Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:1490
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Date: Tue, 13 May 2003 22:34:01 +0200
To: internals@lists.php.net
Message-ID: <20030513223357.6A42.MAXIM@php.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable
Subject: Re: [PHP-DEV] exceptions instead of errors
From: maxim@php.net (Maxim Maletsky)


Just to throw some fuel into the fire: I still fancy the idea discussed mon=
ths
ago about a new error reporting logic that would have some sort of
unique codes for errors supported by strings. Something comparable to
the way Oracle, C# and 'others' handles them.

here is some of that:
http://groups.google.com/groups?q=3Dmaxim+%2Berror+group:php.dev&hl=3Den&lr=
=3D&ie=3DUTF-8&oe=3DUTF-8&group=3Dphp.dev&selm=3D20021121143311.63B5.MAXIM%=
40php.net&rnum=3D3
http://groups.google.com/groups?q=3Dmaxim+%2Berror+group:php.dev&hl=3Den&lr=
=3D&ie=3DUTF-8&oe=3DUTF-8&group=3Dphp.dev&selm=3D000401c29448%2473f3f840%24=
9d10fea9%40coogle&rnum=3D1


Maxim Maletsky
maxim@php.net




On Tue, 13 May 2003 21:28:26 +0200
marcus.boerger@t-online.de (Marcus B=F6rger) wrote:

> At 18:45 13.05.2003, Zeev Suraski wrote:
> >Unless I'm missing something Marcus' patch only deals with non fatal=20
> >errors.  But if I am missing something and it somehow tries to be smart=
=20
> >about E_ERRORs - it shouldn't, whether or not an error is fatal should b=
e=20
> >left entirely to the author of the error call...
>=20
> Atm the following error cannot be turned into an exception, all others ca=
n be.
> E_ERROR
> E_CORE_ERROR
> E_COMPILE_ERROR
> E_USER_ERROR
> E_PARSE
>=20
> The reason i did not allow E_ERROR is the fact that i knew some few error=
s
> which leave an unstable engine state.
>=20
> So perhaps a new error code isn't such a bad idea as i first thought. But=
=20
> it makes
> only sense when we change some codes which are aren't really E_ERROR and
> that would be another BC issuue since it would also affect non exception =
code.
> OR we convert the new error code into its original E_ERROR code before sp=
itting
> it out as a normal error.
>=20
> marcus
>=20
>=20
> --=20
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: http://www.php.net/unsub.php
>=20


--------------------- Original Message Ends --------------------