In my previous blog on 30 Sep 2016, I had just discovered the plotly package and was blown away by how easy it was to use plotly to make ggplot2 plots reactive to the user’s mouse.

Well I’ve just stumbled across another package to achieve this called ggiraph, and I ggiraph like even more!!

In my opinion ggiraph integrates with ggolot2 in a nicer way than plotly to achieve basic reactivity to the user’s mouse.

Below is some code to make same fake data as I used in my last blog post, but this time I use ggiraph to turn the static ggplot2 plot from my last blog post into a plot with user reactivity.

The code below uses functions from the “ggplot2”, “stringr”, “dplyr”, “knitr” and “ggiraph” packages.

Use the same fake revenue data as my last blog post, but this time add a column for State Total Revenue ($) per year:

#Make up some fake data
df<-data_frame(state=rep(c("New South Wales", 
                 "Victoria", 
                 "Queensland",
                 "Western Australia",
                 "South Australia",
                 "Tasmania"), 36)) %>%
    group_by(state) %>%
    mutate(year=c(rep(2012, 9), rep(2013,9),rep(2014, 9),rep(2015, 9))) %>%
    group_by(state, year) %>%
    mutate(`store ID` = str_c("shop_#",as.character(seq_along(state)))) %>%
    group_by(state, year, `store ID`) %>%
    mutate(`Revenue ($)` =  ifelse(state=="New South Wales", sample(x=c(100000:900000), 1),
                            ifelse(state=="Victoria", sample(x=c(100000:700000), 1),
                            ifelse(state=="Queensland", sample(x=c(100000:500000), 1),
                            ifelse(state=="Western Australia",sample(x=c(10000:200000), 1),
                            ifelse(state=="South Australia",sample(x=c(10000:90000), 1),       
                            ifelse(state=="Tasmania", sample(x=c(10000:200000), 1), NA))))))) %>%
    group_by(state, year) %>%
    mutate(`State Total Revenue ($)` = sum(`Revenue ($)`))

Peak at the first 6 rows of the fake revenue data:

kable(head(df))

state	year	store ID	Revenue ($)	State Total Revenue ($)
New South Wales	2012	shop_#1	271022	3885088
Victoria	2012	shop_#1	207434	4209586
Queensland	2012	shop_#1	236501	3018283
Western Australia	2012	shop_#1	35100	1168335
South Australia	2012	shop_#1	15847	344081
Tasmania	2012	shop_#1	50989	1012868

Get the same ggplot from my last blog most, but this time, add user reactivity using ggiraph…

by replacing ‘geom_boxplot’ and ‘geom_point’ (from the plot in my last blog post) with ‘geom_boxplot_interactive’ and ‘geom_point_interactive’ from the ggiraph package, and then using ggiraph() to print the plot:

#Create the text you want displayed in the tooltop for geom_point_interactive
  df$tooltip_point <- paste0(
    "<b>", df$`store ID`, "</b>",
    "<br>(", df$state, ")",
    "<br>Revenue:",
    "<br>$", df$`Revenue ($)`,
    "</span></div>"
  )  

#Create the text you want displayed in the tooltop for geom_boxplot_interactive
  df$tooltip_boxplot <- paste0(
    "<b>", df$state, "</b>",
    "<br>State Total Revenue:",
    "<br>$", df$`State Total Revenue ($)`,
    "</span></div>"
  )    


#Make plot reactive to user's mouse with 
#'geom_boxplot_interactive' and 'geom_point_interactive'
point_interactive_plot <- ggplot(df, aes(state, `Revenue ($)`)) +
        geom_boxplot_interactive(aes(colour=state, tooltip=tooltip_boxplot, data_id=tooltip_boxplot)) +
        geom_point_interactive(aes(colour=state, tooltip=tooltip_point, data_id=tooltip_point)) +
        theme(axis.title.x =  element_blank(),
              axis.text.x  =  element_blank(), 
              axis.title.y = element_text(face="bold", size=12),
              axis.text.y  = element_text(angle=0, vjust=0.5, size=11),
              legend.title = element_text(size=12, face="bold"),
              legend.text = element_text(size = 12, face = "bold"),
              plot.title = element_text(face="bold", size=14)) + 
        ggtitle("Store Revenue per State from 2012 to 2015") +
        facet_wrap(~year) 


# htmlwidget call
ggiraph(code = {print(point_interactive_plot)}, zoom_max = 5,
        tooltip_offx = 20, tooltip_offy = -10, 
        hover_css = "fill:red;",
        tooltip_opacity = 0.7,
        pointsize = 12)

These plots using the ggiraph and ggplot2 packages are cool because you can hover mouse over data point and a lable appears with information about that data point or box. Notice I’ve made different information appear depending on whether the mouse is hovered over an individual data point or a state box-plot, using the tooltip_point and tooltip_boxplot variables I made. Also notice the “zoom_max = 5” in the ggiraph() function allows the user to mouse scroll to zoom in and zoom out on the plot. David Gohel and Mike Bostock have done a great job with the ggiraph package! imho :)

The ggiraph package has a bunch of other interactive geom functions too:

• geom_bar_interactive
• geom_point_interactive
• geom_polygon_interactive
• geom_map_interactive
• geom_path_interactive
• geom_rect_interactive
• geom_segment_interactive
• geom_text_interactive
• geom_boxplot_interactive

Go check it out!