help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: octave from java


From: frank buloup
Subject: Re: octave from java
Date: Fri, 15 Apr 2011 15:18:06 +0200
User-agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; fr; rv:1.9.2.15) Gecko/20110303 Thunderbird/3.1.9

Le 15/04/11 14:08, Kim Hansen a écrit :
On Fri, Apr 15, 2011 at 10:47, frank buloup <address@hidden> wrote:
Hello,

I'm trying to use Octave from java but I'm facing a strange problem that
have already
been encountered. From java program, I'm using Runtime exec to start an
external
process, getting back input, output and error streams; sending command using
output stream and parsing the result from input stream. Everything is
working fine
until a bad command is sent to Octave (a command containing a call to a
function
from a package that has not been installed for instance) : octave seems to
crash
and I 've not found any log file that could help me to debug something.

This problem has already been encountered, please see thread :

http://octave.1599824.n4.nabble.com/Octave-from-Java-td1630041.html#a1630055
The problem with handling bad commands has been solved in JavaOctave
(http://kenai.com/projects/javaoctave/). Have you tried JavaOctave?

Regards,
Hello Kim,

thanks a lot for your reply, you gave me the solution : wrap every
sent commands in an "eval()" function. THANKS YOU VERY MUCH !!!!!!!

I was aware of the existence of javaoctave, but never tried it. The solution
that I've used is not so generic as your and is very linked to our needs.

But when I was looking for the solution, I've downloaded src of javaoctave and
had a look at it. I've seen that you've wrapped almost every octave type in a specific
octave class. Why did you choose to do that ? Was it impossible to wrap octave double, or
matrix of doubles in native java type double and double[] for instance ? Have you ever test your
code in manipulating a huge amount of data (say 500*60000 array of double for instance) ?

Again thanks a lot for your help,
With best regards,
frank

--

Frank Buloup

04 91 17 22 71

Institut des Sciences du Mouvement

UMR 6233 CNRS & Université de le Méditerranée

163 avenue de Luminy

13288 Marseille Cedex 9

Fax : 04 91 17 22 52


reply via email to

[Prev in Thread] Current Thread [Next in Thread]