CODEWITHSUNDEEP
r
nouse
nouse

Reputation: 3471

Looking up and replacing column names

df <- data.frame(A=1, B=2, C=3)
new.names <- data.frame(old=c("A","B","C"), new=c("X", "Y", "Z"))

How can i look up names(df) in new.names$old and replacing them with new.names$new?

Desired output:

names(df)
[1] "X" "Y" "Z"

Upvotes: 2

Views: 42

Answers (4)

Onyambu
Onyambu

Reputation: 79348

library(tidyverse)
df <- rename(df, !!invoke(set_names, unname(new.names)))
df
  X Y Z
1 1 2 3

Upvotes: 1

Leonardo Hansa
Leonardo Hansa

Reputation: 334

Two ways come to mind: base R and data.table. I changed the order of the names just to show that the look-up is working.

Base R

Instead of a data.frame with the names, I would use a named vector (using the old names as names of the components of this vector):

df <- data.frame(A=1, B=2, C=3)
new_names <- setNames(c("Y", "X", "Z"), c("B","A","C"))

names(df) <- new_names[names(df)]

data.table

More straightforward:

library(data.table)
df <- data.frame(A=1, B=2, C=3)
setnames(df, old = c("B","A","C"), new = c("Y", "X", "Z"))
# Note that setnames() doesn't return anything. You can 
# run names(df) for seeing the result.s

Upvotes: 1

TarJae
TarJae

Reputation: 79311

In plyr package we could use mapvalues function

library(plyr)
df_new <- df
  colnames(df_new) <- plyr::mapvalues(colnames(df),
                                   as.character(new.names$old),
                                   as.character(new.names$new))

Output:

> df_new  
  X Y Z
1 1 2 3

Upvotes: 1

tmfmnk
tmfmnk

Reputation: 40171

One way to do it could be:

names(df) <- new.names$new[match(new.names$old, names(df))]

  X Y Z
1 1 2 3

Upvotes: 2

Related Questions