How to replace data.frame column names with string in corresponding lookup table in R

Question

I have the following data.frame:

set.seed(126)
df <- data.frame(a=sample(c(1:100, NA), 10), b=sample(1:100, 10), c=sample(1:100, 10), d = c(1:10))
    a  b  c  d
1  18 27 53  1
2  44 16 66  2
3  58 47  3  3
...

And the following lookup table:

varnames <- data.frame(old = c("a", "b", "c"), new = c("dog", "cat", "mouse"))
  old   new
1   a   dog
2   b   cat
3   c mouse

What I am trying to do is replace the names(df) with the corresponding varnames$new... If a names(df) is not in varnames$old, then retain the colname in df...

The resulting data.frame I would like returned would look like this:

   dog cat mouse  d
1   57  10    83  1
2   53  99    94  2
3   99  60    39  3
...

Any and all help appreciated.

Answer 1

How about using the match() function

mm <- match(names(df), varnames$old)
names(df)[!is.na(mm)] <- as.character(varnames$new[na.omit(mm)])
head(df)
#   dog cat mouse d
# 1  65  48    19 1
# 2  46  15    80 2
# 3  NA  47    84 3
# 4  68  34    46 4
# 5  23  75    42 5
# 6  92  87    68 6

If you are interested, you could also use the dplyr rename() function

library(dplyr)
df %>% rename_(.dots=with(varnames, setNames(as.list(as.character(old)), new)))

Or one more option, the data.table package has a setnames function

library(data.table)
setnames(df, as.character(varnames$old), as.character(varnames$new))

Answer 2

One more option is mapvalues() from the plyr package:

library(plyr)
names(df) <- mapvalues(names(df),
                             from = varnames$old,
                             to = as.character(varnames$new))

If you are using the dplyr package, you can call this with plyr::mapvalues() so you don't have to load plyr on top of dplyr (which causes problems).