Read a transposed (variables in rows) CSV file into R correctly

read.tscv.R

## Based on
## https://stackoverflow.com/a/17289991/4168169
read.tcsv = function(file, header=TRUE, sep=",", ...) {
  
  n = max(count.fields(file, sep=sep), na.rm=TRUE)
  x = readLines(file)
  
  .splitvar = function(x, sep, n) {
    var = unlist(strsplit(x, split=sep))
    length(var) = n
    return(var)
  }
  
  x = do.call(cbind, lapply(x, .splitvar, sep=sep, n=n))
  x = apply(x, 1, paste, collapse=sep) 
  ## empty strings are converted to NA
  out = read.csv(text=x, sep=sep, header=header, na.strings = "", ...)
  return(out)
  
}

## If we have a transposed CSV file
## (rows are data, should be columns)
cat(readLines("tdata.csv"), sep = "\n")
#> var1,1,2,3,4
#> var2,0.8354,0.8359,0.5590,0.8812
#> var3,apple,banana,,orange

## We can read this in and transpose using read.tcsv above
(tdata <- read.tcsv("tdata.csv"))
#>   var1   var2   var3
#> 1    1 0.8354  apple
#> 2    2 0.8359 banana
#> 3    3 0.5590   <NA>
#> 4    4 0.8812 orange

## Which interprets values correctly
str(tdata)
#> 'data.frame':    4 obs. of  3 variables:
#>  $ var1: int  1 2 3 4
#>  $ var2: num  0.835 0.836 0.559 0.881
#>  $ var3: Factor w/ 3 levels "apple","banana",..: 1 2 NA 3

## Including not making factors if we wish
(tdata <- read.tcsv("tdata.csv", stringsAsFactors = FALSE))
#>   var1   var2   var3
#> 1    1 0.8354  apple
#> 2    2 0.8359 banana
#> 3    3 0.5590   <NA>
#> 4    4 0.8812 orange
str(tdata)
#> 'data.frame':    4 obs. of  3 variables:
#>  $ var1: int  1 2 3 4
#>  $ var2: num  0.835 0.836 0.559 0.881
#>  $ var3: chr  "apple" "banana" NA "orange"