Re: problem with sum (...)

[Ben Abbott]

On Nov 4, 2007, at 1:00 PM, Søren Hauberg wrote:

> Ben Abbott skrev:
> > Søren,
> > The patch didn't work for me. However, I was able to apply in  
> > manually. What exactly should I have done to have it apply cleanly?
> The patch is against the CVS version. Did you apply it to a recent  
> CVS checkout?

I updated my cvs yesterday. One problem may be that I don't use the  
same directory structure as you, so I tried to apply the patch  
directly the ismember.m. For example,

patch ismember.m ismember.patch

Does this work for you?

> > In any event, you've done quite a bit more work on this than I. I  
> > had added some additional tests to my version that produced  
> > failures in the original. They pass in yours.
> Are these tests part of the current set of tests? If not, then  
> please send them so they can be part of the test suite.

I'm happy to send a patch, but would prefer that I'm able to create  
and apply patches correctly first ... although for such a simple  
change, I can send the tests to you. I'll look through the present  
tests to see if any others might be added for completeness.

> > I was attempting to simplify the "Magic" portion ... actually  
> > trying to incorporate numel(a)==1 and numel(a)>1 into the same code.
> To be honest, I don't truly get the magic part. So, if you have an  
> easier-to-understand idea that is just as fast, then it would be  
> great.

I'm sure I can come up with a simpler algorithm, but will have to  
test it in order to determine if it is as efficient.
> > I'll play with the idea a bit longer, but since everything appears  
> > to work, I'm not sure it makes sense to tweak a solution that  
> > already works.
> The part of my implementation that handles the 'rows' argument, is  
> just horrible. It's an evil-and-dirty hack, and it really should be  
> improved. So if you have a better idea, it would be great.

I'll take a look.

Ben