You can install the development version from GitHub with:
# install.packages("remotes")
remotes::install_github("yjunechoe/ggtrace")More on the 📦 package website: https://yjunechoe.github.io/ggtrace
{ggtrace} is a functional interface to ggplot2 internals. You need
to know a little bit about {ggplot2} internals to make the most out of
{ggtrace}.
If you prefer watching videos, here are three talks on the topic in increasing complexity:
Broadly speaking, {ggtrace} was designed with two goals in mind:
To help users write more expressive layer code using delayed aesthetic
evaluation. The
family of layer_*() extractor functions return snapshots of layer data
in the internals, to help scaffold a mental model of {ggplot2}
internals as a data wrangling pipeline. See ?`sublayer-data`
To empower users to start developing their own extensions This is
achieved via a family of inspect_*(), capture_*(), and
highjack_*() workflow functions, which provide a functional interface
into the object-oriented design of the internals (the <ggproto> OOP).
You can read the full motivation behind {ggtrace} in the Pedagogical
Philosophy
vignette.
The best way to learn {ggtrace} and {ggplot2} internals is by
playing around and breaking things as you go.
In that spirit, you can take {ggtrace} for a spin on the ggplot2
layer explorer.
🚀 Live App: https://yjunechoe.github.io/ggplot2-layer-explorer/
🔗 Repo: https://github.com/yjunechoe/ggplot2-layer-explorer
