slice(df,n) for n > nrow(df) inconsistency when using data.tables

[franknarf1]

In this SO post, I noticed this difference:

df<-data.frame(Initials=c("a","a","b"),data=c(2,3,4))
my_rows = 1:2

df %>% group_by(Initials) %>% slice(my_rows)
#   Initials  data
#     (fctr) (dbl)
#1        a     2
#2        a     3
#3        b     4

library(data.table)
setDT(df) %>% group_by(Initials) %>% slice(my_rows)
#   Initials  data
#     (fctr) (dbl)
#1        a     2
#2        a     3
#3        b     4
#4        b    NA

Personally, I'd prefer to see the latter behavior and think it's more in keeping with standard R behavior, e.g., in (3:1)[c(5,6)]. However, based on #1269 , I think it clashes with your design goals.

In any case, the two cases should give the same result if dplyr is to play nice with data.table objects.