data.table: group-by, sum, name new column, and slice columns in one step

问题

This seems like it should be easy, but I've never been able to figure out how to do it. Using data.table I want to sum a column, C, by another column A, and just keep those two columns. At the same time, I want to be able to name the new column. My attempts and desired output:

library(data.table)
dt <- data.table(A= c('a', 'b', 'b', 'c', 'c'), B=c('19', '20', '21', '22', '23'),
C=c(150,250,20,220,130))

# Desired Output - is there a way to do this in one step using data.table? #
new.data <- dt[, sum(C), by=A]
setnames(new.data,'V1', 'C.total')
new.data
   A C.total
1: a     150
2: b     270
3: c     350

# Attempt 1: Problem is that columns B and C kept, extra rows kept #
new.data <- dt[, 'C.total' := sum(C), by=A]
new.data
   A  B   C C.total
1: a 19 150     150
2: b 20 250     270
3: b 21  20     270
4: c 22 220     350
5: c 23 130     350

# Attempt 2: Problem is that new column not named #
new.data <- dt[, sum(C), by=A]
new.data
   A  V1
1: a 150
2: b 270
3: c 350

回答1:

Use list (or .):

> dt[, list(C.total = sum(C)), by=A]
   A C.total
1: a     150
2: b     270
3: c     350

来源：https://stackoverflow.com/questions/32124630/data-table-group-by-sum-name-new-column-and-slice-columns-in-one-step