R - How do I add a new column to a dataframe that is the calculated result of another column

Question

I'm trying to write a function in R that maps input information to outputs, but retains the input in the same data frame. I have a function that I wrote that will take the input and provide the correct output for single values, but I lack the R knowledge to know how to rewrite this so I can pass in a vector.

I want to pass it in a frame, and add a new column to the dataframe. Example code below:

d <- data.frame(Type=c("_Swap_","Nothing","CAP", "FLOOR", "FLOOR", "BLAH", "Digital", "Something!"))

d$newType= createTags(d$Type)

createTags = function(Type){
   map = data.frame(input=c("Swap","Note","Option", "Floor", "FLOOR", "CAP", "Digital", "Bond"), 
                  output=c("Swap","Note","Option", "CapFloor", "CapFloor", "CapFloor", "CapFloor", "Bond"))

   tag="Unknown"

   print(length(Type))

   for (j in 1:NROW(map)) {    
     input = map[j,]$input
     output = map[j,]$output

     ifelse (grepl(toupper(input), toupper(Type))){
       tag = output
     }    
   }

   return(tag)  
 }

So in the output I want d to have two columns, one is the original Type, and then at the end there should be a second column called newType.

Answer 1

You can use the switch function to map inputs to outputs:

# Create sample data
d <- data.frame(
  type = c("_Swap_","Nothing","CAP", "FLOOR", "FLOOR", "BLAH", "Digital", "Something!")
)

# Use the switch function to map inputs to outputs
create_tags <- function(type) {
  type <- gsub("[[:punct:]]", "", tolower(type)) # standardize the letter case and remove punctuation
  switch(
    type,
    "swap"    = "Swap",
    "note"    = "Note",
    "option"  = "Option",
    "floor"   = "CapFloor",
    "cap"     = "CapFloor",
    "digital" = "CapFloor", 
    "bond"    = "Bond",
    "Unknown"
  )
}  

# Switch isn't vectorized so we have to sapply our create_tags function onto d$type
d$new_type <- sapply(d$type, create_tags)

Answer 2

I ended up playing around with this alot. Thanks Mark Timms for the help, although I realised I needed to use grepl to make sure that I could cover all the cases

SRTLondon <- 
   data.frame(Type=c("_Swap_","Nothing","CAP", "FLOOR", "FLOOR", "BLAH", "Digital", "Something!"))

createTags = function(Type){
   #browser()
   map = data.frame(input=c("Swap","Note","Option", "Floor", "FLOOR", "CAP", "Digital", "Bond"), 
               output=c("Swap","Note","Option", "CapFloor", "CapFloor", "CapFloor", "CapFloor", "Bond"),stringsAsFactors = F)

   work = data.frame(x=toupper(Type),stringsAsFactors = F)
   work$tag="NA"

   for (i in 1:NROW(work)) {  
    check = work$x[i]

    for (k in 1:nrow(map)){
      input = map[k,'input']
      output = map[k,'output']

      if(grepl(toupper(input),check))
      {
        work$tag[i]=output
        break
      }

    }

  }
  return(work$tag)
}

SRTLondon$type.tags= createTags(SRTLondon$Type)