In devOpifex/g2r: Interactive Grammar of Graphics

library(g2r)

Aspects (asp()) are analogous to ggplot2::aes but may work in somewhat different ways in places.

library(g2r)

g2(trees, asp(Girth, Volume, size = Height)) %>% 
  fig_point(asp(shape = "circle", label = Height))

Aspects define the visualisation, these aspects can be further manipulated with the gauge_* functions.

cb <- "(data, mappingData, index) => {
  return data.Height + 'ft'
}"

g2(trees, asp(Girth, Volume, size = Height)) %>% 
  fig_point(asp(shape = "circle", label = Height)) %>% 
  gauge_label(
    content = htmlwidgets::JS(cb), 
    style = list(fill = "darkgray")
  )

Callback functions have to be wrapped with htmlwidgets::JS

There are some differences with ggplot2 though, e.g.: the use of JavaScript functions as shown above. Another such difference is that one use the same aspect multiple times (not all aspects).

df <- data.frame(
  x = c(letters, letters),
  y = runif(52),
  grp = rep(c("A", "B"), each = 2)
)

The gauge is only used when a corresponding aspect has been defined, to configure a global feature of the chart there are other functions, e.g.: tooltip.

Warning

The graph below is __erroneous.__

g2(df, asp(x, y, color = grp)) %>% 
  fig_interval() %>% 
  tooltip(
    shared = TRUE,
    showMarkers = FALSE
  )

Some aspects (few) are defined outside of the asp function, e.g.: to adjust bars on a chart or points on a scatter plot.

g2(df, asp(x, y, color = grp)) %>% 
  fig_interval(adjust("dodge")) %>% 
  tooltip(
    shared = TRUE,
    showMarkers = FALSE
  )

g2(cars, asp(speed, dist)) %>% 
  fig_point(asp(shape = "circle")) %>% 
  fig_point(adjust("jitter")) %>% 
  fig_point(adjust("jitter")) %>% 
  fig_point(adjust("jitter"))

Aspects also define the shape of points, lines, and areas.

df <- data.frame(
  x = 1:100,
  y = rnorm(100, mean = rep(c(10, 14), each = 50))
)

g2(df, asp(x, y)) %>% 
  fig_smooth(asp(shape = "smooth"), method = "gaussian") %>% 
  fig_point(asp(shape = "square"))