pivot_wider 具有复杂名称 R 的数据框

Question

所以我有一个如下所示的数据框：


datInput <- tibble(id = 1:2,
             c.0.opt = c("a,b", "c,d"),
             c.0.optI = c("1,2", "3,4"),
             c.0.sel = c("a", "c"),
             c.1.opt = c("e,f", "g,h"),
             c.1.optI = c("5,6", "7,8"),
             c.1.sel = c("e", "g"))

datInput
#     id c.0.opt c.0.optI c.0.sel c.1.opt c.1.optI c.1.sel
 
#1     1 a,b     1,2      a       e,f     5,6      e      
#2     2 c,d     3,4      c       g,h     7,8      g

我需要它看起来像这样：


datOutput <- tibble(id = c(1,1,2,2),
                   c_opt = c("a,b", "e,f", "c,d", "g,h"),
                   c_optI = c("1,2", "5,6", "3,4", "7,8"),
                   c_sel = c("a", "e", "c", "g"))

#     id c_opt c_optI c_sel

#1     1 a,b   1,2    a    
#2     1 e,f   5,6    e    
#3     2 c,d   3,4    c    
#4     2 g,h   7,8    g

我通常使用 dplyr::pivot_longer 来完成这类任务，但我不知道如何处理那些复杂的列名，行标识符在中间。有办法吗？

谢谢

Answer 1

datInput %>% 
  gather(colname, val,-1 ) %>% 
  mutate(colname = gsub("\.\d\.","_",colname)) %>% 
  pivot_wider(id_cols = id, names_from = colname, values_from = val, values_fn = list) %>% 
  unnest(cols = c(colnames(.)))

# A tibble: 4 x 4
     id c_opt c_optI c_sel
  <int> <chr> <chr>  <chr>
1     1 a,b   1,2    a    
2     1 e,f   5,6    e    
3     2 c,d   3,4    c    
4     2 g,h   7,8    g

Answer 2

我们也可以使用 pivot_longer 和 names_sep 作为正则表达式查找来匹配列名中数字

之后的 .

library(dplyr)
library(tidyr)
library(stringr)
pivot_longer(datInput, cols = -id, names_to = c("grp", ".value"), 
         names_sep = "(?<=\d)\.") %>%
    select(-grp) %>%
    rename_with(~ str_c('c_', .), -id)
 # A tibble: 4 x 4
 #   id c_opt c_optI c_sel
 #  <int> <chr> <chr>  <chr>
#1     1 a,b   1,2    a    
#2     1 e,f   5,6    e    
#3     2 c,d   3,4    c    
#4     2 g,h   7,8    g

Answer 3

我用 zimia 的评论修改了 Akrun 的回答，如下所示：

datOutput <- datInput %>% 
  pivot_longer(-id, names_to = "colname", values_to = "val") %>%
  mutate(colname = gsub("\.\d\.","_",colname)) %>% 
  pivot_wider(id_cols = id, names_from = colname, values_from = val, values_fn = list) %>% 
  unnest(cols = c(colnames(.)))

效果很好。谢谢你们。

pivot_wider 具有复杂名称 R 的数据框

pivot_wider a dataframe with complex names R

r

tidy

dataframe

dplyr