Re: saving function handles

[Richard Crozier]

On 15/11/2012 14:04, Jordi Gutiérrez Hermoso wrote:
> On 14 November 2012 12:55, Muhali <address@hidden> wrote:
> > But if during the saving of f `foo´ is known from a relative path, e.g. path
> > = ... fun ... (instead of path = ... /path/to/fun ...), isn't it more
> > logical if only the relative path is saved?
> This proposal might make sense, but I don't see how to implement it.
> Do you have precise suggestions how to do so? How can we know in
> advance if an absolute or a relative path should be used? Why and when
> should a relative path be preferred? What should the path be relative
> to?
>
> - Jordi G. H.

I have had similar problems with saving function handles on one computer 
and trying to load them on another with a different directory path for 
the same function. I got around the problem using strings and feval 
instead, but this might not be so easy in some cases.

It might be nice if octave fell back on 'foo' in the local path (with a 
turn-offable warning maybe?) if /path/to/fun/foo.m is not accessible?

I can see how this might cause problems though. I don't see how relative 
paths could work.

Richard

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.