How to fix spaces in column names of a data.frame (remove spaces, inject dots)?

Question

After importing a file, I always try try to remove spaces from the column names to make referral to column names easier.

Is there a better way to do this other then

Answer 1

It's often convenient to change the names of your columns within one chunk of dplyr code rather than renaming the columns after you've created the data frame. Piping in rename_all() is very useful in these situations:

ctm2 %>% rename_all(function(x) gsub(" ", "_", x))

The code above will replace all spaces in every column name with an underscore.

Answer 2

There is an easy way to remove spaces in column names in data.table. You will have to convert your data frame to data table.

setnames(x=DT, old=names(DT), new=gsub(" ","",names(DT)))

Country Code will be converted to CountryCode

Answer 3

To replace only the first space in each column you could also do:

names(ctm2) <- sub(" ", ".", names(ctm2))

or to replace all spaces (which seems like it would be a little more useful):

names(ctm2) <- gsub(" ", "_", names(ctm2))

or, as mentioned in the first answer (though not in a way that would fix all spaces):

spaceless <- function(x) {colnames(x) <- gsub(" ", "_", colnames(x));x}
newDF <- spaceless(ctm2)

where x is the name of your data.frame. I prefer to use "_" to avoid issues with "." as part of an ID.

The point is that gsub doesn't stop at the first instance of a pattern match.

Answer 4

Assign the names like this. This works best. It replaces all white spaces in the name with underscore.

names(ctm2)<-gsub("\\s","_",names(ctm2))

Answer 5

There exists more elegant and general solution for that purpose:

tidy.name.vector <- make.names(name.vector, unique=TRUE)

make.names() makes syntactically valid names out of character vectors. A syntactically valid name consists of letters, numbers and the dot or underline characters and starts with a letter or the dot not followed by a number.

Additionally, flag unique=TRUE allows you to avoid possible dublicates in new column names.

As code to implement

d<-read_delim(urltxt,delim='\t',)
names(d)<-make.names(names(d),unique = TRUE)

Answer 6

There is a very useful package for that, called janitor that makes cleaning up column names very simple. It removes all unique characters and replaces spaces with _.

library(janitor)

#can be done by simply
ctm2 <- clean_names(ctm2)

#or piping through `dplyr`
ctm2 <- ctm2 %>%
        clean_names()