Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:92658
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: error (pb1.pair.com: domain fleshgrinder.com from 212.232.25.162 cause and error)
Reply-To: internals@lists.php.net
References: <BY2PR0201MB178479B3AF8BD49583233034BF6E0@BY2PR0201MB1784.namprd02.prod.outlook.com>
 <CAESVnVrNqpqsTUCdHiQi8j3d4KVz8ZqRh6pEVJegR-EsmFN3_g@mail.gmail.com>
To: Sara Golemon <pollita@php.net>, internals <internals@lists.php.net>,
 "dmitry@zend.com >> Dmitry Stogov" <dmitry@zend.com>
Message-ID: <571BA0F0.2030400@fleshgrinder.com>
Date: Sat, 23 Apr 2016 18:21:04 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101
 Thunderbird/38.7.2
MIME-Version: 1.0
In-Reply-To: <CAESVnVrNqpqsTUCdHiQi8j3d4KVz8ZqRh6pEVJegR-EsmFN3_g@mail.gmail.com>
Content-Type: multipart/signed; micalg=pgp-sha256;
 protocol="application/pgp-signature";
 boundary="W60ChiNPxcP5FCSP9FcwKNHov8nB1gX9K"
Subject: Re: [PHP-DEV] [RFC] PHP Attributes
From: php@fleshgrinder.com (Fleshgrinder)

--W60ChiNPxcP5FCSP9FcwKNHov8nB1gX9K
Content-Type: multipart/mixed; boundary="rSLeut9M0lVNQHkbmhavEHMSM7UeVfbIa"
From: Fleshgrinder <php@fleshgrinder.com>
Reply-To: internals@lists.php.net
To: Sara Golemon <pollita@php.net>, internals <internals@lists.php.net>,
 "dmitry@zend.com >> Dmitry Stogov" <dmitry@zend.com>
Message-ID: <571BA0F0.2030400@fleshgrinder.com>
Subject: Re: [PHP-DEV] [RFC] PHP Attributes
References: <BY2PR0201MB178479B3AF8BD49583233034BF6E0@BY2PR0201MB1784.namprd02.prod.outlook.com>
 <CAESVnVrNqpqsTUCdHiQi8j3d4KVz8ZqRh6pEVJegR-EsmFN3_g@mail.gmail.com>
In-Reply-To: <CAESVnVrNqpqsTUCdHiQi8j3d4KVz8ZqRh6pEVJegR-EsmFN3_g@mail.gmail.com>

--rSLeut9M0lVNQHkbmhavEHMSM7UeVfbIa
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

On 4/22/2016 4:15 AM, Sara Golemon wrote:
> All that said, I love the proposal overall, and I can't wait to
> propose builtin annotations like <<__Memoize>>, <<__Mock>>, and
> similar.
>=20

I'd rather see these two functionalities added as modifiers at the
language level instead since they change the code behavior rather
dramatically. Not sure if knowledge of them is of interest in stages
where annotations might not be readily available; this was a driving
factor in Kotlin's decision what becomes a modifier or annotation.[1]

<?php

final class A {

  private memoized function f() {}

}

test class B extends A { // OK, despite final.

  public function f() {} // Has access to private methods.

}

test class C extends B {} // OK

class D extends B {} // Fatal Error!

?>

I know that this is true for `@invariant`, `@require`, and `@ensure` as
well, they are fully-fledged assertions after all. Especially the
`@invariant` could be easily added as a magic method.

[assert]
assert.invariants =3D 1

<?php

final class Object {

  public function __invariant() {}

}

?>

Both `@require` and `@ensure` are a bit harder to implement at language
level. The following is the best I can come up with and is inspired by
the Eiffel syntax.[2]

[assert]
assert.preconditions =3D 1
assert.postconditions =3D 1

<?php

class PreconditionError extends AssertionError {}
class PostconditionError extends AssertionError {}

final class Account {

  public function withdraw(int $sum): void {
    $this->balance -=3D $this->balance - $sum;
  }
  require {
    $sum >=3D 0;
    $sum <=3D ($this->balance - self::MIN_BALANCE);
  }
  ensure {
    $this->balance =3D=3D=3D (old::$balance - $sum);
  }

}

$account =3D new Account;

$account->withdraw(-10);
/*
Uncaught PreconditionError: $sum >=3D 0
*/

$account->withdraw(10);
/*
Uncaught PreconditionError: $sum <=3D ($this->balance - self::MIN_BALANCE=
)
*/

$account->deposit(10);
$account->withdraw(5);
/*
Uncaught PostconditionError: $this->balance =3D=3D=3D (old::$balance - $s=
um)
*/

?>

I actually think now that it would be better to go for this than for
annotations.

[1] http://blog.jetbrains.com/kotlin/2015/08/modifiers-vs-annotations/
[2]
https://archive.eiffel.com/doc/online/eiffel50/intro/language/invitation-=
07.html

--=20
Richard "Fleshgrinder" Fussenegger


--rSLeut9M0lVNQHkbmhavEHMSM7UeVfbIa--

--W60ChiNPxcP5FCSP9FcwKNHov8nB1gX9K
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
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=JkwZ
-----END PGP SIGNATURE-----

--W60ChiNPxcP5FCSP9FcwKNHov8nB1gX9K--