Home Convert multiple rows into one column for a specific start of row name starting

Questions

Convert multiple rows into one column for a specific start of row name starting

June 23, 2023

From a dataframe like this:

df <- data.frame(row_name = c("sha1", "sha2", "edt1", "edt2", "perform1", "perform2"),
                 value = c(10, 20, 30, 40, 50, 60))

How is it possible to convert rows to columns to receive this result?

df <- data.frame(sha = c(10, 20), edt = c(30, 40), perform = c(50, 60))

>Solution :

You can do it first by separating the number from the words that you want to be the variable names. Then you can pivot_wider() on the names and then remove the variable that captured the number.

library(dplyr)
library(tidyr)
df <- data.frame(row_name = c("sha1", "sha2", "edt1", "edt2", "perform1", "perform2"),
                 value = c(10, 20, 30, 40, 50, 60))
df %>% 
  separate_wider_regex(row_name, c(name = "[^\\d]*", obs = "\\d")) %>%
  pivot_wider(names_from = "name", values_from = "value") %>% 
  select(-obs)
#> # A tibble: 2 × 3
#>     sha   edt perform
#>   <dbl> <dbl>   <dbl>
#> 1    10    30      50
#> 2    20    40      60

^{Created on 2023-06-23 with reprex v2.0.2}