r2r

r2r provides a flexible implementation of hash tables in R, allowing for:

• arbitrary R objects as keys and values,
• arbitrary key comparison and hash functions,
• customizable behaviour (throw or return a default value) on missing key exceptions.

Installation

You can install the released version of r2r from CRAN with:

install.packages("r2r")

and the development version from my R-universe repository, with:

install.packages("r2r", repos = "https://vgherard.r-universe.dev")

Usage

library(r2r)
m <- hashmap()

# Insert and query a single key-value pair
m[[ "user" ]] <- "vgherard"
m[[ "user" ]]
#> [1] "vgherard"

# Insert and query multiple key-value pairs
m[ c(1, 2, 3) ] <- c("one", "two", "three")
m[ c(1, 3) ]
#> [[1]]
#> [1] "one"
#> 
#> [[2]]
#> [1] "three"

# Keys and values can be arbitrary R objects
m[[ lm(mpg ~ wt, mtcars) ]] <- c(TRUE, FALSE, TRUE)
m[[ lm(mpg ~ wt, mtcars) ]]
#> [1]  TRUE FALSE  TRUE

Getting help

For further details, including an introductory vignette illustrating the features of r2r hash maps, you can consult the r2r website. If you encounter a bug, want to suggest a feature or need further help, you can open a GitHub issue.

Comparison with hash

CRAN package {hash} also offers an implementation of hash tables based on R environments. The two tables below offer a comparison between {r2r} and {hash} (for more details, see the benchmarks Vignette)

Features supported by {r2r} and {hash}.

Performances of {r2r} and {hash} for basic hash table operations.