R/mutate_dt.R

In tidyfst: Tidy Verbs for Fast Data Manipulation

#' @title Mutate columns in data.frame
#' @description Adds or updates columns in data.frame.
#' @param .data data.frame
#' @param ... List of variables or name-value pairs of summary/modifications
#'   functions.
#' @param by (Optional) Mutate by what group?
#' @return data.table
#' @seealso \code{\link[dplyr]{mutate}}
#' @examples
#'
#' iris %>% mutate_dt(one = 1,Sepal.Length = Sepal.Length + 1)
#' iris %>% transmute_dt(one = 1,Sepal.Length = Sepal.Length + 1)
#' # add group number with symbol `.GRP`
#' iris %>% mutate_dt(id = 1:.N,grp = .GRP,by = Species)
#'
#' @rdname mutate
#' @export

mutate_dt = function(.data,...,by){
  dt = as.data.table(.data)
  eval.parent(substitute(dt[,`:=`(...),by][]),1)
}

# mutate_dt = function(.data,...,by){
#   dt = as.data.table(.data)
#   substitute(dt[,`:=`(...),by][]) %>% eval()
# }

#' @rdname mutate
#' @export

transmute_dt = function(.data,...,by){
  dt = as_dt(.data)
  eval.parent(substitute(dt[,.(id_ = 1:.N,...),by]),1)[,id_:=NULL][]
}

# transmute_dt = function(.data,...,by){
#   dt = as_dt(.data)
#   substitute(dt[,.(id_ = 1:.N,...),by][]) %>% eval() %>%
#     .[,id_:=NULL] %>% .[]
# }

globalVariables("id_")

# transmute_dt = function(.data,...,by){
#   dt = as_dt(.data)
#   substitute(dt[,.(...),by][]) %>% eval()
# }