Re: adopting functions from Octave-forge

[John W. Eaton]

On 29-Sep-2005, Quentin Spencer wrote:

| Did you mean moved from octave-forge to octave?

Yes.

| >Ignoring sparse functions, we currently have the following overlap:
| >
| >  builtin   dispatch       hankel    mu2lin     randn       struct  
| >  cellstr   dispatch_help  isa            ndims      rmfield     tf2zp       
    
| >  char           double         isequal   orient     setdiff     toeplitz
| >  chol           fieldnames     isfield   polyder    sortrows    tril        
  
| >  complex   full        ismember  polyderiv  str2double  triu          
| >  deal           gammaln        issparse  polygcd    strcmpi     unique  
| >  detrend   gammaln        isunix    print   strmatch    unix          
| >  dispatch  grid        lin2mu    rand       strncmp     zp2tf   
| >
| >Some of these may still be different in Octave-forge, so the list
| >should also keep the current status of the function so that we don't
| >end up with two diverging versions of each of these.
| 
| Maybe we should use the wiki for this. However, I'm under the impression 
| that the wiki has become much less active since the security was 
| tightened. Maybe we need to consider a new way of dealing with it. 

The wiki is fine, if it is working.  Or, I can maintain a list and
keep it in the Octave CVS.  It doesn't matter to me.  I just wondered
whether anyone was keeping list.  I think we should have one, or we
are asking for more future confusion about which functions are the
latest and greatest.  Also, after a function has been moved to Octave,
that version should be the "master" source.  How can we make sure that
we won't modify the forge version and forget to update the Octave
version?

jwe