Dplyr: Suggestion: recode to accept named list

[gowerc]

It would be really convenient if recode could accept a named list as an argument instead of having to specify each of the input parameters.

Currently to use the function you have to do something like

x <- c( "a" , "b" , "c")
recode( x , a = "1" , b = "2" , c = "3") 

I propose that the following be valid

MAP <- list(
    a = "1" ,
    b = "2" ,
    c = "3"
)

recode( x , MAP )

This allows for standardized maps to be setup and used across a project which is useful when you are dealing with short variable names or coded parameter terms which need to be transformed into more wordy labels later on.

[hadley]

Part of #2477

[ahcyip]

@hadley Has this been implemented?
I've been trying to use named lists in dplyr::recode() but it doesn't seem to work.
I suppose I'll stick with the tried and true left_join lookup table for now.

x <- c("a", "b", "c")

MAP <- list(a = "1", b = "2", c = "3")

dplyr::recode(x, MAP)

#> Error: Argument 2 must be named, not unnamed

[ahcyip]

Here is my left_join solution

  x <- c("a", "b", "c")
  MAP <- list(a = "1", b = "2", c = "3")
  dplyr::recode(x, MAP)
#> Error: Argument 2 must be named, not unnamed

  x_lj <- tibble::tibble(x = x)
  MAP_lj <- dplyr::bind_cols(x = names(MAP), y = as.character(MAP))
  dplyr::left_join(x_lj, MAP_lj, by = "x")
#> # A tibble: 3 x 2
#>       x     y
#>   <chr> <chr>
#> 1     a     1
#> 2     b     2
#> 3     c     3

[ahcyip]

@CG1122 I figured it out via http://dplyr.tidyverse.org/articles/programming.html:

x <- c("a", "b", "c")
MAP <- list(a = "1", b = "2", c = "3")
dplyr::recode(x, !!!MAP)
#> [1] "1" "2" "3"

Thanks @hadley @lionel- and tidyverse !

[billdenney]

I was about to suggest a recode_ function that does what @ahcyip suggested. Can that example or one largely similar be added to the documentation, please?