数据表-分组实验中j表达的有效评价

有没有一种方法可以使用/计算字符串表达式而不是 eval(parse(text = ...))
据我所知, eval(parse()) i 以及 j by 在使用中。我曾经 verbose=TRUE GForce 在使用表达式时不使用,所以我想问题是如何同时使用两者 G力

library(data.table)

N = 10**5
DT = data.table(x1 = sample(1:1000, size = N , replace = TRUE),
                x2 = sample(1:500, size = N , replace = TRUE),
                y1 = runif(N,1,10),
                y2 = runif(N,0,1))

system.time({
  DT_agg = DT[, .(sum_y1 = sum(y1)), by = c('x1','x2'), verbose=TRUE]
})
# Making each group and running j (GForce TRUE)
# user  system elapsed 
# 0.02    0.00    0.02 

expr = "sum_y1 = sum(y1)"
system.time({
  DT_agg = DT[, .(eval(parse(text = expr))), by = c('x1','x2'), verbose=TRUE]
})
# Making each group and running j (GForce FALSE)
# user  system elapsed 
# 27.72    0.00   28.11