Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:96073
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 209.85.213.182 as permitted sender)
MIME-Version: 1.0
Sender: morrison.levi@gmail.com
In-Reply-To: <CAFMT4NoAEg+f_26d7Gt2RPTwUDRT65qRYMmhK0vBPiheFYBzGA@mail.gmail.com>
References: <CAOWwgpn1_ZHxQq9FpiwCYoqT3kVHH0k3LgO7E+eX5N2ybjDtrg@mail.gmail.com>
 <CAF+90c9XW7G-7CtKCWt+JOm5vE1XTnC92AkHvmBe9fohTOWeKA@mail.gmail.com>
 <CAOWwgpm6YL+kiwT=n8QdeKHVYg=KG7aCp=2Cv5qn0nRVNwR0bA@mail.gmail.com>
 <CAObuZduNr1oKRx0MKdGG=9XLADZF_tP4YORTk486kWa6CC_=Hg@mail.gmail.com>
 <CAOWwgpmjaYnsOcJ2_vACCpR95NNQk8C_PwyWtdwOA-s5FbfpSg@mail.gmail.com>
 <CAObuZdvjesE9=C8sFk_8wdbpUzu7g2CTUY6Js_mSitO6SbB2=Q@mail.gmail.com>
 <CAOWwgp=Cha7y0goSOkQPqCOEr1rwmG0+-6JGaiDZmqkhMfyrpQ@mail.gmail.com> <CAFMT4NoAEg+f_26d7Gt2RPTwUDRT65qRYMmhK0vBPiheFYBzGA@mail.gmail.com>
Date: Wed, 21 Sep 2016 12:17:40 -0600
Message-ID: <CAFMT4NqziQoD69sKxhmVKHW+TAUreYsndc6S4jrMtbf670ixRg@mail.gmail.com>
To: Nicolas Grekas <nicolas.grekas@gmail.com>
Cc: Ryan Pallas <derokorian@gmail.com>, internals <internals@lists.php.net>
Content-Type: text/plain; charset=UTF-8
Subject: Re: [PHP-DEV] Fix ReflectionType::__toString() BC break
From: levim@php.net (Levi Morrison)

On Wed, Sep 21, 2016 at 12:07 PM, Levi Morrison <levim@php.net> wrote:
> On Wed, Sep 21, 2016 at 11:13 AM, Nicolas Grekas
> <nicolas.grekas@gmail.com> wrote:
>>> To handle this in code written around current __toString seems pretty
>> simple
>>
>> Yes it is, but that's not what we're talking about:
>> BC is about having perfectly fine code working in e.g. 7.0 be still working
>> fine on 7.1 *without any change*.
>>
>> Right now, we have red test suites on php7.1rc2.
>> This is the symptom of a BC break, by definition.
>> And the issue is not the existing code we have, but the new one that is
>> changing the behavior of the engine.
>
> This was understood when the decision was made. You seem to not be
> understanding the bigger issue and instead focusing on the BC break
> for a *single minor release, and a dot zero at that*. If we keep the
> BC compat this method is redundant and useless forever. If we fix it
> we break your code for *one single minor release, and a dot zero at
> that*. Which is the bigger disruption?
>
> This is why the decision was made. It is better to have the useful
> functionality from here on out than to preserve BC with a single minor
> release, and a dot-zero at that.

I should also add that the dot-zero is the same version it was
released. Having backwards compatibility breaks in the version after
its release in some cases is not that uncommon (though unfortunate).

Essentially I understand the frustration and empathize with having
your code broken. However, I still think it is better for the overall
language health to prepend the question mark, which is why it is there
in the RC releases despite the previous discussion thread.