RCaller Java Premature end of file aka XML file is empty

Question

I am writing program using RCaller 2.1.1-SNAPSHOT. Problem is when I use code from external library it says Routput file is empty. Here is my code:

  Random random = new Random();


  RCaller caller = new RCaller();
  RCode code = new RCode();

  caller.setRscriptExecutable("C:/Program Files/R/R-2.14.2/bin/x64/Rscript.exe");

  double[] data = new double[100];

  for (int i = 0; i < data.length; i++) {
    data[i] = random.nextGaussian();
  }

  code.addDoubleArray("x", data);

  code.addRCode("setwd('C:/Radek/')");
  code.addRCode("wd=list.files()");
  code.addRCode("library(Biobase)");
  //code.addRCode("targets=read.AnnotatedDataFrame('targets.txt',row.names=1,as.is=TRUE)");// WHEN I USE THIS IT CRASH
  code.addRCode("my.mean<-mean(x)");
  code.addRCode("my.var<-var(x)");
  code.addRCode("my.sd<-sd(x)");
  code.addRCode("my.min<-min(x)");
  code.addRCode("my.max<-max(x)");
  code.addRCode("my.standardized<-scale(x)");

  code.addRCode(
          "my.all<-list(mean=my.mean, variance=my.var, sd=my.sd, min=my.min, max=my.max, std=my.standardized)");
  caller.setRCode(code);
  caller.runAndReturnResult("wd");

  String[] results;
  results = caller.getParser().getAsStringArray("wd");
  System.out.println("Mean is " + results[0]);

I checked this: 1. RUniversal is installed and loaded 2. library for using function is downloaded and installed. 3 When I put rCaller request from file generated by RCaller into R it works. 4. slashes in RScript path are good because I checked results with commented problematic line and it works.

Can someone help me with this?

Answer 1

this is generally about the differences of the installer and the loader user of package. The latest and experimental RCaller 2.2.0 does not require Runiversal. If is your problem still current you can try it out and write here if problem still exists.You can follow the links for download at official blog page of RCaller

Answer 2

In the part of your code

code.addRCode(
      "my.all<-list(mean=my.mean, variance=my.var, sd=my.sd, min=my.min, max=my.max, std=my.standardized)");
caller.setRCode(code);

String[] results;
results = caller.getParser().getAsStringArray("wd");

change the part

caller.runAndReturnResult("wd");

to

caller.runAndReturnResult("my.all");

so your prepared list is returned from R to Java. Then use the code

double[] results;
results = caller.getParser().getAsDoubleArray("mean");

and, finally print your returned mean to screen

System.out.println("Mean is " + results[0]);