CODEWITHSUNDEEP
rdataframejoinmergerow
Hawk McFadzen
Hawk McFadzen

Reputation: 1

Joining rows while maintaining all columns

folx. Long time listener, first-time caller. I have a df that looks like:

ID thing_1 thing_2 thing_3
A001 NA YES NA
A001 YES NA NA
A002 NA NA YES
A002 NA YES NA
A002 YES NA YES
A003 NA YES NA
-------- -------- -------- --------

I want it to look like:

ID thing_1 thing_2 thing_3
A001 YES YES NA
A002 YES YES YES
A003 NA YES NA
-------- -------- -------- --------

All of the resources I found about joining rows concatenate all the columns' data into one column separated by commas or else they dealt with adding numerical values into a sum. I am dealing with characters. How do I maintain separate columns but squash all of the data into one row per ID?

Thanks!

Upvotes: 0

Views: 38

Answers (1)

Andre Wildberg
Andre Wildberg

Reputation: 19088

This is one approach with tidyverse

library(dplyr)
library(tidyr)

df %>% 
  group_by(ID) %>% 
  fill(-ID, .direction="updown") %>% 
  distinct() %>% 
  ungroup()
# A tibble: 3 × 4
  ID    thing_1 thing_2 thing_3
  <chr> <chr>   <chr>   <chr>
1 A001  YES     YES     NA
2 A002  YES     YES     YES
3 A003  NA      YES     NA

Data

df <- structure(list(ID = c("A001", "A001", "A002", "A002", "A002",
"A003"), thing_1 = c(NA, "YES", NA, NA, "YES", NA), thing_2 = c("YES",
NA, NA, "YES", NA, "YES"), thing_3 = c(NA, NA, "YES", NA, "YES",
NA)), class = "data.frame", row.names = c(NA, -6L))

Upvotes: 1

Related Questions