How to select the names of variables which are of class POSIXct in R?

Question

I have a data frame, df, which has columns of different class:

df <- structure(list(ID = 1:6, weight = c(74, 60, 49, 75, 75, 57), 
    date = structure(c(1488931200, 1501459200, 1493251200, 1494201600, 
    1494201600, 1493251200), class = c("POSIXct", "POSIXt"), tzone = "UTC"), 
    group = c("B", "B", "A", "A", "A", "A")), row.names = c(NA, 
6L), class = "data.frame")

df
  ID weight       date group
1  1     74 2017-03-08     B
2  2     60 2017-07-31     B
3  3     49 2017-04-27     A
4  4     75 2017-05-08     A
5  5     75 2017-05-08     A
6  6     57 2017-04-27     A

> sapply(df, class)
$ID
[1] "integer"

$weight
[1] "numeric"

$date
[1] "POSIXct" "POSIXt" 

$group
[1] "character"

I want to be able to select which columns have a certain class. For example:

names(which(sapply(df, class) == "numeric" | sapply(df, class) == "character"))
[1] "weight" "group" 

However, this doesn't work with class of the date column.

dput(class(df$date))
c("POSIXct", "POSIXt")

names(which(sapply(df, class) == c("POSIXct", "POSIXt")))
character(0)

Can someone advise the best way to give me the name of columns that have the POSIXct class?

Answer 1

Using your method with sapply you can do :

names(df)[sapply(df, function(x) all(class(x) %in% c("POSIXct", "POSIXt")))]
#[1] "date"

You can also use Filter :

names(Filter(function(x) all(class(x) %in% c("POSIXct", "POSIXt")), df))