<!-- README.md is generated from README.Rmd. Please edit that file -->

ggdollar

The goal of ggdollar is to provide a more intuitive way to set theme() arguments when using the package ggplot2.

theme() is a complex function, we provide nested lists of functions that should guide the users to what they through a sequence of limited choices thanks to autocomplete.

It’s a draft, but works ok already.

Installation

remotes::install_github("moodymudskipper/ggdollar")

Intro

ggset, ggdo and ggremove are nested lists of functions.

ggset lets you dig to the element you want to change, and apply those changes, so that ggset$panel$background(fill = "blue") is the same as theme(panel.background = element_rect(fill = "blue")). The arguments are most of the time ultimately forwarded to an element_*() function and you can change many attributes of a component at once. ?margin will provide the description of most parameters.

ggdo on the other hands starts from the action you want to perform, For instance the equivalent to the above would be ggdo$fill_background$panel("red"). Typically with ggdo we set only a single or a couple of attributes at once.

ggremove is like ggset but helps removing components, it sets the relevant theme() arguments to element_blank(), NULL or "none".

For instance ggremove$panel$background() is the same as theme(panel.background = element_blank())

There would have been many other ways to organize these lists and we went for an partially arbitrary reorganization rather than a fully systematic one.

library(ggdollar)
library(ggplot2) 
ggplot(iris, aes(Sepal.Width, Sepal.Length, color = Species)) + geom_point() + 
  ggset$panel$background(fill = "white", color = "black") + # top down approach to change theme
  ggdo$set_text$angle$axis$x$bottom$text(45) + # bottom up approach
  ggremove$legend$all() # removing element

<img src="man/figures/README-example-1.png" width="100%" />

To do :

• find better names
• support in ggdo attributes that aren’t related to element_*() functions
• Play around and see what doesn’t feel right
• better doc
• cleaner code
• display native ggplot2 code