data.table 子集条件 on.SD 列？

Question

I want to create the same result as the below minimum reproducible example, but you can see the subsetting condition on the last code is basically remove rows where.SD columns should not be all zero .

library(dplyr)

library(data.table)

set.seed(1)

t1 <- sample(-1:1, 999999, replace = T) %>% matrix(ncol = 9) %>% as.data.table %>% {cbind.data.frame(id = 1:nrow(.), .)}

t1[V1 != 0|V2 != 0|V3 != 0|V4 != 0|V5 != 0|V6 != 0|V7 != 0|V8 != 0|V9 != 0, lapply(.SD, sum), by = id]

I have tried

t1[sum(abs(.SD)) != 0, lapply(.SD, sum), by = id]

Answer 1

We can create the i with Reduce

i1 <- t1[, .I[Reduce(`|`, lapply(.SD, `!=`, 0))], .SDcols = V1: V9]
out2 <- t1[i1, lapply(.SD, sum), by = id]

-checking with OP's output

out1 <- t1[V1 != 0|V2 != 0|V3 != 0|V4 != 0|V5 != 0|V6 != 0|V7 != 0|V8 != 0|V9 != 0, lapply(.SD, sum), by = id]  
identical(out1, out2)
#[1] TRUE

Answer 2

Using rowSums() :

t1[t1[, rowSums(.SD != 0) > 0, .SD = patterns("V\\d")]
   ][, lapply(.SD, sum), by = id]