Re: [Pkg-octave-devel] Bug#567975: FP error when filtering an empty vector

[Thomas Weber]

On Tue, Feb 02, 2010 at 01:26:41AM -0500, John W. Eaton wrote:
> On  1-Feb-2010, Bas Zoetekouw wrote:
> 
> | Octave crashes when trying to filter() an empty vector:
> | 
> | | octave:1> a=1; b=ones(10,1)/10; foo=filter(b,a,[]);
> | | panic: Floating point exception -- stopping myself...
> | | attempting to save variables to `octave-core'...
> | | save to `octave-core' complete
> | | zsh: floating point exception  octave
> | 
> | and I'm dropped back into the shell...
> | 
> | Of course, filtering an empty vector doesnt make much sense, but it 
> certainly
> | shouldn't cause a crash.
> | FWIT, Matlab simple returns an empty vector.
> | 
> | -- System Information:
> | Debian Release: 5.0.4
> |   APT prefers stable
> |   APT policy: (500, 'stable')
> | Architecture: amd64 (x86_64)
> | 
> | Kernel: Linux 2.6.26-2-amd64 (SMP w/1 CPU core)
> | Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
> | Shell: /bin/sh linked to /bin/bash
> 
> I checked in the following change:
> 
>   http://hg.savannah.gnu.org/hgweb/octave/rev/a277ba5da4dc
> 
> I'm not sure this fix places the return in exactly the right
> location.  Should it be later in the function, after other dimension
> checks?  It would be nice if someone who is more familiar with how
> filter is expected to work would comment.
> 
> Also, in the future, please report bugs in Octave itself directly to
> the address@hidden mailing list and use the Debian bug reporting
> system for problems specific to the Debian package.

Attached is a test case for this. No changelog entry, I don't think this
warrants one.

        Thomas

filtertest.hg
Description: Text document