Hi,
I use the Rserve package (RserveEngine.jar) and
the Java client (REngine.jar) from Rforge.
I manipulate R matrices from within Java.
When calling a R function from the Java client on
a R matrix, Rserve gives an error :
"CStack too close to the limit"
That error happens when the matrix has a minimum
size (about 29x10000).
Here is the Java stacktrace I obtain from the execution :
org.rosuda.REngine.Rserve.RserveException: eval failed
at org.rosuda.REngine.Rserve.RConnection.eval(RConnection.java:233)
at main(ConnectToRserve.java:17)
When I call the same function from within R, with
exactly the same environment (I dumped my matrix in a file
which is then loaded in R), there is no error.
Thus, I believe the error comes from the Rserve protocol.
I have written a small Groovy script producing the error.
It is attached to this mail. Please contact me if you want
the complete matrix dump also.
I have already tried to contact Simon Urbanek, the author
of RServe, one month ago. Without success.
Does someone has encountered (and resolved) the same problem as me ?
Is there a more robust alternative to Rserve ?
Before beginning to work to bind the whole R core to the JVM, I would
like to explore all the possibilities to connect to R seamless.
If you know a more appropriate list for that kind of request, I would be
glad to know it.
Thanks for your help,
Matthieu Decorde
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
