At some point I’ll probably get round to putting {tidyods} and {shrthnd} on CRAN, but in the meantime I’ve set up a repository on R-universe that hosts these two packages.

About a year ago I started development of a new R package to deal with dirty data, specifically to process the use of shorthand, symbols and other annotations in columns of spreadsheets and tables. I realised I’ve not blogged about it, and thus if one subscribes to “blog-driven-development”1 then if I’ve not blogged about it the package doesn’t exist.

I created govukhugo in my last couple of years working for the Cabinet Office’s Analysis & Insight Team. But in July last year I left that role, and government. But I’m aware, via sources1, there have been some requests within government for updates, including support for Quarto. So I’ve thought a bit about where next for the project2. This post has turned out to be a bit of a stream of consciousness/potted history of the development of govukhugo, so there’s also a TLDR summary if you want to skip to the end.

I’m working on a project where one of the outputs is a deck of Microsoft PowerPoint slides that act as a reference guide1. Outside of contents, introductory and annex slides the bulk of the slides follow the same format with a simple table providing metadata about a suite of around 90 indicators. Using the gtcars dataset I’ll demonstrate how I’ve used Quarto and the {officer} package to automate some of the workflow.

Last summer I started the development of {tidyods}, an R package for reading cells in ODS (OpenDocument Spreadsheet) files, and effectively a {tidyxl} equivalent for ODS files. After developing the first iteration of the package in June I left it alone, but I’ve recently revisited the package and refactored the code which has improved performance.