Question

I’ve got a data.frame with 3 columns. The third column has either numbers or character type “1:5”, “30:20”, “1:10”, etc. I’m trying to split the values which have those characters into two and then divide between them but I’m stuck with this:

    datos[,3]=gsub(":", " ", datos[,3])
    if (datos[,1]==TRUE)
    {
    s=datos[,3]
    chr.pos <- which(unlist(strsplit(s,NULL)) == " ") 
    chr.count <- length(chr.pos)
    one=as.numeric(substr(s,1,chr.pos-1))
    two=as.numeric(substr(s,chr.pos+1,nchar(as.character(s))))
    datos[,3]=round(two/one,5)
    }

Answer 1

Try this. Note: Added ‘col.names’ to suppress default handling of rownames.

x=c("1", "2", "3", "2:3","4","5","3:2")
 datos <- data.frame(1:7, 1:7, x=x)
newframe <- cbind( datos[1:2], 
                 read.table(text= as.character(datos[[3]]), sep=":",
                            fill=TRUE, colClasses="numeric", 
                           col.names=c("V3", "V4")
                           )
                  )

> newframe
  X1.7 X1.7.1 V3 V4
1    1      1  1 NA
2    2      2  2 NA
3    3      3  3 NA
4    4      4  2  3
5    5      5  4 NA
6    6      6  5 NA
7    7      7  3  2