R/transform_coordinates.R

In skgrange/gissr: Tools To Make Working With R and Spatial Data Easier

#' Function to convert coordinates within a data frame. 
#' 
#' \code{transform_coordinates} converts coordinate pairs from one projection 
#' system to another. \code{transform_coordinates} is useful when tabular data 
#' are supplied and the coordinate system is different that what is desired. 
#' 
#' \code{transform_coordinates} works by coercing the input data frame to a 
#' spatial object, applies \code{spTransform} to convert the coordinates, 
#' converts the spatial object back to a data frame and then returns the data 
#' frame with the transformed coordinates. The transformed coordinates can be 
#' optionally renamed and reordered. 
#' 
#' \code{transform_coordinates} requires a CRS projection string for the 
#' \code{from} and \code{to} arguments. \code{to} by default is set as 
#' WGS 84/EPSG:4326 (\code{+proj=longlat +datum=WGS84}).
#' 
#' @param df Data frame with coordinates to be transformed. 
#' 
#' @param latitude Name of latitude/y variable. 
#' 
#' @param longitude Name of longitude/x variable. 
#' 
#' @param from A proj4 string which represents what coordinate system the
#' data frame's coordinates are in. 
#' 
#' @param to A proj4 string which represents what coordinate system the 
#' converted coordinates will be converted to.
#' 
#' @seealso \code{\link{sp_transform}}, \code{\link{sp_from_data_frame}}
#' 
#' @author Stuart K. Grange
#' 
#' @return Tibble.
#' 
#' @examples 
#' 
#' \dontrun{
#' 
#' # Convert British National Grid/Ordnance Survey National Grid/OSGB36/EPSG:7405
#' # to latitude and longitude (WGS 84/EPSG:4326)
#' 
#' data_oxford_transform <- transform_coordinates(
#'   data_oxford, 
#'   latitude = "latitude", 
#'   longitude = "longitude", 
#'   from = projection_bng()
#' )
#'   
#' }
#'
#' @export
transform_coordinates <- function(df, latitude = "latitude", 
                                  longitude = "longitude", from, to) {
  
  # Get variable order
  variables <- names(df)
  
  # Promote
  sp <- sp_from_data_frame(
    df, 
    latitude, 
    longitude, 
    projection = from, 
    type = "points"
  )
  
  # Do the projection conversion
  sp <- sp_transform(sp, to)
  
  # Back to data frame and drop optional variable if exists
  df <- sp %>% 
    data.frame() %>% 
    select(-matches("optional"))
  
  # Rename coordinates that come from row names
  names(df) <- if_else(names(df) == "coords.x1", "longitude", names(df))
  names(df) <- if_else(names(df) == "coords.x2", "latitude", names(df))
  
  # Arrange variables in original order
  df <- df %>% 
    select(!!variables) %>% 
    as_tibble()
    
  return(df)
  
}