R 在 data.table 中每组返回多行

发布于05月18日

在data.table的分组命令中，每个组可以返回多行吗？在dplyr中，这是通过reframe完成的:

y <- c("a", "b", "d", "f")
df <- tibble(
  g = c(1, 1, 1, 2, 2, 2, 2),
  x = c("e", "a", "b", "e", "f", "c", "a")
)

library(dplyr)
df %>%
  reframe(x = setdiff(x, y), .by = g)
# g x    
# 1 e    
# 2 e    
# 2 c

在data.table中，这将返回错误:

library(data.table)
dt <- setDT(df)
dt[, x := setdiff(x, y), g]

Error in [.data.table(df, , :=(x, intersect(x, y)), g) :
Supplied 2 items to be assigned to group 1 of size 3 in column 'x'. The RHS length must either be 1 (single values are ok) or match the LHS length exactly. If you wish to 'recycle' the RHS please use rep() explicitly to make this intent clear to readers of your code.

不管怎么说，要拿到相当于reframe分的data.table分？

推荐答案

换成.(...)，用=代替:=(因为它在.(..)以内).

as.data.table(df)[, .(x = setdiff(x, y)), by = g]
#        g      x
#    <num> <char>
# 1:     1      e
# 2:     2      e
# 3:     2      c

请注意，在幕后，.(.)实际上只是list(.)，所以我们也可以使用任何返回类似list的对象，包括:

as.data.table(df)[, list(x = setdiff(x, y)), by = g]
as.data.table(df)[, data.table(x = setdiff(x, y)), by = g]
as.data.table(df)[, data.frame(x = setdiff(x, y)), by = g]