[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: The future of Octave
From: |
Kevin Straight |
Subject: |
Re: The future of Octave |
Date: |
Sun, 10 Dec 2000 10:42:35 -0800 (PST) |
On Sat, 9 Dec 2000, Paul Kienzle wrote:
> I've been hesitant to pursue this option for three reasons.
>
> 1) The need to learn a second language for programming callbacks
> 2) The overhead of installing and running a second large interpreter
> 3) The ungainliness of tcl/tk
>
Tcl is too slow, especially since everyone is obviously concerned about
execution time. I say we should use C++m which Octave can already
dynamicly load.
The embedded interpreter model sounds like it has merit, though. Let me
see if I understand you:
*) we would have a second process in memory that knew how to draw
all the widgets and process callbacks.
*)We would write an API of functions called in Octave, which would
communicate with this process (with pipes, I guess, but I only know
unix/linux and don't know if this would be portable) to ask for widgets,
then request their states (weather a button was pressed, for instance)
Again, I only know Linux, but I would say that (as long as you could use
dynamic linking) you could run something like this in less than one meg of
user RAM. I would worry about stability, though.
> It occurs to me, though, that we can apply the same approach that I would
> recommend for an embedded GUI, which is a widget register defining the
> range of widgets available, and their properties and callbacks (including
> some packing widgets). The widget definition can include all the bits
> of the foreign interpreter that are needed, leaving your Octave scripts
> with pure Octave code.
>
> The overhead of the second interpreter is offset by the benefit of having
> the GUI in a separate process. Even if we did our own specialized
> GUI in a separate process, we would still have to incur that overhead
> (though presumably the specialized interpreter would be very simple).
>
==========================
Kevin Straight
University of Idaho
www.uidaho.edu/þstra9456
==========================
-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.
Octave's home on the web: http://www.octave.org
How to fund new projects: http://www.octave.org/funding.html
Subscription information: http://www.octave.org/archive.html
-------------------------------------------------------------
- Re: The future of Octave, (continued)
Re: The future of Octave, John W. Eaton, 2000/12/08
Re: The future of Octave, Stef Pillaert (KAHO), 2000/12/08
- Re: The future of Octave, Paul Kienzle, 2000/12/09
- Re: The future of Octave, Rafael Laboissiere, 2000/12/09
- Re: The future of Octave, j . logsdon, 2000/12/09
- Re: The future of Octave,
Kevin Straight <=
- Re: The future of Octave, Przemek Klosowski, 2000/12/11
- Re: The future of Octave, Trond Eivind GlomsrØd, 2000/12/11
- Re: The future of Octave, João Cardoso, 2000/12/11
- Re: The future of Octave, Trond Eivind GlomsrØd, 2000/12/11
- Re: The future of Octave, Kevin Straight, 2000/12/13
Re: The future of Octave, Paul Kienzle, 2000/12/08
Re: The future of Octave, Andy Adler, 2000/12/08
Re: The future of Octave, Kevin Straight, 2000/12/08