Style guide 🧑🏻‍💻

R style guide ✅

The tidyverse style guide

pipes — |> · 🌬️

• one line for each step in a pipe
• first line pipe with name of new object and assignment operator (only)
• use Base-R pipe |> over Tidyverse-R pipe %>% for R >= 4.2
  • placeholder _ (|>) instead of . (%>%)
  • check Use native pipe operator in RStudio

data · 🔢

• use coherent prefixes for groups of data // dt dt_ctry dt_ctz
• prefix raw_ for raw data read into object with

plots · 📊

• prefix pl for plot objects // pl pl1 pl2
• data for plot only into pl_dt object

models · 🔬

• prefix mo for model objects // mo mo_lm mo1

workflow · ⚙️

• .gitignore sources with source__ prefix
  • [ single option to ignore copyright protected, personal or sensible sources ]
• final checks
  • restart R session and run all
  • check all object names
  • check console messages and warnings
• reproducible environments with Rocker
  • see Dockerfile and docker-compose.yml examples

packages · 📦 // see snippet below

• use conflicted to avoid function name conflicts
  • load first with library(conflicted)
  • specify preferred functions (e.g. conflicts_prefer(dplyr::filter()))
• load library(tidyverse) after conflicted
• add brief description why package is loaded
  • library(broom) # models // tidy results
• order packages loaded by thematic groups
  • plot — patchwork ggrepel
  • models — broom · ggeffects
• specify single usage of package with ::
  • comment below loading packages block // # janitor::
  • no comment for usage tidyverse of packages that are not loaded with library(tidyverse) (e.g. glue readxl)
  • [ reduce potential name conflicts and code completion options ]

---

Snippets · ✂️

Loading packages

library(conflicted) # create errors for function name conflicts
conflicts_prefer(dplyr::filter, .quiet = TRUE)

library(tidyverse)

# order alphabetically in section and provide category with brief description
library(broom)      # models // tidy results
library(ggeffects)  # models // visualize model effects
library(patchwork)  # plots  // arrange plots
library(reactable)  # layout // interactive data tables
library(sf)         # maps   // spatial data tools

# DT:: skimr:: viridis::

---

Note — These are my personal best practices for code formatting in Tidyverse R. They have evolved over time and not all my previous code may follow these guidelines.