Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:110095
Content-Type: text/plain;
	charset=us-ascii
In-Reply-To: <171dee3baf3.bef5753260554.3262181919550021853@manuelcanga.dev>
Date: Fri, 8 May 2020 23:40:22 +0200
Cc: Internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
Message-ID:
 <VI1PR02MB4703DC86644E4D8952E5E334E2A20@VI1PR02MB4703.eurprd02.prod.outlook.com>
References: <171dee3baf3.bef5753260554.3262181919550021853@manuelcanga.dev>
To: Manuel Canga <php@manuelcanga.dev>
MIME-Version: 1.0
X-MS-Exchange-MessageSentRepresentingType: 1
Received: from [IPv6:2001:7e8:cb47:6400:7457:1108:e8c:f12b] (2001:7e8:cb47:6400:7457:1108:e8c:f12b) by PR3P191CA0040.EURP191.PROD.OUTLOOK.COM (2603:10a6:102:55::15) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.2979.27 via Frontend Transport; Fri, 8 May 2020 21:40:24 +0000
X-Mailer: Apple Mail (2.3445.104.11)
X-Microsoft-Original-Message-ID:
 <36A0BD75-9A0F-4C89-8172-84943746F4FB@hotmail.com>
X-TMN: [xknXYw/CIxRQENRbxjQSM5OM4ADKvj8P0yyTY7ZrpqvAHeRVO3drJxGxSnlSZVA7]
X-MS-PublicTrafficType: Email
X-IncomingHeaderCount: 50
X-EOPAttributedMessage: 0
X-MS-Office365-Filtering-Correlation-Id: 3fd38c87-c51e-46ae-9e7b-08d7f3986aec
X-MS-TrafficTypeDiagnostic: HE1EUR04HT175:
X-Microsoft-Antispam: BCL:0;
X-Microsoft-Antispam-Message-Info:
	Utu86hbgN44OSE8b/drFiQQS4bFl7H4bqME9r5dLO0nE0TknqM7L7pVV/BOm0IARTQn+bekZZIK4jWse3VpXyWdgsGW1IGJ7VCMAVDemrPRrI+9oSJE4tRxU82U1hueiwrXQnewTCQTwH71YTd+1Fpc4qzsYx3NHRsUtPEs39qqF0wbJNCf65Eam7ASiEHN3LX9/IUiolFPTWO/4Fe4vhQ==
X-Forefront-Antispam-Report:
	CIP:255.255.255.255;CTRY:;LANG:en;SCL:0;SRV:;IPV:NLI;SFV:NSPM;H:VI1PR02MB4703.eurprd02.prod.outlook.com;PTR:;CAT:NONE;SFTY:;SFS:;DIR:OUT;SFP:1901;
X-MS-Exchange-AntiSpam-MessageData:
	mtAbIyrHxC7lrM6LyV/kuH3ItY291m5Vl94+LH/bAjMD5tCk0TTgIN0oq8TGJjV5XuP09tLwX6OGDiF9tnpMW4No3xv27TIX1ouRfrLblEHFaWgABTo9Gpf/hI5qxJPd425RWMcWF1qpbsbJgglx5LBah6BDBIuI0VHj/VfI/EV4IKF4PrdY/gaPK7/qBCTKoTUhSfwqeU05uGpyIJjU0Q==
X-OriginatorOrg: hotmail.com
X-MS-Exchange-CrossTenant-Network-Message-Id: 3fd38c87-c51e-46ae-9e7b-08d7f3986aec
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 08 May 2020 21:40:26.8172
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 84df9e7f-e9f6-40af-b435-aaaaaaaaaaaa
X-MS-Exchange-CrossTenant-FromEntityHeader: Internet
X-MS-Exchange-CrossTenant-RMS-PersistedConsumerOrg:
	00000000-0000-0000-0000-000000000000
X-MS-Exchange-Transport-CrossTenantHeadersStamped: HE1EUR04HT175
Subject: Re: [PHP-DEV] [RFC] Keep type of reference params
From: bobwei9@hotmail.com (Bob Weinand)

> Am 04.05.2020 um 10:53 schrieb Manuel Canga <php@manuelcanga.dev>:
>=20
> Hi internals,
>=20
>=20
>=20
> I would like to present a possible new RFC( "keep type of reference param=
s" ) for your
>=20
> consideration.
>=20
>=20
>=20
> Firstly, an example:
>=20
>=20
>=20
> ```
>=20
> <?php
>=20
>=20
>=20
> function my_array_shift( array & $array ) {
>=20
>     $array =3D "string";
>=20
> }
>=20
>=20
>=20
> $array =3D [ 0, 1, 2, 3, 4 ];
>=20
>=20
>=20
> my_array_shift($array);
>=20
>=20
>=20
> count( $array );
>=20
> ```
>=20
>=20
>=20
> The result of this code is a warning( in count line ) because of $array i=
s a string.=20
>=20
> However, I think it should be an error or exception when a string is assi=
gned to $array var.=20
>=20
> In my opinion, $array var would have to keep its type when function ends.
>=20
>=20
>=20
> What is your opinion ? Do you see it useful ?
>=20
> Thanks and I'm sorry for my English( I'm a Spanish ).
>=20
> Regards
>=20
> --
>=20
> Manuel Canga

Hey Manuel,

the primary issue (apart from the BC break) here is leaking the reference a=
cross the function boundary.

function a(array &$a) {
    $GLOBALS["globalA"] =3D &$a;
}

funcition b() {
    $GLOBALS["globalA"] =3D 10;
}

$a =3D 1;
a($a);
b();
// $a is magically changed to 10

Yes, you can here verify, that $a is an array at the function boundaries, b=
ut you cannot afterwards.

If we had proper inout parameters (which do not leak a reference, but assig=
n the value of the variable (in callee scope) back to the passed variable f=
rom caller), then we could easily enforce it.

But as it stands now, this is not an option. (Especially due to the false p=
romise this seems to make.)

Bob=