This may be what prevent R from installing xgboost.

Define variables, such as colors (e.g. $primary: #337ab7) in Sass (styles.scss) then compile to css for web.

R library "bootstraplib" built on foundation of "sass".

Use "run_with_themer()" to get a live preview GUI for customizing bootstrap theme.

Also, use "shinyOptions(plot.autocolors=TRUE)" at top of app to get plot outputs that respect Dark Mode.

Comparison of dashboard tech stack as of 10/2020:

Dablander, F. (2020, June 11). Interactive exploration of COVID-19 exit strategies. R-Bloggers. https://www.r-bloggers.com/interactive-exploration-of-covid-19-exit-strategies/

The same basic structure for all Shiny apps:

1. ui object.
2. server function.
3. call to the shinyApp function.

---> examples <---

UI example of a Shiny app (check the code below)

Server example of a Shiny app (check the code below):

• random distribution is plotted as a histogram with the requested number of bins
• code that generates the plot is wrapped in a call to renderPlot

Using Python scripts inside R Shiny (in 6 steps):

1. In ui.R create textOutput: textOutput("textOutput") (after plotoutput()).
2. In server.R create handler: output$textOutput <- renderText({ }].
3. Create python_ref.py and insert this code:
4. Import reticulate library: library(reticulate).
5. source_python() function will make Python available in R:
6. Make sure you've these files in your directory:

• app.R
• python_ref.py and that you've imported the reticulate package to R Environment and sourced the script inside your R code.

Hit run.

Shiny > Dash:

• Dash isn't yet as stabilised
• Shiny has much more layout options, whereas in Dash you need to utilise HTML and CSS
• developing new components in Dash needs ReactJS knowledge (not so easy)

Typical tools used for working with Shiny