将长格式拆分为行，但将重复项合并为以“;”分隔的字符串？

Question

Suppose I have a dataframe that I want to split into rows.

temp = data.frame ( group=c('a','b','c'), fruits = c('apple', 'orange', 'none'), days=c('mon','tues','wed') )
reshape2::dcast(temp , days ~ group, value.var=c ( "fruits") )
  days     a      b    c
1  mon apple   <NA> <NA>
2 tues  <NA> orange <NA>
3  wed  <NA>   <NA> none

This works fine, however when I add a duplicated row as such.

temp = rbind ( temp, c('a','orange','mon')  )

the casting would failed and only show the total. What I really want is something like this.

 days     a      b    c
1  mon apple;orange   <NA> <NA>
2 tues  <NA> orange <NA>
3  wed  <NA>   <NA> none

thanks!

Answer 1

Try this tidyverse solution. You can aggregate your data using paste0() to get a proper structure in order to transform to wide format:

library(tidyverse)
#Code
temp %>%
  group_by(group,days) %>%
  summarise(fruits=paste0(fruits,collapse = ';')) %>%
  pivot_wider(names_from = group,values_from=fruits)

Output:

# A tibble: 3 x 4
  days  a            b      c    
  <chr> <chr>        <chr>  <chr>
1 mon   apple;orange NA     NA   
2 tues  NA           orange NA   
3 wed   NA           NA     none