Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:110103
Date: Sat, 09 May 2020 20:44:03 +0200
To: "Bob Weinand" <bobwei9@hotmail.com>
Cc: "Internals" <internals@lists.php.net>
Message-ID: <171fac045cf.124a18ecf610619.2220256095841059487@manuelcanga.dev>
In-Reply-To: <VI1PR02MB4703DC86644E4D8952E5E334E2A20@VI1PR02MB4703.eurprd02.prod.outlook.com>
References: <171dee3baf3.bef5753260554.3262181919550021853@manuelcanga.dev> <VI1PR02MB4703DC86644E4D8952E5E334E2A20@VI1PR02MB4703.eurprd02.prod.outlook.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Importance: Medium
User-Agent: Zoho Mail
Subject: Re: [PHP-DEV] [RFC] Keep type of reference params
From: php@manuelcanga.dev (Manuel Canga)






 ---- En vie, 08 may 2020 23:40:22 +0200 Bob Weinand <bobwei9@hotmail.com> =
escribi=C3=B3 ----
 > > 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 pa=
rams" ) 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 $arra=
y is a string.=20
 > >=20
 > > However, I think it should be an error or exception when a string is a=
ssigned to $array var.=20
 > >=20
 > > In my opinion, $array var would have to keep its type when function en=
ds.
 > >=20
 > >=20
 > >=20
 > > What is your opinion ? Do you see it useful ?
 > >=20
 > > Thanks and I'm sorry for my English( I'm a Spanish ).
 > >=20
 > > Regards
 > >=20
 > > --
 > >=20
 > > Manuel Canga
 >=20
 > Hey Manuel,
 >=20
 > the primary issue (apart from the BC break) here is leaking the referenc=
e across the function boundary.
 >=20
 > function a(array &$a) {
 >     $GLOBALS["globalA"] =3D &$a;
 > }
 >=20
 > funcition b() {
 >     $GLOBALS["globalA"] =3D 10;
 > }
 >=20
 > $a =3D 1;
 > a($a);
 > b();
 > // $a is magically changed to 10
 >=20
 > Yes, you can here verify, that $a is an array at the function boundaries=
, but you cannot afterwards.
 >=20
 > If we had proper inout parameters (which do not leak a reference, but as=
sign the value of the variable (in callee scope) back to the passed variabl=
e from caller), then we could easily enforce it.
 >=20
 > But as it stands now, this is not an option. (Especially due to the fals=
e promise this seems to make.)
 >=20
 > Bob

Thanks Bob, that was a great example

I explain...I am very forgetful, so it's very easy  for me to write somethi=
nk like this:

```
function filter_something( &string $to_filter) {
  $to_filter  =3D strtolower($to_lower);
  $to_filter  =3D sanitize_this($to_filter);
  .....
}=20

filter_something($my_string);
```

Sometimes, one of these functions can returns a false|null|numeric|... inst=
ead of expected type.=20
Also, If any function in 'filter_somethid' would throw an exception, value =
in $to_filter would finish inconsistent

However, with:

```
function filter_something( inout string $to_filter) {
  $to_filter  =3D strtolower($to_lower);
  $to_filter  =3D sanitize_this($to_filter);
  .....
}=20

filter_something(inout $my_string);
```

PHP could in the end produce exception when any function in `filter_somethi=
ng` modify type of $my_string or =20
PHP could keep value of `$to_filter` as well if any function  throwed an ex=
ception.

Regards
--
Manuel Canga