Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:106569
Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (1.0)
In-Reply-To: <3db68c5e-54d1-812f-bbf5-1b089cba1bff@gmail.com>
Date: Wed, 14 Aug 2019 00:49:41 +0700
Cc: internals@lists.php.net
Content-Transfer-Encoding: quoted-printable
Message-ID: <722174DE-F28F-41A3-BFF2-23E3CA8B44AA@koalephant.com>
References: <CAF+90c_Ctjja=8fR-uBWL1wDxZge771yMujMhaR_R2seKXciYw@mail.gmail.com> <CAOWwgpmsHMjnHuH4tfv_5-0f+pGKv2jJmA5v21FbkdpSHTbTJQ@mail.gmail.com> <25d774e8-183b-d39c-4ac4-81c1b7770229@gmail.com> <5d5298a5.1c69fb81.b4ed1.2d97SMTPIN_ADDED_MISSING@mx.google.com> <3db68c5e-54d1-812f-bbf5-1b089cba1bff@gmail.com>
To: Rowan Collins <rowan.collins@gmail.com>
Subject: Re: [PHP-DEV] [RFC] Namespace-scoped declares, again
From: php-lists@koalephant.com (Stephen Reay)



> On 14 Aug 2019, at 00:05, Rowan Collins <rowan.collins@gmail.com> wrote:
>=20
>> On 13/08/2019 12:01, Mark Randall wrote:
>>> On 13/08/2019 10:02, Rowan Collins wrote:
>>> I really like this approach. It allows a package definition file to exis=
t, without either the language or the header of each file having to define i=
ts location.
>>=20
>> #
>> # File: /lib/company/project1/a/b/MyClass.php
>> #
>>=20
>> <?php
>>=20
>>   declare_import(company[:label]);
>>=20
>>   namespace company/project1/a/b;
>>=20
>>   ...
>>=20
>>=20
>> #
>> # File: /lib/company/__nsmeta.php
>> #
>>=20
>> <?php
>>=20
>>   namespace company;
>>=20
>>   namespace_declare [:label] {
>>     strict_types=3D1;
>>     strict_operators=3D1;
>>     upgrade_errors_to_exceptions=3DE_ALL;
>>   }
>>=20
>>   final class __nsmeta {
>>     /* does nothing except be findable by the autoloader */
>>     /* and defaults to private constructor so it can't be used */
>>   }
>=20
>=20
> This seems to be more complicated than Nicolas's version, and involve much=
 more special magic, like the name __nsmeta, and the class that does nothing=
. I'm also not clear on how you're picturing the relationship between namesp=
aces and packages.
>=20
> The way I understood the suggestion, you'd end up with something more like=
 this, which feels much simpler and cleaner:
>=20
> #
> # File: /lib/company/project1/a/b/MyClass.php
> #
>=20
> <?php
>=20
> declare(package=3Dcompany/project1);
> // or with a new keyword
> package company/project1;
>=20
> namespace company/project1/a/b;
>=20
> ...
>=20
>=20
> #
> # File: /lib/company/project1.php
> #
>=20
> <?php
>=20
> namespace company;
>=20
> class project1 extends \PHP\Package {
>     public function getParserDirectives() {
>         return [
>             'strict_types' =3D> 1,
>             'strict_operators' =3D> 1,
>             'upgrade_errors_to_exceptions' =3D> E_ALL
>         ];
>      }
> }
>=20
> Regards,
>=20
> --=20
> Rowan Collins
> [IMSoP]
>=20
>=20
> --=20
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: http://www.php.net/unsub.php
>=20

I think this approach is easier to follow (less magic) but has a (small) pot=
ential for conflict with existing code that may then require renaming or mor=
e (eg if a class existed with that proposed name, and is either a singleton a=
nd/or already extends another class).

Is it feasible to instead require the class implements an interface that def=
ines a static method?=20


Building on that, could the concept of =E2=80=9Cdo something to initialise t=
his package=E2=80=9D be made more generic: when a file declares it=E2=80=99s=
 a member of a given package, the static method on the class that matches th=
e package name is called, and it=E2=80=99s up to the developer what it does i=
n that method. That would of course also then require a new function to set t=
he declares with package scope to be useful for this particular problem,  bu=
t it also allows more control and power for the concept of packages in gener=
al.

Cheers

Stephen=20=