cache

The cache package provides a consistent interface for writing and restoring data. It is designed to object permanance easy, functional, extensible and consistent.

• easy: there are two functions that you need, cache and uncache
• functional:
  • functions work well with pipes.
  • use for building data marts for analyses
  • replaces ProjectTemplates cache functionality
  • encryption built in.
• extensible:
  • Add new save/load and read/write by writing your own cache_read_* and cache_write_* functions.
• consistent
  • standard arguments and meanings

It should be noted that the cache exists on a per-project basis.

Together with the -tk package. Those data sets can be saved and easily refreshed.

NB This packges is not intended for importing data. Packages such as readr, foreign or any of the database packages are well-suited for this purpose.

Motivation

A very common pattern when building models or performing analysis is to build ad-hoc data marts that support your analysis.

In many ways this performs a more generalized function of what ProjectTempate does but in a more general and pipe-friendly way.

or performing analyses you would like to create a local store of data that can improve model building and analysis. The cache package makes that easy

Installation

install_github( "decisionpatterns/cache" )

Usage

cache(object)    # save object to cache
cache("object")  # same (by name)
 
uncache(object)
uncache(object)
 
use_cache()    # Creates a cache directory like devtools::use_* functions

DESIGN

cache is designed to be modular in its ability to save and retrieve objects

What is a Cache?

A cache is just a directory for storing objects. The cache package handles the saving and restoring of those objects.