In caprice-j/ggconf: Simpler Appearance Modification of 'ggplot2'

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-"
)

ggconf

ggconf provides theme2(), a flexible ggplot2::theme() interface.

library(ggplot2)

Usage

g <- ggplot(iris) + geom_point(aes(Sepal.Width, Sepal.Length))

g + theme2(ax.txt(sz=20, f="bold"),
           ax.line(col='gray60', sz=2),
           panel.bg(fill="white")
          )

g <- ggplot(iris) + geom_point(aes(Sepal.Width, Sepal.Length))
g + theme(       axis.text = element_text(size=20),
                 axis.line = element_line(colour="gray60", size=2),
          panel.background = element_rect(fill="white"))
# gg iris + point Sepal.W Sepal.L col=Spec siz=Petal.W + theme text: size=20 face="bold"

The following ggplot2 command generates the same plot.

g + ggplot2::theme(axis.text = element_text(size=20, face="bold"),
                   axis.line = element_line(colour="gray60", size=2),
                   panel.background = element_rect(fill="white")
                  )

Getting Started

If you replace your ggplot2::theme() with ggconf::theme2(), ggconf would work. All of the followings return the same plot, and you can use the style you prefer the most.

g + theme( axis.text = element_text(size=20, face="bold")) # Style 1: ggplot2 default (50 characters)
g + theme2(axis.text = element_text(size=20, face="bold")) # Style 2: ggconf
g + theme2(axis.text(size=20, face="bold"))                # Style 3: ggconf without element_text()
g + theme2(ax.txt(sz=20, face="bold"))                     # Style 4: ggconf shorter but readable
g + theme2(at(z=20, f="bold"))                             # Style 5: ggconf shortest (25 characters)

Features

Partial Match

Even if the unique identification is not possible for specified elements (e.g. theme element names or arguments), ggconf tries to execute its best estimate instead of just returning an error.

For the input theme2(ax.txt(sz=20, fc="bold"), ax.ln(c='gray60'), panel.bg(fill="white")), ggconf performs partial matches six times.

• theme element names

  • ax.txt matches axis.text. You can even write a.t or at.
  • ax.ln matches axis.line. You can even write a.l or al.
  • panel.bg matches panel.background. You can even write pnl.bg.
    • p.bg matches plot.background according to edit distance.

• theme configuration arguments

  • sz or z match size.
  • f matches face (fontface).
    • fill needs to write not just f but fi.
  • c matches colour.

Installation

# install.packages("devtools")
devtools::install_github("caprice-j/ggconf")

• If you get no appender.console() error, please install.packages('rly').

Example

The goal of ggconf is to make it less stressful to finalize your plots.

Raw ggplot2 plot

The following plot uses the ggplot2 default appearance settings.

suppressPackageStartupMessages(library(dplyr))
gg <- ggplot(mtcars[1:20, ] %>% tibble::rownames_to_column() %>% 
             mutate(car_name = rowname, maker = gsub(" .*", "", car_name) ) ) + 
      #geom_label(aes(mpg, qsec, label = substr(car_name, 1, 13), color=maker),
      geom_point(aes(mpg, qsec, color=maker), size=8) +
      geom_text(aes(mpg, qsec, label=substr(maker, 1, 2)), color="white", fontface="bold") +
      labs(title = "Motor Trend Car Road Tests",
           subtitle = "Top 20 rows are extracted for demonstration", 
           caption = "Source: 1974 Motor Trend US magazine") + 
           scale_x_continuous(breaks=seq(10,34, 4))
gg

suppressPackageStartupMessages(library(dplyr))
gg <- ggplot(mtcars[1:20, ] %>% tibble::rownames_to_column() %>% 
             mutate(car_name = rowname, maker = gsub(" .*", "", car_name) ) ) + 
      #geom_label(aes(mpg, qsec, label = substr(car_name, 1, 13), color=maker),
      geom_point(aes(mpg, qsec, color=maker), size=8) +
      geom_text(aes(mpg, qsec, label=substr(maker, 1, 2)), color="white", fontface="bold") +
      labs(title = "Motor Trend Car Road Tests",
           subtitle = "Top 20 rows are extracted for demonstration", 
           caption = "Source: 1974 Motor Trend US magazine") + 
           scale_x_continuous(breaks=seq(10,34, 4))
gg

• When we consider making a presentation, this plot has several issues:
  • Axis titles and numeric values are too small to see through remote screen sharing (i.e. low-resolution)
  • Y-axis title is rotated and does not jump out at you unless carefully looking at it
  • Title and subtitle fonts are not aesthetically appealing
  • Does not necessarily conform to company-specific styles (e.g. white background)

For resolving these issues, you would add the following theme() configurations:

# If using ggplot2::theme():
gg + theme(
                   text = element_text(face="bold", size=24, family="Times New Roman"),
       panel.background = element_rect(fill="white"),
      legend.box.margin = margin(0.2,0.2,0.2,0.2,"cm"),
  legend.box.background = element_rect(colour="black"),
             legend.key = element_rect(fill="white"),
        legend.position = "bottom",
            legend.text = element_text(size=rel(0.8)),
           legend.title = element_text(family="Consolas", colour="royalblue"),
             axis.title = element_text(family="Consolas", colour="royalblue"),
           axis.title.y = element_text(angle=0, vjust=0.5),
              axis.text = element_text(size=rel(1.1)),
              axis.line = element_line(arrow=arrow(type="open",angle=20), size=2),
             axis.ticks = element_line(size=1),
      axis.ticks.length = grid::unit(0.5,"cm"),
          plot.subtitle = element_text(face="plain", hjust=1),
            plot.margin = margin(0.3,0.3,0.3,0.1,"inch")
  )

ggconf enables modifying these parameters with concice notations.

With ggconf

gg + 
  theme2(
       txt(f="bold", sz=24, family="Times New Roman"),    # make all text thicker/larger 
       pnl.bg(fill="white"),
       lgd.box.bg(c="black"),
       lgd.box.margin(.2, .2, .2, .2, "cm"),
       lgd.key(fill="white"),
       lgd.pos("bottom"),
       lgd.txt(z=rel(.8)),
       lgd.title(family="Consolas", c="royalblue"),       # equally-spaced font
       axs.title(family="Consolas", c="royalblue"),       # colorize axis titles
       axs.title.y(angle=0, vjust=.5),                    # rotate and centerize y axis label
       axs.txt(z=rel(1.1)),
       axs.line(arrow=arrow(type="open", angle=20), z=2), # 
       axs.tick(sz=1),                                    # tick or ticks? It doesn't matter
       axs.tick.len(.5, "cm"),
       plt.subtitle(f="plain", hjust=1),
       plt.margin(.3, .3, .3, .1, "inch")                 # adjust margins
  )

Other Works

ggconf draws inspiration from some other higher level programming languages including Bash, CoffeeScript, Lisp, and Ruby.

Current Implementation Status

ggconf is first released on August 24, 2017.

• DONE:
  • version 0.1 : lightweight port from ggbash

caprice-j/ggconf documentation built on May 23, 2019, 11:33 p.m.