Using R in web development


Statistic Tool: R

I have been want to use R in my work for a long time. But seem it is not so easy to combine it into the normal work flow very easily.

Now we have a web development using R and Bootstrap: Shiny

Shiny seems would be very useful for rapid development for some web project.

I can pass all the statistic job to R, which is more suitable for it, then I can develop other business logic in javascript. And maybe I can learn some statistic knowledges by using R. That is great!

Shiny setup

Shiny is a package in R. So to install shiny:


install.packages("shiny")

Then in a directory, prepare 2 R files:

  1. ui.R
  2. server.R

Bootstrap2.3.2 components and other libs

According to the source, shiny integrate following javascript libs:


▾ inst/
  ▾ www/
    ▾ shared/
      ▸ bootstrap/
      ▸ datatables/
      ▸ datepicker/
      ▸ font-awesome/
      ▸ highlight/
      ▸ jqueryui/
      ▸ selectize/
      ▸ showdown/
      ▸ slider/
        jquery.js

Shiny use Bootstrap2.3.2. And it is very easy to use the components available.

In ui.R:


shinyUI(pageWithSidebar(
  headerPanel("Minimal Example"),
  sidebarPanel(
               textInput(inputId = 'comment',
                         label = "Say",
                         value = "")),
  mainPanel(
            h3("This is you saying it"),
            textOutput('textDisplay')
            )

))

Notice that we use headerPanel, sidebarPanel and mainPanel, they are defined in R/bootstrap.R . And there are more UI componts defined in that file.

In server.R, we will update the output:


library(shiny)
shinyServer(function(input, output) {
  output$textDisplay <-renderText({
    paste0("You said ", input$comment, " there are ", nchar(input$comment), " characters in this.")
  })
})

It does 2 things:

  1. gives the output name: testDisplay
  2. it tells shiny that the content contained whthin is reactive
  3. paste0 links strings without spaces

run gist


> runGist(1233456)

show all widgets


> runGist(6571951)

more about reactivity

To understand the reactive programming used in Shiny, I need to read the Shiny tutorial page

There are 3 kinds of objects in reactive programming:

  1. Reactive source
  2. Reactive conductor
  3. Reactive endpoint

In the simple case, we only need Reactive source and the Reactive endpoint. But in the case that we don’t want to calculate the same middle result several times, we can use Rective conductor. The following is an example for fib():


fib <- function(n) ifelse(n<3, 1, fib(n-1)+fib(n-2))

shinyServer(function(input, output) {
  currentFib         <- reactive({ fib(as.numeric(input$n)) })

  output$nthValue    <- renderText({ currentFib() })
  output$nthValueInv <- renderText({ 1 / currentFib() })
})

More notes for reactive object

We can’t access the reactive value or expressions from outside a reactive context.

The following code can’t work:


shinyServer(function(input, output) {
  # Will give error
  currentFib      <- fib(as.numeric(input$n))
  output$nthValue <- renderText({ currentFib })
})

But we can define a function, then call it in the reactive world:


shinyServer(function(input, output) {
  # OK, as long as this is called from the reactive world:
  currentFib <- function() {
    fib(as.numeric(input$n))
  }

  output$nthValue <- renderText({ currentFib() })
})

Reactive object implementation

  1. Reactive values are an implementation of Reactive sources; that is, they are an implementation of that role.

  2. Reactive expressions are an implementation of Reactive conductors. They can access reactive values or other reactive expressions, and they return a value.

    A reactive expressions can be useful for caching the results of any procedure that happens in response to user input, including:

    1. accessing a database
    2. reading data from a file
    3. downloading data over the network
    4. performing an expensive computation

  3. Observers are an implementation of Reactive endpoints. They can access reactive sources and reactive expressions, and they don’t return a value; they are used for their side effects.

    manipulate CSV

import CSV into data frame


data <- read.table("simulation1.csv", header=TRUE, sep=",")

export sample data into csv


> write.csv(women, file = "MyData.csv")

query

league_table$team[league_table$home_wins > 8]

merge


> league <- merge(league_table, points, by='team')

sorting


> with(league, league[order(-pts),])

insert more rows


rbind(with(league, league[order(-pts),]), another_table)

format date string


alldates <- format(as.Date(sent_data$date), '%Y-%m')

user login

We should be able to use nginx or apach to do the http basic authentication, then foward the request to local Shiny server

sample data of R

load sample data


> data(women)
> women
   height weight
1      58    115
2      59    117
3      60    120
comments powered by Disqus