An update to our package RcppNLoptExample arrived on CRAN earlier today marking the first update since the intial release more than four year ago. The nloptr package, created by Jelmer Ypma, has long been providing an excellent R interface to NLopt, a very comprehensive library for nonlinear optimization. In particular, Jelmer carefully exposed the API entry points such that other R packages can rely on NLopt without having to explicitly link to it (as one can rely on R providing sufficient function calling and registration to make this possible by referring back to nloptr which naturally has the linking information and resolution). This package demonstrates this in a simple-to-use Rcpp example package that can serve as a stanza.
More recent NLopt versions appear to have changed behaviour a little so that an example we relied upon in simple unit test now converges to a marginally different numerical value, so we adjusted a convergence treshold. Other than that we did a number of the usual small updates to package metadata, to the README.md file, and to continuous integration.
The (very short) NEWS entry follows:
Changes in version 0.0.2 (2025-03-09)
Updated tolerance in simple test as newer upstream nlopt change behaviour ever so slightly leading to an other spurious failure
Numerous small and standard updates to DESCRIPTION, README.md, badges, and continuous integration setup
Courtesy of my CRANberries, there is also a diffstat report for this release. For questions, suggestions, or issues please use the issue tracker at the GitHub repo.
This post by Dirk Eddelbuettel originated on his Thinking inside the box blog. If you like this or other open-source work I do, you can now sponsor me at GitHub.