R/arrange_left.R

In skgrange/threadr: Tools to Thread Pieces Of Data Analysis Together

#' Function to arrange variables to the left of a data frame.
#' 
#' \code{arrange_left} is useful when some variables/columns want to be moved to
#' the left-side of a data frame. \code{arrange_left} will silently skip 
#' variables if they do not have a match in \code{names(df)}. 
#' 
#' @param df Data frame which is to be transformed. 
#' @param variable Variable(s) to be moved to the left side of \code{df}. 
#' 
#' @author Stuart K. Grange
#' 
#' @examples
#' \dontrun{
#' 
#' # Move a single variable to the first position
#' data_ozone <- arrange_left(data_ozone, "date")
#' 
#' # Move many variables to left hand side of data frame
#' data_ozone <- arrange_left(data_ozone, c("date", "site", "site_name", "sensor"))
#' 
#' }
#' 
#' @export
arrange_left <- function(df, variable) {
  
  # Soon to be dropped
  .Deprecated(msg = "`arrange_left` is deprecated, please use dplyr::relocate.")
  
  # Bound strings
  variable <- stringr::str_c("\\b", variable, "\\b")
 
  # Create index
  index <- lapply(variable, function(x) grep(x, names(df)))
  # Remove non-matches
  index <- index[lapply(index, length) > 0]
  index <- unlist(index)
  
  # Rearrange
  df <- df[, c(c(index), (1:ncol(df))[-index])]
  
  # Return
  df
  
}