split column into two with "/" separator in r

Question

I have following type of data, although data.frame is very large.

A1 <- c("A/B", "A/A", "B/B", "A/A")
B1 <- c("B/B", "C/C", "C/B", "D/A")
C1 <- c("B/B", "C/C", "C/B", "D/A")
mydf <- data.frame (A1, B1, C1)
mymat <- as.matrix (mydf)
mymat
         A1    B1    C1   
[1,] "A/B" "B/B" "B/B"
[2,] "A/A" "C/C" "C/C"
[3,] "B/B" "C/B" "C/B"
[4,] "A/A" "D/A" "D/A"

I want split each columns into two "/" is sperator. Thus output would look like:

    A1a  A1b   B1a  B1b    C1a  C1b   
[1,] A   B      B    B     B     B 
[2,] A   A      C    C     C     C
[3,] B   B      C    B     C     B
[4,] A   A      D    A     D     A

Answer 1

Here's a mildly devious way. Add "/"'s between the characters and read in with that as separator:

splitdf <- read.table(text=capture.output(write.table(mydf, quote=FALSE, 
                                        sep="/",col.names=FALSE,row.names=FALSE)),
                       sep="/")
names(splitdf) <- paste( rep(names(mydf), each=2), c("a","b"), sep="")
 splitdf
#------------
  A1a A1b B1a B1b C1a C1b
1   A   B   B   B   B   B
2   A   A   C   C   C   C
3   B   B   C   B   C   B
4   A   A   D   A   D   A

Answer 2

Here's one solution using reshape2's colsplit:

require(reshape2)
mymat <- as.data.frame(mymat)

LS <- lapply(seq_along(mymat), function(i){
    colsplit(mymat[, i], "/", paste0(colnames(mymat)[i], letters[1:2]))
    }
)

do.call('cbind', LS)

Which yields:

  A1a A1b B1a B1b C1a C1b
1   A   B   B   B   B   B
2   A   A   C   C   C   C
3   B   B   C   B   C   B
4   A   A   D   A   D   A