Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:44017
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 74.125.46.29 as permitted sender)
DomainKey-Status: bad
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        b=xu1zZ1n/gqMt0Ey0eUXHu2GkUCNYAd+sGgMWoTJ6dC4t/wW3nASZdi2xCDqexPwD8F
         eDYxvaadz70tLRa4h9jr/sZcBgGRr5xuy5CMkA6HJKRnYzePz19gLfCVumWcIQXlq10e
         m5NEjJan1b7Jmf8GMP5YMd+bR/o3qKIJTayPY=
MIME-Version: 1.0
In-Reply-To: <b9339a730905152303r3e1c6bddi582d8cac5dc05e2c@mail.gmail.com>
References: <b9339a730905152303r3e1c6bddi582d8cac5dc05e2c@mail.gmail.com>
Date: Sun, 17 May 2009 15:54:08 +0800
Message-ID: <b9339a730905170054o44148080t81d2faf6451c330f@mail.gmail.com>
To: php-dev <internals@lists.php.net>
Content-Type: multipart/alternative; boundary=001e680f1ac47e78f6046a16fb0e
Subject: Re: Suggestion on static field inheritance
From: diogin@gmail.com (Jingcheng Zhang)

--001e680f1ac47e78f6046a16fb0e
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Nobody is interested in this topic? Another confusing feature is "static
function", which is allowed in interface, but not allowed in abstract class,
any explanation or reason for this? Thanks :-)

2009/5/16 Jingcheng Zhang <diogin@gmail.com>

> Hello all,
>
> Maybe I have not found its detailed description on PHP's official manual,
> but PHP does allow static field inheritance. However there is a little
> difference between dynamic field inheritance and static field inheritance,
> as the following codes shows:
>
> <?php
> class static_a {
>     public static function change($name) {
>         self::$name = $name;
>     }
>     public static $name = 'a';
> }
> class static_c extends static_a {}
> class static_d extends static_a {}
>
> echo static_a::$name; // a
> static_c::change('c');
> echo static_a::$name; // c
> static_d::change('d');
> echo static_a::$name; // d
>
>
> class dynamic_a {
>     public function change($name) {
>         $this->name = $name;
>     }
>     public $name = 'a';
> }
> class dynamic_c extends dynamic_a {}
> class dynamic_d extends dynamic_a {}
>
> $a = new dynamic_a();
> $c = new dynamic_c();
> $d = new dynamic_d();
> echo $a->name; // a
> $c->change('c');
> echo $a->name; // a
> $d->change('d');
> echo $a->name; // a
> ?>
>
> The result of static inheritance test can be meaningful on some
> way(especially on class-based programming perspective), but when considering
> the static class as "object" in prototype-based programming(which I doubt is
> some people's favourite who comes from prototype-based OOP community), this
> result can be confusing. On JavaScript, this example can be:
>
> <script type="text/javascript">
> function extends(parent) {
>     var T = function () {};
>     T.prototype = parent;
>     return new T();
> }
> var static_a = {
>     name: 'a',
>     change: function (name) {
>         this.name = name;
>     }
> };
> var static_c = extends(static_a);
> var static_d = extends(static_a);
>
> alert(static_a.name); // a
> static_c.change('c');
> alert(static_a.name); // a
> static_d.change('d');
> alert(static_a.name); // a
> </script>
>
> This looks more meaningful. So my suggestion is, could PHP's static
> inheritance rule follow the dynamic inheritance rule ( $this on dynamic, or
> prototype delegation on JavaScript) ?
> Thanks :-)
>
> --
> Best regards,
> Jingcheng Zhang
> P.R.China
>



-- 
Best regards,
Jingcheng Zhang
Room 304, Dormitory 26 of Yuquan Campus, Zhejiang University
P.R.China

--001e680f1ac47e78f6046a16fb0e--