Overwriting only NAs in column A with values from column B

Question

I have column Aand column B in a dataframe:

A = structure(c(NA, NA, NA, 1559401558, 1559413729, 1559417798), class = c("POSIXct", 
"POSIXt"), tzone = "")

B = structure(c(1559379600, 1559388600, 1559397600, 1559406600, 1559415600, 
1559424600), class = c("POSIXct", "POSIXt"), tzone = "UTC")
>

As you can see, only column A has missing Dates. I want now, that only the missing Dates in Aare replaced with related values (same indices) in B. I know this should work with indices, but I can't find a solution for this. Thanks for the help!

Also see https://stackoverflow.com/q/15629885/5325862 – camille Dec 23 '19 at 18:56 — camille, Dec 23 '19 at 18:56

score 5 · Accepted Answer · answered Dec 23 '19 at 16:56

5

Using Base R:

A[is.na(A)] <- B[is.na(A)]

answered Dec 23 '19 at 16:56

akash87

3,398
2
11
28

akrun · Answer 2 · 2019-12-23T18:23:31.117

3

We can use coalesce

library(dplyr) 
coalesce(A, B)

As @RuiBarradas mentioned in the comments, we can set the tz to NULL before doing the coalesce

library(lubridate)
coalesce(A, `tz<-`(B, ""))

edited Dec 23 '19 at 18:23

answered Dec 23 '19 at 16:57

akrun

674,427
24
381
486

1

Note that `A` doesn't have a timezone assigned, so the result may be a surprise. – Rui Barradas Dec 23 '19 at 17:23

score 1 · Answer 3 · answered Dec 23 '19 at 17:54

1

With tidyverse:

library(tidyverse)

tibble(A = A, B = B) %>%
 mutate(A = ifelse(is.na(A), B, A))

answered Dec 23 '19 at 17:54

Ben G

2,903
1
13
33

Overwriting only NAs in column A with values from column B

3 Answers3