Rserve
About Rserve
|
Please read before downloading/using the Windows version of Rserve!
Binaries for R 2.4.1 and higher are now available via CRAN. Use install.packages("Rserve") in R to install the latest Rserve release. The R package contains both regular and debug version of Rserve, see ?Rserve. Installation: Binaries are now provided on CRAN, so you can use install.packages("Rserve") to install and library(Rserve); Rserve() to start Rserve. This sets PATH according to the R used and then starts the corresponding exe file. Alternatively you can bopy the binary Rserve.exe to the same directory where R.dll is located (e.g. C:\Program Files\R\R-2.12.1\bin\i386). Rserve automatically detects latest installed version from the registry. If you un-install any version of R, you have to re-install R, because the un-installation process removes entries from the registry even if you have another version of R installed. (Optionally Rserve can be started from anywhere as long as the R.dll is in the PATH, but this is not required if Rserve is copied into the bin directory). Compiling your own Rserve
Rserve can be compiled from sources just like any other R package using simply Details concerning Windows version of Rserve
Since the Windows operating system doesn't support fork method for spawning copies of a process, it is not possible to initialize R and use initialized copies for all subsequent connections in parallel. Therefore the Rserve for Windows supports no concurrent connections. This implies that all subsequent connections share the same namespace and sessions (as in >=0.4 version on unix) cannot be supported. It is still possible to start multiple Rserves to handle multiple connections (just make sure you use different port for each one). Note: if you use non-interactive mode, you may have to set options(error=function() NULL) in order to prevent R from quitting on error - see FAQ. |