On Mon, May 17, 2010 at 10:26 AM, Alois Schlögl
<[email protected]> wrote:
> However, if there is no agreement that having
> "free toolboxes for matlab" is a worthwhile goal, it would expect similar
> opposition than trying to make the source code compatible.

I think the opposition is that your proposing rewriting octave code
for the sole purpose of compatibility with a translator that targets
octave. The problem I have with these changes is that some of them
defeat useful semantics that many of us view as improvements over
matlab's language.

Don't take it too harshly. Sanitizing input is a valid way to approach
this sort of problem. I think you just underestimated how much others
may like octave's extensions.

> The latter would be much easier to realize than using a parser, I think.

It may seem easier at first. The problem is that I don't think there
is any way to validate that the translated code will actually work in
Matlab without running it in Matlab. In the long run a fully working
parser could be modified to compile m files directly to oct files. Who
needs JIT then?


--judd

------------------------------------------------------------------------------

_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to