Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:108231
To: Nikita Popov <nikita.ppv@gmail.com>
CC: Mark Randall <marandall@php.net>, "internals@lists.php.net"
	<internals@lists.php.net>
Thread-Topic: [PHP-DEV] Re: [RFC] "use global functions/consts" statement
Thread-Index:
 AQHVwNgGtev425C1jEaW2/fYCDQJoKfqjUCogAA6vwCAABZSCIAF7qHQgAA+1YCAABKbIoAI/p4AgAAhc/4=
Date: Fri, 24 Jan 2020 18:11:29 +0000
Message-ID:
 <DM5PR07MB30673A86761F13DB3CCCD8D9F90E0@DM5PR07MB3067.namprd07.prod.outlook.com>
References:
 <DM5PR07MB3067B7578E3F04C5A20A221AF9210@DM5PR07MB3067.namprd07.prod.outlook.com>
 <CY4PR07MB30633FFAAF6E9CC45B3ECB43F9340@CY4PR07MB3063.namprd07.prod.outlook.com>
 <CAF+90c9svofd0rJV9ooDUb+rgsgE4+cbaahUnLzYqrksY+d=WA@mail.gmail.com>
 <DM5PR07MB3067239059A936661E1100B5F9340@DM5PR07MB3067.namprd07.prod.outlook.com>
 <DM5PR07MB3067A4F37F4D8E6738582859F9300@DM5PR07MB3067.namprd07.prod.outlook.com>
 <b8e7ced4-9cec-4bde-a009-622c477a0e58@DB3EUR04FT030.eop-eur04.prod.protection.outlook.com>
 <DM5PR07MB306711A0E8F1CADB23C17CB5F9300@DM5PR07MB3067.namprd07.prod.outlook.com>,<CAF+90c-Rmmh3+iYqFmn1E+W5-S1qT4BpPhN9wgroz=9ZudeWpg@mail.gmail.com>
In-Reply-To:
 <CAF+90c-Rmmh3+iYqFmn1E+W5-S1qT4BpPhN9wgroz=9ZudeWpg@mail.gmail.com>
Accept-Language: en-CA, en-US
Content-Language: en-CA
x-ms-exchange-messagesentrepresentingtype: 1
x-tmn: [ZQdIxtmdk4YousmCLieCf1zT5tk8l+Vb]
x-ms-publictraffictype: Email
x-incomingheadercount: 46
x-eopattributedmessage: 0
x-ms-office365-filtering-correlation-id: c8fe9a4b-180a-4799-e5be-08d7a0f8d5f6
x-ms-traffictypediagnostic: DM6NAM10HT168:
x-microsoft-antispam: BCL:0;
x-microsoft-antispam-message-info:
 nETu6j3HeMW/GY8OdEqu/Uc3Imjj/HjbOp6co9gf8y4zQrHjGS8BI7d6afc8+/GFnL2ScMa06rgc/KM1Q1ploM4ltmi4aANeRiQid24bIVXYvDhfrPgaJEmLZMR+xP2VlIZUurydP5bQsL4kOZ59xiTlei72q9WNzVnAh8G3N0+QqyTtfMDiiQawtp2m7PwONjM+FuIPEnEGDxltXGI/fEwz06DaINt6+CPRe7Twsxw=
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-RMS-PersistedConsumerOrg: 00000000-0000-0000-0000-000000000000
X-MS-Exchange-CrossTenant-Network-Message-Id: c8fe9a4b-180a-4799-e5be-08d7a0f8d5f6
X-MS-Exchange-CrossTenant-rms-persistedconsumerorg: 00000000-0000-0000-0000-000000000000
X-MS-Exchange-CrossTenant-originalarrivaltime: 24 Jan 2020 18:11:29.7206
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Internet
X-MS-Exchange-CrossTenant-id: 84df9e7f-e9f6-40af-b435-aaaaaaaaaaaa
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM6NAM10HT168
Subject: Re: [PHP-DEV] Re: [RFC] "use global functions/consts" statement
From: tysonandre775@hotmail.com (tyson andre)

> I like the idea of using a meaningful value here, but think that this=0A=
> should be using a string, i.e. declare(function_and_const_lookup=3D'globa=
l')=0A=
> rather than declare(function_and_const_lookup=3Dglobal). =0A=
=0A=
Also, I could make that a voting option, and it could be changed before 8.0=
 alphas if new concerns come up after a vote.=0A=
=0A=
> This looks a bit peculiar... We could also say that declaring a symbol al=
so automatically brings it into scope.=0A=
> I'm not really sure which is better.=0A=
=0A=
Neither am I. There were edge cases with doing that that may cause more con=
fusion=0A=
=0A=
    namespace NS;=0A=
    function f() { /* calls g() - which g() */ }=0A=
    function g() { /* calls f() */ }=0A=
=0A=
    if (!function_exists('NS\h')) {=0A=
        function h() {}=0A=
    } else {=0A=
        // what will h() refer to=0A=
    } =0A=
    // what will h() refer to=0A=
=0A=
> As a general note on the overall proposal: I think this is a very pragmat=
ic proposal. It addresses a real issue, and it addresses it in the most str=
aightforward way. You can just drop the declare in your code, and it will k=
eep working exactly as it did=0A=
> before, just faster. Assuming you weren't doing anything stupid, of cours=
e.=0A=
>=0A=
> The big downside is that this makes class and function symbol resolution =
complete opposites. Classes only look in the current namespace. Functions o=
nly look in the global namespace. This reflects the way they are most commo=
nly used, but is not a great=0A=
 place to end up from a language design perspective.=0A=
>=0A=
There are two alternatives that keep class and function symbol resolution t=
he same: One is to always only look in the namespace. This has been suggest=
ed a few times in the past, including in this thread, and the RFC=0A=
https://wiki.php.net/rfc/fallback-to-root-scope-deprecation. I think this i=
s a fairly unpopular option, because using fully qualified names for functi=
ons tends to be quite ugly (more=0A=
 so than for classes, due to the positions they're used in), or requires a =
large number of imports, as even a small piece of code can easily use a doz=
en different string/array functions.=0A=
=0A=
Agreed=0A=
=0A=
> One option that I haven't seem much discussion on is the opposite: Always=
 only look in the global namespace. Any unimported unqualified usages will =
be treated as fully qualified names.=0A=
> This would match the proposed semantics for functions/consts and change=
=0A=
 the resolution rules for classes.=0A=
> I think this would have relatively little impact on how code is written, =
as classes already tend to make extensive use of cross-namespace references=
 and the import is usually IDE managed.=0A=
=0A=
Also, about the note on classes.=0A=
=0A=
It's fairly common for NS\SubNS\ClassName to mention other classes from NS\=
SubNS\OtherClassName right now,=0A=
(more commonly than use Exception, use Throwable, etc in some cases),=0A=
and changing that default would require changing a lot of third party code.=
=0A=
A separate option such as `declare(lookup_classes=3Dglobal)` would allow mi=
grating to that,=0A=
but would confuse developers switching between codebases using different se=
ttings of lookup_classes,=0A=
and introduce similar confusion about the rare case of multiple classes in =
one file.=