Disordered vectors in R: introducing the disordR package

TL;DR

The disordR package in R introduces a way to handle unordered map objects, preventing forbidden operations and enabling transparent manipulation of disordered vectors, with applications demonstrated in polynomial manipulation.

Contribution

It provides the first implementation of disordered vectors in R, facilitating safe handling of unordered map objects and integrating with existing packages.

Findings

DisordR enforces safe operations on unordered vectors.

The package simplifies manipulation of disordered data structures.

DisordR is a dependency for several R packages.

Abstract

Objects in the {\tt stl map} class of {\tt C++} associate a value to each of a set of keys. Accessing values or keys of such an object is problematic in the R programming language because the value-key pairs are not stored in a well-defined order. This document motivates and discusses the concept of "disordered vector" as implemented by the {\tt disordR} package which facilitates the handling of {\tt map} objects. Values and keys of a map are stored in an implementation-specific way so certain extraction and replacement operations should be forbidden. For example, if values are real, then the "first" value is implementation specific\ldots but the maximum value has a well-defined result. The {\tt disordR} package makes forbidden operations impossible while allowing transparent R idiom for permitted operations. An illustrative R session is given in which the package is used abstractly, without reference to any particular application, and then shows how it can be used to manipulate multivariate polynomials. The {\tt disordR} package is a dependency of {\tt clifford}, {\tt freealg}, {\tt hyper2}, {\tt mvp}, {\tt spray}, {\tt stokes}, and {\tt weyl}. The {\tt disordR} package is available on CRAN at \url{https://CRAN.R-project.org/package=disordR}.