Easily Subset Elements of Objects

Alternatives to purrr::keep(), purrr::discard(), and purrr::compact(), but without having to load the purrr namespace. The syntax and behavior is generally the same, with some exceptions (see Details). For example, compact_it() is similar to purrr::compact(), however only supports the default behavior where non-empty elements are retained.

Usage

keep_it(x, lgl, ...)

discard_it(x, lgl, ...)

compact_it(x)

Arguments

x

A list, data frame, or vector.

lgl

logical(n). A vector or a function that returns a logical vector when applied to the elements of x.

...

Named arguments passed to lgl if it is a function.

Details

These functions are not a simple drop-in replacement, as they do not support quasi-quotation or formula syntax, but should be a sufficient replacement in most cases.

Functions

• keep_it(): keeps elements corresponding to lgl.

• discard_it(): the inverse of keep_it().

• compact_it(): subsets elements that have non-zero length.

purrr analogues

helpr	purrr
keep_it()	purrr::keep()
discard_it()	purrr::discard()
compact_it()	purrr::compact()

Examples

# pass a logical vector
lst <- list(A = 1, B = 2, C = 3)
keep_it(lst, c(TRUE, FALSE, TRUE))
#> $A
#> [1] 1
#> 
#> $C
#> [1] 3
#> 

# logical vector on-the-fly
vec <- unlist(lst)
keep_it(vec, vec != 2)
#> A C 
#> 1 3 

# subset itself
keep_it(c(a = TRUE, b = FALSE, c = TRUE), identity)
#>    a    c 
#> TRUE TRUE 

# pass a simple function
lst <- replicate(10, sample(10, 5), simplify = FALSE)
keep_it(lst, function(x) mean(x) > 6)
#> [[1]]
#> [1]  5  3  7  9 10
#> 
#> [[2]]
#> [1]  9 10  5  8  4
#> 
#> [[3]]
#> [1]  9 10  3  2  8
#> 
#> [[4]]
#> [1]  8  5  7  1 10
#> 

# will work on data frames
df <- data.frame(a = 5, b = 2, c = 10)
keep_it(df, function(x) x >= 5)
#>   a  c
#> 1 5 10

df <- data.frame(a = "A", b = "B", c = 10, d = 20)
keep_it(df, is.numeric)
#>    c  d
#> 1 10 20

# compact_it() selects elements with non-zero length
lst <- list(A = 5, B = character(0), C = 6, D = NULL, E = NA, F = list())
compact_it(lst)
#> $A
#> [1] 5
#> 
#> $C
#> [1] 6
#> 
#> $E
#> [1] NA
#> 
# discard_it() works nicely with `be_safe()`
.f <- be_safe(log10)
res <- .f("5")
discard_it(res, is.null)
#> $error
#> [1] "non-numeric argument to mathematical function"
#>