Move characters from beginning of column name to end of column name (additonal question)

Question

I need an additional solution to the previous question/answer Move characters from beginning of column name to end of column name

I have a dataset where column names have two parts divided by _ e.g.

pal036a_lon 
pal036a_lat 
pal036a_elevation

I would like to convert the prefixes into suffixes so that it becomes:

lon_pal036a 
lat_pal036a 
elevation_pal036a

The answer to the previous question

names(df) <- sub("([a-z])_([a-z]+)", "\\2_\\1", names(df))

does not work for numbers within the prefixes.

I think in general I would opt to use a negated character class in such case like `([^_]+)_([^_]+)` for example, or if these column names are typically just word-characters: `(\w+)_(\w+)` — JvdV, Feb 16 '21 at 08:52

s_baldur · Answer 1 · 2021-02-16T08:47:49.657

2

Assuming your names have a single _. You could also you strsplit():

sapply(strsplit(names(df), '_'), function(x) paste(rev(x), collapse = '_'))

If you have more than one you could modify the above as suggested by jay.sf:

sapply(strsplit(x, "_"), function(x) paste(c(x[length(x)], x[-length(x)]), collapse="_"))

edited Feb 16 '21 at 08:47

answered Feb 16 '21 at 08:38

s_baldur

29,441
4
36
69

Thanks! in this case, I do not have to change upper case letter into lower case letter before – Marcin Koprowski Feb 16 '21 at 09:06
@MarcinKoprowski This solution doesn't change modify upper/lower case. – s_baldur Feb 16 '21 at 09:18

score 1 · Accepted Answer · answered Feb 16 '21 at 08:35

You can include alphanumeric characters in the first group:

names(df) <- sub("([a-z0-9]+)_([a-z]+)", "\\2_\\1", names(df))

For example :

x <- c("pal036a_lon","pal036a_lat","pal036a_elevation")
sub("([a-z0-9]+)_([a-z]+)", "\\2_\\1",x)
#[1] "lon_pal036a"       "lat_pal036a"       "elevation_pal036a"

Move characters from beginning of column name to end of column name (additonal question)

2 Answers2

Linked