J. Miguel Marin schrieb:
I hoped before, that something like
sl<-comCreateObject("matlab.Application")
should work with SciLab (e.g. "SciLab.Application")
But as R> sl<-comCreateObject("STATCONNECTORSRV.StatConnector")
sl
<pointer: 0x0006fe34>
attr(,"class")
[1] "COMObject"
R> comInvoke(sl,"Init","SciLab")
NULL
R> x<-1:10
R> dim(x)<-c(2,5)
R> comInvoke(sl,"SetSymbol","x",x)
NULL
R> y<-comInvoke(sl,"Evaluate","x'")
R> y
NULL
I cannot understand why matlab is identified but the parameter
"SciLab" is not, even when the Scilab's console is under execution.
I wonder whether this is related with the Registry of Windows, or the
current version of SciLab (I use the latest version 5.01, just linked
yesterday), or I have any problem with rcom, because I usually call R
from MatLab with D(COM), but I have not tried in the opposite
direction: matlab from R. Although, with
comCreateObject("matlab.Application") the matlab console is openned
so I assumed rcom worked fine.
My main intention was to substitute propietary code, like MatLab, by
free programs like SciLab or Octave, but perhaps things are not so
easy.
Anyway, thank you for your kind responses.
Hello,
excuse for my insistence. I have only a little experience in MatLab, so
I am not an expert.
If I write sl<-comCreateObject("STATCONNECTORSRV.StatConnector")
it works
but comInvoke renders "NULL"
In my example, I can see comInvoke to be called 4 times.
Only the one I'm assigning to y should return a result.
As a simple question, method="Scilab" is refered to the main *.exe
file from SciLab? Or another "thing"?
Which is the method "Scilab"?
I am afraid I do not understand the logic behind comInvoke, because I
do not know where rcom must find the method SciLab.
Aaaah, I see. You mean comInvoke(sl,"Init","SciLab")
First of all, this should tell us how important reasonable quoting is.
Second, this calls the method "Init" on the object "sl" and passes
the parameter "SciLab".
Sorry for my lack of experience.
Best
ACTXSERVER Creates a COM Automation server
In SciLab? If yes, you should be able to easily access R from Scilab.
Thomas
_______________________________________________
Rcom-l mailing list
Rcom-l@mailman.csd.univie.ac.at
http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
More information (including a Wiki) at http://rcom.univie.ac.at
jm~
_______________________________
J. Miguel Marin
http://www.est.uc3m.es/jmmarin
Dep. of Statistics
University Carlos III of Madrid
Spain (E.U.)
_______________________________
_______________________________________________
Rcom-l mailing list
Rcom-l@mailman.csd.univie.ac.at
http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
More information (including a Wiki) at http://rcom.univie.ac.at
