Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:72596
Return-Path: <ingwie2000@googlemail.com>
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Delivered-To: mailing list internals@lists.php.net
Received: (qmail 17748 invoked from network); 14 Feb 2014 13:09:34 -0000
Received: from unknown (HELO lists.php.net) (127.0.0.1)
  by localhost with SMTP; 14 Feb 2014 13:09:34 -0000
Authentication-Results: pb1.pair.com smtp.mail=ingwie2000@googlemail.com; spf=pass; sender-id=pass
Authentication-Results: pb1.pair.com header.from=ingwie2000@googlemail.com; sender-id=pass
Received-SPF: pass (pb1.pair.com: domain googlemail.com designates 74.125.83.45 as permitted sender)
X-PHP-List-Original-Sender: ingwie2000@googlemail.com
X-Host-Fingerprint: 74.125.83.45 mail-ee0-f45.google.com  
Received: from [74.125.83.45] ([74.125.83.45:52427] helo=mail-ee0-f45.google.com)
	by pb1.pair.com (ecelerity 2.1.1.9-wez r(12769M)) with ESMTP
	id A1/91-09381-C851EF25 for <internals@lists.php.net>; Fri, 14 Feb 2014 08:09:33 -0500
Received: by mail-ee0-f45.google.com with SMTP id b15so5664094eek.32
        for <internals@lists.php.net>; Fri, 14 Feb 2014 05:09:29 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=googlemail.com; s=20120113;
        h=content-type:mime-version:subject:from:in-reply-to:date:cc
         :content-transfer-encoding:message-id:references:to;
        bh=sqO0L9fqYpfVLXyBESMaJrv8qAZMVrATEHBZKg3kq10=;
        b=PZdd4OSeUP4atNueGDOUGc6eYHCuOtpaZ/vB7IJzxBk9xYj439hM2oCVE6On+oFEJg
         4BjQbEQMCqTM6RCxcoOFVyRCwOvPL50TDJng4WmsmgXcgRML2bp6QtqW4tTjqy9ETuHw
         ClCKThqzZ+734rmkBKy1h1IpuidEP6xbmlsCAnymY3LFP9+YBSZSR1ZjN72pwSrVKlBo
         7ukqIL76RPJsQTY2jjnrgq1pxBgOQmpKc9qTw6GLqXBVwvdBwbneyUfvUKoS01jwxIZD
         4IRftUedSf+PqE89NYJDffNePXP5sdXEBae+PdqmEEhxI+2fxnBrewnrfqTLe5AZUe1j
         VjHQ==
X-Received: by 10.14.203.197 with SMTP id f45mr334721eeo.90.1392383369465;
        Fri, 14 Feb 2014 05:09:29 -0800 (PST)
Received: from [192.168.200.19] (dslb-088-068-169-060.pools.arcor-ip.net. [88.68.169.60])
        by mx.google.com with ESMTPSA id j42sm19686198eep.21.2014.02.14.05.09.28
        for <multiple recipients>
        (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128);
        Fri, 14 Feb 2014 05:09:28 -0800 (PST)
Content-Type: text/plain; charset=windows-1252
Mime-Version: 1.0 (Mac OS X Mail 7.0 \(1822\))
In-Reply-To: <CA+wXisV6++6J7u-epv=FTOAggbECas3E+3-SyKR3NAtV35zRNA@mail.gmail.com>
Date: Fri, 14 Feb 2014 14:09:26 +0100
Cc: internals <internals@lists.php.net>,
 Sebastian Krebs <krebs.seb@gmail.com>
Content-Transfer-Encoding: quoted-printable
Message-ID: <3B2D0F80-FACF-4D6E-A11F-28009F5ED0FE@googlemail.com>
References: <CA+wXisVVDaQYL+drdonNhxD=+=CWpES5K9UPsx40EvrXAAG3Yw@mail.gmail.com> <CALtuQzBh-9gWv843tRNie6=C_w4eFcxyTB9692XFz+XrVT352g@mail.gmail.com> <1373879683.17031.4.camel@guybrush> <CA+wXisVXVxquXm+MZrOwoC_+PqWptdzhxv-VtRQoyn8Hwuxsug@mail.gmail.com> <CA+wXisUWGf3jA8ZWWkSHYdjUq46Ez1O7-Fk9VXG5Nj-mw-2euQ@mail.gmail.com> <CADyq6sJT5bjW0PbETLCaiLdqiX+jzSRAx4L3_ZoeT2FPOVVXoQ@mail.gmail.com> <CA+wXisWnBwcHxSHha3Vk9KcPXbzL8BCWhDEpHrr+EMdLPF2e4Q@mail.gmail.com> <CA+wXisXDgsHYTahyBsqYtA5kdo9JMGi5jJK92W1-=RFBkSte_A@mail.gmail.com> <CA+wXisWy8_Bo=ZEDxXaVa3vv5oW097qWqzDUO43H2JG9Z-5XFg@mail.gmail.com> <CA+wXisV6++6J7u-epv=FTOAggbECas3E+3-SyKR3NAtV35zRNA@mail.gmail.com>
To: Giuseppe Ronca <giuseppe.ronca2@gmail.com>
X-Mailer: Apple Mail (2.1822)
Subject: Re: [PHP-DEV] PHP proposal on modular systems
From: ingwie2000@googlemail.com (Kevin Ingwersen)

If we could just override the require function, so we would be able to =
mimic the behavior of pytons import, and nodejs=92 require. Porting the =
architecture is rather easy, but without being able to override =
functions such as require, this is going to be harder.
Actually, this is an idea for an RFC i was thinking about=85 o.o; Thing =
is just, I dont know how to mod the core to allow overriding of the =
require function.

Kind regards.
Am Fr. Feb. 14 2014 13:53:37 schrieb Giuseppe Ronca:

> Sorry for "necroposting" but i would bumping this because since =
package
> managers such as composer/bower etc. are growing , the hell of
> dependencies conflicts is increasing proportionally.
>=20
> here you are other examples:
>=20
> =
http://stackoverflow.com/questions/12656982/resolving-dependency-hell-with=
-composer
>=20
> =
http://stackoverflow.com/questions/15256416/symfony2-how-solve-two-bundles=
-requirements-conflicts-when-using-composer
>=20
> it's a real problem and not always caused by "bad programming".
>=20
> i think that the language itself must offer a solution in this case ( =
see
> javascript and no-conflicts techniques )
>=20
> so , as explained in my first post, namespaces could be a nice =
solution if
> they can be nested or dynamically added in some way.
>=20
>=20
> 2013-07-17 0:29 GMT+02:00 Giuseppe Ronca <giuseppe.ronca2@gmail.com>:
>=20
>> Sorry , i think you're misunderstanding the main concept of this =
proposal
>> ..
>>=20
>> take a look at this post: http://news.php.net/php.internals/68118 it
>> explains how you could solve this problem using namespace.
>> You can actually have 2 version of the same library loaded inside =
same
>> process . But namespacing libraries ,using version number encoded in
>> namespace, is not a common technique...and manually namespacing =
libraries
>> is often a long job.
>>=20
>> It could be solved by precedent language features IMHO i've exposed. =
And
>> many people can gain on this.. Composer too.
>>=20
>>=20
>> 2013/7/16 Giuseppe Ronca <giuseppe.ronca2@gmail.com>
>>=20
>>> You can have multiple objects from the same type in different
>>>> implemenations in the same process?
>>>>=20
>>>=20
>>>=20
>>> No i'm not saying it.
>>>=20
>>>=20
>>>=20
>>>> What should happen, when I instanciate an object of "test" in =
version X
>>>> and
>>>> pass it to a function, that expect it as version Y? To be =
consistent PHP
>>>> _must_ trigger an error too, if something from the locally included =
file
>>>> leaves its scope
>>>>=20
>>>> class foo {
>>>>     public static function bar () {
>>>>          local_include "test.php" / / file which includes "test" =
class;
>>>>          return new test (); // <-- error here
>>>>     }
>>>> }
>>>> Because else it is not a local scope anymore.
>>>>=20
>>>=20
>>>=20
>>> the concept as i said is the same of "nesting" a class inside =
another (
>>> or dynamically namespacing a file )
>>>=20
>>>=20
>>> if two different components need "Library v1.0" and "Library v2.0" =
..
>>> they will probably have same classes and namespaces ( but the code =
is not
>>> the same ). So how could we use both lib at the same time?   =
Actually php
>>> can't load classes in a scope way so you can't do it in any way on =
global
>>> scope.
>>>=20
>>> nesting dynamically libraries inside a namespace or a class , we can
>>> create objects that can be accessed only throught their root
>>> namespace/class...something like:
>>>=20
>>> new Version1\ExternalLibrary\FooClass();
>>>=20
>>> new Version2\ExternalLibrary\FooClass();
>>>=20
>>> and also Composer will thankyou about it since they can solve
>>> automatically problems .
>>>=20
>>>=20
>>>=20
>>>> Backward compatibility (or compatibility in general) between =
component is
>>>> something the plugin maintainer should take care of, not the =
language.
>>>> Regards,
>>>> Sebastian
>>>=20
>>>=20
>>>=20
>>> i'm talking about backward compatibility of php version...not =
external
>>> plugins etc...i think that my idea doesn't require any change to php =
apps
>>> to adapt their code.
>>>=20
>>>=20
>>=20