Extract separate columns from data frame

Question

I am trying to extract the first names of the titles of the columns such as pack_10, pack_18 and pack_20 and group all of them with sum. Below you can see my data

df<-data.frame(  
              packs_10_value5=c(100,0,0,0,0),
              packs_18_value9=c(200,0,0,0,0),
              packs_20_value13=c(300,0,0,0,0),
              packs_10_value15=c(100,0,0,0,0),
              packs_18_value17=c(200,0,0,0,0),
              packs_20_value18=c(300,0,0,0,0)
              )

df

So can anybody help me with how to solve this?

Answer 1

You can split the columns and apply rowSums by group:

library(purrr)
split.default(df, f = gsub("_value.*", "", names(df))) %>% 
  map_dfc(rowSums)

# A tibble: 5 × 3
  packs_10 packs_18 packs_20
     <dbl>    <dbl>    <dbl>
1      200      400      600
2        0        0        0
3        0        0        0
4        0        0        0
5        0        0        0