Build robust and maintainable software with object-oriented design patterns in R. Design patterns abstract and present in neat, well-defined components and interfaces the experience of many software designers and architects over many years of solving similar problems. These are solutions that have withstood the test of time with respect to re-usability, flexibility, and maintainability. R6P provides abstract base classes with examples for a few known design patterns. The patterns were selected by their applicability to analytic projects in R. Using these patterns in R projects have proven effective in dealing with the complexity that data-driven applications possess.

A fast JSON parser, generator and validator which converts JSON, NDJSON (Newline Delimited JSON) and GeoJSON (Geographic JSON) data to/from R objects. The standard R data types are supported (e.g. logical, numeric, integer) with configurable handling of NULL and NA values. Data frames, atomic vectors and lists are all supported as data containers translated to/from JSON. GeoJSON data is read in as simple features objects. This implementation wraps the yyjson C library which is available from .

mlr3: Machine Learning in R - next generation. Contribute to mlr-org/mlr3 development by creating an account on GitHub.

A guide to R development.

When developing a package it is essential to track the changes you make to your code. This is especially vital if they are breaking changes which have implications for any code written that depends on your package, i.e. a major version bump. Although you can always look back at your version control history in git, it is also convenient to have documentation which summarises the changes. This is where the NEWS file comes in.

A fresh approach to package installation. Contribute to r-lib/pak development by creating an account on GitHub.

Task View: PackageDevelopment. Contribute to ropensci/PackageDevelopment development by creating an account on GitHub.