README.md

In taylordunn/dunnr: Personal R Package of Taylor Dunn

dunnr

The goal of dunnr is to compile various helper functions, templates, and colors that I frequently use.

Installation

# install.packages("remotes")
remotes::install_github("taylordunn/dunnr")

When installing to a project using renv, may have pass the --no-multiarch option as follows:

remotes::install_github("taylordunn/dunnr", INSTALL_opts = c("--no-multiarch"))

This has something to do with renv checking for multiple architectures when building packages, see this issue.

In order to link the cache of dunnr within an renv project, use the following code:

renv::install("taylordunn/dunnr")

Plotting functions

Themes

My go-to ggplot2 theme can be applied with theme_td():

library(dunnr)
library(ggplot2)
library(dplyr)
library(palmerpenguins)

p1 <- penguins %>%
  filter(!is.na(bill_length_mm)) %>%
  ggplot(aes(x = bill_length_mm, y = bill_depth_mm, color = species)) +
  geom_point(aes(shape = species), size = 3, alpha = 0.5, show.legend = FALSE) +
  geom_smooth(method = "lm", formula = "y ~ x",
              se = FALSE, show.legend = FALSE) +
  scale_color_brewer(palette = "Set1") +
  labs(title = "Penguin bill dimensions",
       subtitle = "Bill length and depth for different penguin species",
       x = "Bill length (mm)", y = "Bill depth (mm)",
       color = "Penguin species", shape = "Penguin species",
       caption = "Data from the palmerpenguins package.") +
  facet_wrap(~species, nrow = 1)
p1 + theme_td()

Note that the default font family here is Roboto Condensed, which can be downloaded from Google Fonts (to save headaches on Windows, use the “Install for all users” option so that R can find the font easily). Then to import and load the font:

extrafont::font_import(pattern = "Roboto", prompt = FALSE)
extrafont::loadfonts(device = "win", quiet = TRUE)

Update 2021-08-05: I had some issues importing fonts with extrafont, where it would say “No FontName. Skipping” for every font. It turns out this has to do with the Rttf2pt1 package, which can be fixed be installing an older version. See this GitHub issue.

Alternatively, give another font as an argument, e.g. theme_td(base_family = "Arial").

theme_td_grid() is based on theme_bw() with minor tweaks, such as slightly softer colors and Droid Serif as the default font:

p2 <- penguins %>%
  filter(!is.na(flipper_length_mm)) %>%
  ggplot(aes(x = flipper_length_mm, y = species)) +
  geom_violin(aes(fill = species), show.legend = FALSE) +
  labs(x = "Flipper length (mm)", y = NULL) +
  facet_wrap(~year) +
  scale_fill_brewer(palette = "Set1") +
  scale_x_continuous(breaks = seq(180, 220, 20))
p2 + theme_td_grid() +
  theme(panel.grid.major.y = element_blank())

theme_td_grey() is a low contrast theme that emphasizes title and facet text:

p1 + theme_td_grey(base_size = 12)

theme_td_minimal() is based on theme_minimal():

p1 + theme_td_minimal()

Palettes

The td_colors list has some of my commonly used palettes, which can all be accessed via the td_pal() function. A qualitative palette with pastel colors:

scales::show_col(td_pal(palette = "pastel6")())

A diverging blue-to-red palette:

scales::show_col(td_pal(palette = "div5")())

Scales

These palettes can be easily applied to plots with the scale_color_td() and scale_fill_td() functions:

library(patchwork)
p3 <- diamonds %>%
  filter(carat >= 2.2) %>%
  ggplot(aes(x = price, y = cut, fill = factor(cut, ordered = FALSE))) +
  geom_boxplot() +
  theme_td() +
  theme(legend.position = "none") +
  cowplot::background_grid(major = "x")
(p3 + scale_fill_td() + labs(title = "pastel6")) +
  (p3 + scale_fill_td("div5") + labs(title = "div5"))

p4 <- faithfuld %>%
  ggplot(aes(waiting, eruptions, fill = density)) +
  geom_tile() +
  theme_td() +
  scale_x_continuous(expand = c(0, 0)) +
  scale_y_continuous(expand = c(0, 0))
p4 +
  scale_fill_td(palette = "div5", type = "continuous", reverse = FALSE)

remove_axis()

A common tweak I make to plots is to remove unnecessary axes, like in density plots. The remove_axis() function saves a few lines of code:

p5 <- penguins %>%
  filter(!is.na(flipper_length_mm)) %>%
  ggplot(aes(x = flipper_length_mm)) +
  geom_density(aes(fill = species), alpha = 0.5) +
  scale_fill_brewer(palette = "Set1") +
  scale_y_continuous(expand = expansion(c(0, 0.1))) +
  labs(x = "Flipper length (mm)")
p5 + theme_td() +
  remove_axis("y")

add_facet_border()

Sometimes when plotting facets, it helps to have some visual separation in the panels. The add_facet_border() function is a quick way to do so without adjusting theme() elements directly:

p1 + theme_td() + add_facet_borders()

set_geom_fonts()

The set_geom_fonts() function sets the default fonts of geom_text() and geom_label(). If no font is specified, it will automatically take the font from the current ggplot2 theme:

library(glue)
p6 <- penguins %>%
  count(species, island, name = "n_penguins") %>%
  ggplot(aes(y = species, x = n_penguins)) +
  geom_col(aes(fill = island)) +
  geom_label(aes(label = glue("{island}: n = {n_penguins}")),
             position = "stack", hjust = 1) +
  scale_fill_brewer(palette = "Set1") +
  scale_x_continuous(expand = expansion(mult = c(0, 0.1))) +
  labs(title = "Number of penguins per island",
       subtitle = "Different font",
       x = "Number of penguins", y = NULL,
       caption = "Data from the palmerpenguins package") +
  theme(legend.position = "none")

# By default, geom_label() doesn't have the same font as theme_td()
p6 + theme_td() +
  theme(legend.position = "none")

# But if we set the theme
theme_set(theme_td())
# And then set fonts with the helper function
set_geom_fonts()

p6 + theme_td() +
  theme(legend.position = "none") +
  labs(subtitle = "Same font")

set_palette()

The set_palette() function is a convenient way to quickly set the default discrete palette (via ggplot2.discrete.fill and ggplot2.continuous.fill options) and the default continuous palette (via ggplot2.continuous.fill and ggplot2.continuous.color):

set_palette()
(p3 + labs(title = "Discrete palette")) +
  (p4 + labs(title = "Continuous palette"))

Personal website

There are a couple utility functions for working with my website https://tdunn.ca/, which was previously built with distill but recently converted to quarto.

get_quarto_source() and get_distill_source() generate URLs linking to the source code and renv.lock files of my posts:

# By default, returns a HTML tag
get_quarto_source(date = "2021-05-18", slug = "tidytuesday-2021-week-21",
                   html_label = "Link to source code")
#> [1] "<a href=\"https://github.com/taylordunn/tdunn-quarto/tree/main/posts/2021-05-18-tidytuesday-2021-week-21\">Link to source code</a>, <a href=\"https://github.com/taylordunn/tdunn-quarto/tree/main/posts/2021-05-18-tidytuesday-2021-week-21/renv.lock\">R environment</a>"
# Alternatively, just get the URL
get_quarto_source(date = "2021-05-18", slug = "tidytuesday-2021-week-21",
                   html_tag = FALSE, renv_link = FALSE)
#> [1] "https://github.com/taylordunn/tdunn-quarto/tree/main/posts/2021-05-18-tidytuesday-2021-week-21"

There is also a function create_post_tdunn(), which wraps distill::create_post() with some placeholder text and code.

create_post_tdunn("Post title", date = Sys.Date(), date_prefix = TRUE,
                  draft = TRUE, open = FALSE)

Todo

• A create_post_quarto() for my new site that mimics create_post_tdunn()
• Number formatting:
  • round() but keep trailing zeroes via sprintf()

taylordunn/dunnr documentation built on Aug. 8, 2022, 12:28 p.m.