This is half a rant and half a question, so bear with me.
I'm trying to generate some fairly complex publication quality graphs from an XML datasource. I've been using XML::Twig
to munge the data and tab delimited text files. I then take the tab deliminated files, load them into R, run some descriptive stats, and build my graphs. A quick google search reveals that there is a perl interface to R, RSPerl
. In theory, I should be able to completely automate my work, however I haven't had any luck getting RSPerl to work. The installation procedure doesn't install the perl modules anywhere in @INC. I tracked down all the necessary paths, but the R module is still complaining about my environment and dying an ugly death. The RSPerl documentaion hasn't been terribly helpful and R's author's insightful selection of "R" as a name for their program has essentially rendered google useless. I'm getting a bit frustrated, so I thought I would turn to the Monestary for help.
Has any here successfully installed and used RSPerl? If so, could you give me some pointers on how to get this working? If it helps, I've tried getting this working on a fairly vannila Gentoo Linux installation running R 1.61 and perl 5.8.0 compiled from source and a RedHat 7.3 box running R 1.51 and perl 5.6.1 installed from RPM.
My second question is whether there is a better option for producing publication quality graphs from perl. I need to produce semi log scaled graphs with specific dimensions and some ideosyncratic layout conventions. Anyone have any suggestions? Thanks in advance for your help.