After a well-deserved break last month, we are back with great news: initial support for CRAN and Bioconductor packages using C++ code, including those that use the popular Rcpp framework. More and more packages use Rcpp so this is a big deal for us.
Our goal with Renjin is to run plain old R code just as fast as C++, but being able to use existing packages written using native languages is also important. Renjin's tool chain compiles packages that use C++ to pure JVM bytecode, simplifying deployment and allowing R packages to managed with the JVM's rich tooling ecosystem.
Other package news
Besides Rcpp, we have also added drop-in replacements for digest and plyr. Other notable packages to have passed a significant level of compatibility with Renjin are randomForest and foreach.
All of this work has also resulted in a succesful build of ggplot2 which cannot be used in Renjin as we do not support graphics yet, but it was blocking the build of a lot of other packages. So all in all we are making great progress and are happy with the results so far.
We hope that you can give these packages a try and provide us with your feedback!
GNU R source code quote of the month
“Lazy evaluation and method dispatch based on argument types are fundamentally incompatible notions. The results here are ghastly.” — band.c in GNU R
Alex will present at RIOT after useR!2016
If you are attending the useR!2016 conference in Stanford this year and you're interested in alternative engines for the R programming language, then you may want to stick around a little longer to attend the RIOT workshop on Sunday July 3rd.
Alex will be presenting his work on Renjin's JIT loop compiler which is a work-in-progress to speed up the execution of loops.
Full details on the RIOT workshop are available online at http://riotworkshop.github.io/. Alex is also available on Monday July 4th if you'd like to discuss Renjin in person: just reply to this newsletter and we can arrange a meeting in the Stanford / San Francisco area!
Give your input for the new documentation
The documentation for version 0.7 of Renjin has been up for quite a while now at docs.renjin.org but quite a few things have changed in version 0.8 so we are planning to update the documentation very soon.
Let us know what you would like to see added or improved in the documentation by opening an issue on GitHub at https://github.com/bedatadriven/renjin-docs/issues. If you are comfortable with GitHub and reStructuredText, you can also fork the documentation repository, make your changes and send us a pull request.
Using Spark(R)? With Renjin?
We want to hear from you!
We are eager to learn about how people and businesses are using Spark with R so we can provide a seamless Spark-Renjin integration. We also hope that you want to show your love for the project by providing us with your testimonials on using Renjin.
So reply to this message or send us a message at firstname.lastname@example.org if you'd like to help us out.