Re: Octave-maintainers Digest, Vol 58, Issue 38

[Pascal Dupuis]

Le 18/01/11 11:25, address@hidden a écrit :
> Message: 8
> Date: Tue, 18 Jan 2011 05:25:23 -0500
> From: "John W. Eaton" <address@hidden>
> To: octave maintainers mailing list <address@hidden>
> Subject: branching for release?
> Message-ID: <address@hidden>
> Content-Type: text/plain; charset=us-ascii
> 
> Are there any bug reports that should be handled before I create a
> branch for the release?
> 
> I know that there has been a small explosion in bug reports since I
> first mentioned that we should release 3.4 soon.  It would be great to
> fix more of these problems, but I think we are a somewhat understaffed
> to win this whac-a-mole competition.
> 
> Are there any that are serious regressions from previous versions of
> Octave that absolutely must be fixed before the release?
> 
> If we were to release today with the current sources, are there any
> mind blowingly obvious problems that will affect nearly everyone who
> runs Octave?
> 

Sorry for the whac-a-mole. Feeling the closeness of the release urged
peoples to fill bug reports.

Serious regressions: none so far. But lack of progressions: maybe we
could make more functions class aware ? I just reported a bug against
skewness.m. Here and there, there is an implicit assumption that the
output is just a matrix of doubles. Is it important to advertise in the
release notes that most basic functions are now more compatibles with
user-defined classes ?

Regards

Pascal