Crunch.io provides a cloud-based data store and analytic engine. It has a web client for interactive data exploration and visualization. The crunch package for R allows analysts to interact with and manipulate Crunch datasets from within R. Importantly, this allows technical researchers to collaborate naturally with team members, managers, and clients who prefer a point-and-click interface: because all connect to the same dataset in the cloud, there is no need to email files back and forth continually to share results.
Both crunch and the Crunch web application communicate with the same application programming interface (API), served over secure HTTP. Within an R script or interactive session, the crunch package allows you to interact with your data in Crunch with expressive, idiomatic R. Functions in crunch handle the translation between R objects and API requests and responses, so you can typically interact with your datasets as if they were local
data.frames (with some additional metadata), not data on a remote server.
All work with data in Crunch requires users to be authenticated. Thus, the first step after loading crunch is to log in:
Crunch.io password for firstname.lastname@example.org: *enter your password here* ## Logged into crunch.io as email@example.com
This will log you in to the Crunch API. All of these parameters can be also be set as R
options in your .Rprofile so that you can simply
?login for details.
The Crunch data store is built around datasets, which contain variables. Unlike R
data.frames and atomic vectors, Crunch datasets and variables contain additional metadata. The following vignettes illustrate how you can create and work with datasets using crunch.