I have a data.table storing heterogeneous file names in a string column. I want to extract the extension from that column, always taking the characters after the last dot occurrence. Sometimes the filename may contain more dots.
I tried:
files0=data.table(filename=c("simple_file.csv","file with.two dots.xls"))
files0[,chunks:=length(tstrsplit(filename,"\\."))]
files0[,extension:=tstrsplit(filename,"\\.")[chunks]]
How do I make sure that tstrsplit is only applied to each row so that this approach works?
PS: I also managed to generate a column with storing the correct number of text "chunks" with str_count, but the problem remains that when I want to create the "extension" column, the entire "filename" column seems to be used for each row.
No need for sapply or strsplit, those will add unnecessary complexity and inefficiency. We can use tools::file_ext (built-in to R) or just do a sub ourselves.
dat[, tools::file_ext(filename)]
# [1] "csv" "xls"
dat[, sub(".*\\.", "", filename)]
# [1] "csv" "xls"