I’m using KNIME to process some data and to finally create a report from the processed data. My idea was to use the R package “pander” to collect the chapters and finally export everything with pandoc to HTML or PDF or so. This fails rather unexpectedly, because the R node does not find pandoc. Running R directly, it is found, though.
So I searched around a little bit, and found that the system path environment is clobbered in the KNIME node.
As you can see, the R home path is added at the head of the list. (I removed the actual path for privacy, but it is correct.) Also: this is on Windows. So backslashes and semicolons are correct as well.
The trailing “null” makes me think, that something goes wrong when starting up the R process. I also didn’t find a place, where to specify an environment variable and possibly override the value.
Is that a bug? Does someone know how to “reset” the environment variable from inside R, so that packages like pander pick up the changed setting?