In maurolepore/reuse: Reuse R Objects

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

reuse

The goal of reuse is to avoid re-running slow computations, particularly in pipes.

reuse::reuse() combines features of pins::pin_read() and pins::pin_write() with some differences:

• You can use it anywhere in a pipe.
• You don't have to specify the board explicitly.
• You can update the cache inside reuse() or globally with options().

Installation

You can install the development version of reuse like so:

# install.packages("devtools")
devtools::install_github("maurolepore/reuse")

Example

library(reuse)

• Is plays well with the pipe.
• It saves computations, reusing the cache if possible.

one <- 1 %>% reuse("one")
two <- 2 %>% reuse("one")
two

• You can update the cache for one call with the argument update.

two <- 2 %>% reuse("one", update = TRUE)
two

• You can update the cache for one session with the option reuse.update.

options(reuse.update = TRUE)

three <- 3 %>% reuse("one")
three

• The default cache folder uses the appropriate directory for Linux, Mac, and Windows, via the rappdirs package.

cache_dir <- rappdirs::user_cache_dir("reuse")
dir.exists(file.path(cache_dir, "one"))

• It's based on the pins package, which you can use to manage your cache folder a.k.a. "board".

library(pins)

board <- board_folder(cache_dir)
board %>% pin_meta("one")

board %>% pin_delete("one")

• You can use a custom cache folder with the argument board.

custom <- pins::board_folder(tempdir())
1 %>% reuse("abc", board = custom)

pins::pin_read(board = custom, "abc")

• You can use a custom cache folder with the option reuse.board.

options(reuse.board = custom)
1 %>% reuse("def", board = custom)

pins::pin_exists(board = custom, "def")

maurolepore/reuse documentation built on May 1, 2022, 7:34 a.m.