RE: Overriding builtins

[Jurgen Pletinckx]

| >> Is that possible? If so, how do I go about it?
| >I don't think you can overwrite keywords like "false" as I believe
| >they 
| >are part of the language and not just functions (not sure on this, 
| though)
| 
| AFAIK in Octave 2.9.9+ CVS 'false' is not a keyword.
| We can probe this using the function iskeyword().
| iskeyword('false') returns 0.

Thanks, all - 

I will see if I can get 2.9.9 running. However, the system I will be
running octave on is an aging production cluster, running RH 7.2, 
kernel 2.4.17, gcc 2.96 - I expect an uphill struggle getting 2.9.9
compiled. 

(My current tests are with octave 2.1.73, compiled on a slightly less
ancient gcc 3.04, and that already required a bit of source tweaking.)

I suppose it would be worthwile, in addition, to alter the scripts to 
use an alternative for false(n,m), in order to see what other compati-
bility problems are lurking further down the road.


I originally selected 2.1.73 because it was indicated as the current
testing 
version - "Most people who are not actively involved in Octave's
development 
will want to use this version."

However, I now read in the announcement for 2.9.9
https://www.cae.wisc.edu/pipermail/octave-maintainers/2006-October/000
743.html

"If there are no really serious bugs reported in the next week or so,
then this version will replace 2.1.73 as the "testing" version." 
Were there any serious concerns for 2.9.9?



-- 
Jurgen Pletinckx
AlgoNomics NV