you are confusing PhpStorm's inclusion of PhpStorm-specific =
userland attributes where the PhpStorm team are probably trying to drive =
defacto-standard meanings for those attributes they documents with =
attributes that has specific meanings and behavior recognized by PHP =
itself.

Nothing against PhpStorm =E2=80=94 it is my IDE/editor of choice and I =
applaud JetBrain's leadership in this area =E2=80=94 but their =
attributes are still just their attributes, not attributes defined by in =
a PHP RFC.


> If the intention is to add some logic for attributes, then unlikely it =
is going to be accepted because attributes by definition are meta-data. =
It doesn't add any logic for program.

Actually, that is not true either.  PHP already has one attribute that =
PHP itself recognizes, named "Attribute."  As Larry Garfield wrote[1]:  =
"The #[Attribute] attribute tells PHP =E2=80=9Cyes, this class can be =
loaded as an attribute.=E2=80=9D =20

You can also see that future scope was explicitly envisioned[2] that =
some small number of attributes might potentially be more than just =
metadata.


-Mike
[1] https://platform.sh/blog/2020/php-8-0-feature-focus-attributes/ =
<https://platform.sh/blog/2020/php-8-0-feature-focus-attributes/>
[2] https://wiki.php.net/rfc/attributes_v2#future_scope =
<https://wiki.php.net/rfc/attributes_v2#future_scope>=20=

--Apple-Mail=_DC8F3ED2-39F5-4267-8C7A-E6CC4BC8C97A--