How to create a data frame from text submitted in a textarea with FastRWeb

March 8, 2012

(This article was first published on Rronan » R, and kindly contributed to R-bloggers)

In this article, I show you how to create a data.frame from a text submitted in a textarea field with FastRWeb.

Requirements

FastRWeb installed
Knowledge of webforms
?read.table
Experience in HTML5

Submit

This example needs two scripts. The first one contains the webform. I wrote a FastRWeb script in order to work in /var/FastRWeb/web.R/. This form could have been a static HTML page. Paste the code below in /var/FastRWeb/web.R/tablesubmit.R.

run <- function() {
  out("<!doctype html><html lang='en'><head><title>Submit a Table</title>",
      "<meta charset='UTF-8'/></head><body><header><h1>Submit a Table</h1>",
      "<p>This is an example showing how to read a text posted in a ",
      "<code>textarea</code> and create a data frame from it.</p></header>",
      sep = "", eol = "")
  out("<form method='post' action='textareatodf' ",
      "enctype='multipart/form-data' name='form'>", sep = "", eol = "")
  out("<p>Paste the table content in the text area below.</p>")
  out("<textarea name='MyTable' rows=15 cols=80></textarea>")
  out("<br/><input type='submit' /></form>")
  out("</body></html>")
  done()
}

Output

To retrieve the the submitted data, read them and create your data.frame, you’ll respectively use the functions parse.multipart(), textConnection() and read.table(). These last two are installed with R base. parse.multipart() is available if you have installed la version 1.1 de FastRWeb.

Paste the text below in /var/FastRWeb/web.R/textareatodf.R.

run <- function(MyTable) {
  # HTML
  out("<!doctype html><html lang='en'><head><title>Data Frame</title>",
      "<meta charset='UTF-8'/></head><body><header><h1><code>textarea</code> ",
      "to <code>data.frame</code></h1></header>", sep = "", eol = "")

  # Retrieve the data in a 'multipart/form-data' form
  if (grep("^multipart", request$c.type)) {
    pars <- parse.multipart()
  } else {
  }

  # Create the data frame
  try(MyDF <- read.table(file = textConnection(pars$MyTable), header = FALSE,
                         sep = "\t"))

  # HTML output
  if (!exists("MyDF")) {
    out("<h2>Something went wrong…</h2>",
        "<p>Make sure your table has the same number of elements for each ",
        "row.</p>", sep = "", eol = "")
  } else {
    out("<h2>Your data frame</h2><p>This is your data frame.</p>")
    oprint(MyDF)
    oprint(sapply(MyDF, class))
  }

  out("</body></html>")
  done()
}

To run this script, go here : http://localhost/cgi-bin/R/tablesubmit

Testing

Copy and paste the data from à spreadsheet (either LibreOffice Calc or Excel). The non-ASCII characters are handled correctly if the users did not change their browser encoding. You may want to use the options read.table() such as sep, col.names or colClasses.