Reputation: 9018
Whats the best way to detect null values in a vector?
If I have the vector below and want to know that the 4th position is null how would I do that?
vx <- c(1, 2, 3, NULL, 5)
is.null()
returns only FALSE
:
is.null(vx)
# [1] FALSE
and I'd like to get:
FALSE FALSE FALSE TRUE FALSE
Upvotes: 17
Views: 23710
Reputation: 13463
Its possible that someone (like me) may come looking for is.null()
, really needing is.na()
. If so, remind yourself (as I did) that R has both a NULL
and NA
object. NULL
signifies that there is no value whereas NA
signifies that the value is unknown. If you are looking for missing values in a vector of values, what you're probably after is is.na()
.
Upvotes: 10
Reputation: 28461
As mentioned in the comments, NULL
will not appear in length(vx)
. It is a special object in R for undefined values. From CRAN documentation:
NULL represents the null object in R: it is a reserved word. NULL is often returned by expressions and functions whose value is undefined.
But your question can still have learning opportunities with respect to lists. It will show up there. As in:
vlist <- list(1, 2, 3, NULL, 5)
Trying to identify the NULL
value in a very large dataset can be tricky for novice R users. There are different techniques, but here is one that worked for me when I needed it.
!unlist(lapply(vlist, is.numeric))
[1] FALSE FALSE FALSE TRUE FALSE
#or as user Kara Woo added. Much better than my convoluted code
sapply(vlist, is.null)
[1] FALSE FALSE FALSE TRUE FALSE
#suggestion by @Roland
vapply(vlist, is.null, TRUE)
[1] FALSE FALSE FALSE TRUE FALSE
If there are characters, then substitute in is.character
or whatever class applies.
Upvotes: 22