Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:94312
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain trowski.com designates 199.38.81.6 as permitted sender)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\))
In-Reply-To: <CAEKnhAEtLAhn6_6C2Pi22zjD4GGuTPToxQ34HM_H9TY3bC=EyA@mail.gmail.com>
Date: Tue, 28 Jun 2016 13:07:28 -0500
Cc: PHP internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
Message-ID: <8DBD4F28-5C5A-467A-92D9-6953D3FA1E4F@trowski.com>
References: <A447792E-9DA6-4067-9EA5-7B6910520A8F@trowski.com> <CAEKnhAEtLAhn6_6C2Pi22zjD4GGuTPToxQ34HM_H9TY3bC=EyA@mail.gmail.com>
To: Jakub Zelenka <bukka@php.net>
Subject: Re: [PHP-DEV] [RFC][Vote] Throw Error in Extensions
From: aaron@trowski.com (Aaron Piotrowski)

Hi Jakub,

> On Jun 28, 2016, at 12:28 PM, Jakub Zelenka <bukka@php.net> wrote:
>=20
> Hi,
>>=20
>>=20
> Just noticed the openssl case in X509_digest and it's obviously =
oversight
> by whoever added that bit because it should be warning as it's for all
> other similar fails. I'm going to change it to warning to make it
> consistent.

I noticed most others were warnings in openssl, but I did not want to =
make assumptions about what level an error should be. If an error was =
E_ERROR, I assumed there was a reason it was fatal. If you change it to =
a warning I'll be sure not to overwrite this when merging the patch.

> In general I agree with the idea but the patch should be a bit more
> sensible and considers consistency with other errors in the extension. =
It
> should be also reviewed by all active maintainers or regular =
contributors
> to the changed extensions before it gets merged so it might be a bit =
late
> for 7.1

I agree, either the maintainers or someone very familiar with each =
extension should examine the changes before merging. Note that this =
patch is only meant to allow catching and handling of otherwise fatal =
errors, not to modify overall error handling in each extension. I would =
rather individual extension maintainers make decisions on error levels.

Thanks!

Aaron Piotrowski