getLongURL.curl

gistfile1.txt

getLongURL.curl <- function(shorturl){
  # uses curl statement to get expanded url from t.co links (or any link)
  # loop through until there's no location attribute... that's the long link.
  newurl <- shorturl
  url <- ""
  while(url != newurl){
    data <- system(paste0("curl -I ", newurl), intern=T)
    if(sum(grepl("location: ", tolower(data))) == 0){
      url <- newurl
    }else{
      data <- subset(data, tolower(substring(data, 1, 9))=="location:")
      stringurl <- substring(data[1], 11, nchar(data[1])-1)
      # sometimes the location data isn't a url.
      if(substring(stringurl, 1, 4)=="http"){ 
        newurl <- stringurl
      }else{
        url <- newurl
      }
    }
  }
  return(newurl)
}