Note that this page provides installation instructions for R. If you want to install loon for Tcl then follow the instructions here.

Important

• It is best to use loon in RStudio or within a terminal.
  • The Rgui app on OSX does not work well with loon.
  • The default Windows R GUI will not accept debug messages from Tcl, so for now you need to use RStudio or the terminal.
• Mac users need to install XQuartz
  • Do not close XQuartz while R is running! Otherwise you end up crashing the active R session (including RStudio).
• On Ubuntu one should install the libtk-img debian package (e.g. with sudo apt-get install libtk-img) in order to get a wide variety of export formats with the l_export function.

Further Notes

• If you experience difficulties with loon use the Issue tracker on github.
• If you have difficulties with exporting images with the l_export function then take screenshots:
  • On OSX press Command + Shift + 4 and then press space and select the window
  • On Linux install a program called Shutter
  • For Windows use Greenshot
• If you need to install loon manually on a Windows machine with an R version that is less than 3.4.0 then you need install Tcl version 8.6 and link R against it, see the instructions here.

Installation

Installing the loon package

The loon package is available on CRAN. To install the package start your R and run

install.packages('loon')

You can also install the latest development release directly from GitHub with the following R code

devtools::install_github("great-northern-diver/loon", subdir="R")

Packages used for Examples and Demos

The following packages are used in loon's examples and demos. Note that these packages are not needed to install loon, they are just nice to have to run all the examples and demos. You can skip this step and check if your loon installation was successful.

We split the code to install the suggested packages into four sections as not all the packages are easy to install.

First, these packages from CRAN should install without any issues

install.packages(c('loon.data','maps',
                   'sp','RColorBrewer',
                   'rworldmap', 'scales'))

The following packages on Bioconductor should also install without any issues

source("https://bioconductor.org/biocLite.R")
biocLite(c('graph', 'RDRToolbox', 'Rgraphviz'),
    suppressUpdates=TRUE, suppressAutoUpdate=TRUE)

The following packages on CRAN have dependencies that might need special care

• On Ubuntu install the following packages first (from bash)

  sudo apt install mesa-common-dev libglu1-mesa-dev freeglut3-dev\
      libssl-dev libcurl4-gnutls-dev libtk-img libxml2-dev

install.packages(c(
  'dplyr', 'rgl',
  'PairViz', 'scagnostics',
  'kernlab', 'testthat',
  'knitr', 'rmarkdown',
  'formatR'
))

Check if your Installation was Successful

In R enter

library(loon)

p <- with(iris, l_plot(x=Sepal.Length, y=Sepal.Width, color=Species))

If this creates a scatterplot and an inspector loon was correctly installed. Follow the steps in the UI section and the Learn section to learn how to use loon.

Fast Image Resizing for Image Point Glyphs

On Linux and OS X it is advisable to install the ImageScale Tcl extension for fast image resizing. loon will use the compiled C code for image resizing when available.

The TEA setup of ImageScale for Windows does currently not work. If you know how to change the makefile.vc in the win folder so that the ImageScale package also compiles under Windows then please .

More Important Notes

Note that only part of loon's functionality is documented on this webpage. However, most of loon's features are used in the R documentation manuals, R package demos and R package vignettes.

Manual

To get an index of all help entries run

help(package = "loon")

Package Demos

To get a listing of all of loon's demos enter

demo(package="loon")

Run a particular demo as follows

demo("l_timeseries")

To get the location of the source code of a particular demo (e.g. l_timeseries) use

system.file("demo", "l_timeseries.R", package = "loon")

Package Vignettes

To get a list of all package vignettes run

vignette(package = "loon")

To open the minority vignette run

vignette('minority', package = "loon")

Good to know

• There is presently no support for legends
• Most images showing loon's plots are anti-aliased on this webpage. You will probably not see anti-aliased plots on your screen because the canvas widget is only anti-aliased under OS X. Also, the Tcl version that is included in R under OS X does not have an anti-aliased canvas.
• Except for Linux with the libtk-img package installed (e.g. Img Tcl extension): currently only the ps file format works reliably for image exports (although the font mapping is not correct yet). I recommend for now to make screenshots if you need to put a loon plot into your report.