Create a named list with list object's variable names as the list names

Question

I often find myself creating a list from several variables, then renaming the list to use those variable names as the name of the list so I can access them with $ notation or use bind_rows with the .id argument:

a <- 1:10
b <- 11:20
mylist <- list(a, b)
names(mylist) <- c("a", "b")
mylist$a

[1]  1  2  3  4  5  6  7  8  9 10

I can't figure out how to skip this second step and tell R to just use the variable names as the names within the list:

list(a, b)$a

NULL

This seems like it should be doable because it's the default behavior for data.frame:

data.frame(a, b)$a

[1]  1  2  3  4  5  6  7  8  9 10

but using this method only works if the variables all have exactly the same length. It might seem like a simple thing to just use something like

mylist <- list(a=a, b=b)

but this feels redundant, becomes painful with long variable names, and I've often introduced bugs with typos that I don't notice until later.

Is there a way to construct a list in R from some variables that automatically names the list with the variable names?

akrun · Accepted Answer · 2023-03-07 19:14:45Z

7

We may use dplyr::lst

library(dplyr)
lst(a, b)$a
 [1]  1  2  3  4  5  6  7  8  9 10

Or in a chain

lst(a, b)  %>% 
 pluck("a")
 [1]  1  2  3  4  5  6  7  8  9 10

answered Mar 7, 2023 at 19:14

akrun

891k38 gold badges590 silver badges700 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Mikael Jagan · Accepted Answer · 2023-03-08 22:51:26Z

Please, just use base R's mget for that task ...

a <- 1:10
b <- 11:20
mget(c("a", "b"))

$a
 [1]  1  2  3  4  5  6  7  8  9 10

$b
 [1] 11 12 13 14 15 16 17 18 19 20

mget is faster than tibble::lst (also available as dplyr::lst):

microbenchmark::microbenchmark(mget(c("a", "b")), tibble::lst(a, b), times = 10000L)

Unit: nanoseconds
              expr   min    lq       mean median    uq     max neval
 mget(c("a", "b"))   492   615   726.4995    697   738   32390 10000
 tibble::lst(a, b) 36039 37474 42023.1345  37966 38786 1408309 10000

It is also better for programming. Suppose that you want a list containing each variable in your environment whose name begins with zzz. Then you could do mget(grep("^zzz", ls(), value = TRUE)). It is not so clear how you would accomplish that with tibble::lst.

tibble::lst is primarily useful when you want to evaluate nontrivial expressions (i.e., calls rather than symbols) with recursive access to the values of earlier arguments:

tibble::lst(x = a + b, y = x * exp(-b), z = y + 1, z - 1)

$x
 [1] 12 14 16 18 20 22 24 26 28 30

$y
 [1] 2.004204e-04 8.601897e-05 3.616527e-05 1.496752e-05 6.118046e-06
 [6] 2.475774e-06 9.935851e-07 3.959795e-07 1.568783e-07 6.183461e-08

$z
 [1] 1.000200 1.000086 1.000036 1.000015 1.000006 1.000002 1.000001 1.000000
 [9] 1.000000 1.000000

$`z - 1`
 [1] 2.004204e-04 8.601897e-05 3.616527e-05 1.496752e-05 6.118046e-06
 [6] 2.475774e-06 9.935851e-07 3.959795e-07 1.568783e-07 6.183461e-08

Collectives™ on Stack Overflow

Create a named list with list object's variable names as the list names

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related