In DnPeter/rpivotTable: Se les quiere

knitr::opts_chunk$set(echo = TRUE)

rpivotTable: A pivot table for R

The rpivotTable package is an R htmlwidget built around the pivottable library.

PivotTable.js is a Javascript Pivot Table library with drag'n'drop functionality built on top of jQuery/jQueryUI and written in CoffeeScript by Nicolas Kruchten at Datacratic. It is available under a MIT license

Many thanks to everyone that contributes bugs and prs, and of course thanks to Nicolas Krutchen for pivottable.

Installation

The rpivotTable package depends on the development version of the htmlwidgets package so you need to install both packages. You can do this using the devtools package as follows:

# devtools::install_github(c("ramnathv/htmlwidgets", "smartinsightsfromdata/rpivotTable"))

Usage

Just plug in your data.frame, table or data.table to rpivotTable().

It is as simple as this:

library(rpivotTable)

data(mtcars)
rpivotTable(mtcars,rows="gear", cols=c("cyl","carb"),width="100%", height="400px")

The pivot table should appear in your RStudio Viewer or your browser of choice.

For additional technical information please refer to the examples and explanations here.

rpivotTable parameters decide how the pivot table will look like the firs time it is opened:

data can be a data.frame table or data.table. Nothing else is needed. If only the data is selected the pivot table opens with nothing on rows and columns (but you can at any time drag and drop any variable in rows or columns at your leasure)
rows and cols allow the user to create a report, i.e. to indicate which element will be on rows and columns.
aggregatorName indicates the type of aggregation. Options here are numerous: Count, Count Unique Values, List Unique Values, Sum, Integer Sum, Average, Sum over Sum, 80% Upper Bound, 80% Lower Bound, Sum as Fraction of Total, Sum as Fraction of Rows, Sum as Fraction of Columns, Count as Fraction of Total, Count as Fraction of Rows, Count as Fraction of Columns
vals specifies the variable to use with aggregatorName (if any).
renderers dictates the type of graphic rendering used for display, like Table, Treemap etc.
sorters allow to implement a javascript function to specify the ad hoc sorting of certain values. See vignette for an example. It is especially useful with time divisions like days of the week or months of the year (where the alphabetical order does not work)

For example, to display a table with frequency of colour combinations of eyes and hair, you can specify:

library(rpivotTable)
data(HairEyeColor)
rpivotTable(data = HairEyeColor, rows = "Hair",cols="Eye", vals = "Freq", aggregatorName = "Sum", rendererName = "Table", width="100%", height="400px")

This will display the resulting table. Switching the aggregatorName to Sum as Fraction of Rows will give the row percentages (and the column totals will give the percentages over the gran total).

To display the Hair values in reverse order:

library(rpivotTable)
data(HairEyeColor)
rpivotTable(data = HairEyeColor, rows = "Hair",cols="Eye", vals = "Freq", aggregatorName = "Sum", rendererName = "Table", sorters = "
function(attr) { 
var sortAs = $.pivotUtilities.sortAs;
if (attr == \"Hair\") { return sortAs([\"Red\", \"Brown\", \"Blond\", \"Black\"]); }
}", width="100%", height="400px")

This function could be useful for example to sort time divisions like months of the year or days of the week in their proper, non alphabetical order (thanks to palatinuse for its implementation).

Or if you want to include it as part of your dplyr / magrittr pipeline, you can do that also.

suppressMessages(library(dplyr))
iris %>%
tbl_df %>%
filter( Sepal.Width > 3 & Sepal.Length > 5 ) %>%
rpivotTable(rows="Sepal.Width",  rendererName="Treemap")