Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:38301
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 209.85.200.172 as permitted sender)
DomainKey-Status: bad
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=message-id:date:from:to:subject:cc:in-reply-to:mime-version
         :content-type:references;
        b=tDNro7qKH2J7qUjN26Ac6r0ZBLsZ3jkKMDu3xzg5KF2sAjh125rLxTOR1OBuD4qnUs
         nWUJzRtjmopAeMLDPr8jvDKPEUuFvSNWGNmTomY7XKMYfrpqp4CpTR1IvpmQpNfu7HBT
         xsE7DieU7dVSxjHRqteBptCzqMLsHZrp1M2oI=
Message-ID: <e2e02fe0806161504s2838c79dhe85f7cbd541012e2@mail.gmail.com>
Date: Mon, 16 Jun 2008 15:04:54 -0700
To: "Stanislav Malyshev" <stas@zend.com>
Cc: internals@lists.php.net
In-Reply-To: <4856D4EF.4000002@zend.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; 
	boundary="----=_Part_5306_28163889.1213653894254"
References: <46ccd1ab0804170742g142ed562t4f03339da549a820@mail.gmail.com>
	 <48077BD2.2000107@zend.com>
	 <e2e02fe0804170950udd38ab3m21e6d658254f042a@mail.gmail.com>
	 <48078168.1070801@zend.com>
	 <496536410.20080525103752@marcus-boerger.de>
	 <483929DC.2060606@zend.com>
	 <e2e02fe0806151212o3b22ada0wd885cf635961a1e0@mail.gmail.com>
	 <4856D4EF.4000002@zend.com>
Subject: Re: [PHP-DEV] [RFC] Strict type hints (parameter and return value)
From: chrisstocktonaz@gmail.com ("Chris Stockton")

------=_Part_5306_28163889.1213653894254
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

On Mon, Jun 16, 2008 at 2:02 PM, Stanislav Malyshev <stas@zend.com> wrote:

> Hi!
> So you just ignored the "special cases" part. Yes, if you have special
> case where you interface with very brain-dead strictly typed system that
> absolutely can't understand that '1' and 1 is the same - then you need to
> _convert_. So how failing when you get '1' instead of 1 helps you? You'd
> need _conversion_, not _failure_ - and if you write strictly-typed API,
> you'd move the conversion responsibility to the user, instead of having it
> where it belongs - in the API. That's *exactly* why I see strict typing in
> PHP so dangerous - it promotes lazyness and sloppiness in API writing, and
> those APIs will be a nightmare to use, since they would bomb out on
> slightest disagreement about internal engine types, which the API user
> shouldn't care about at all. What happened with "be liberal at what you
> accept"?
> In strict compiled languages, the compiler and IDE will guide you
> through this, in PHP you'd just have it explode in your face in production.
> How this is good for anybody?


I get this, really, I do, again. Like the previous post I (likely we)
understand.

Did you not get or read my post? You say again and again 1 == '1', but you
are forgetting those OTHER operators, === and !==. 1 !== '1' does it?
====>Add numeric hinting like I said 50 posts ago and your 1 == '1' problem
is solved<==== People who want to make more simple API's will make simple
API's offsetting the complexity of your precious 1 '1' to the API level.
People who want to make it strictly integer, will continue to do so as they
do now. You can not, and should not want to, control the way people program.

But I know this will never happen because no one cares what users want...
your logic of

"We should allow user to check for is_int and is_numeric, resource, object,
array, string, etc, within any conditional statement under a function or
method. Allow checking for object or array in functions or methods
paramters. Disallow anything else from being checked in functions or methods
paramters. Follow a philosophy that a user should not be concerned with
internal data types. Make sure the user has access to internal data types.
User should not have to talk to strict typed systems. But if they do have to
make it more difficult then it needs to be because it makes more sense, plus
those systems are brain dead."

How in the world, does this remotely, possibly, even slightly make sense?

-Chris

------=_Part_5306_28163889.1213653894254--