R/summarise_dt.R

In tidyfst: Tidy Verbs for Fast Data Manipulation

#' @title Summarise columns to single values
#' @name summarise_dt
#' @description Summarise group of values into one value for each group. If there is only one group, then only one value would be returned.
#'  The summarise function should always return a single value.
#' @param .data data.frame
#' @param ... List of variables or name-value pairs of summary/modifications
#'   functions for \code{summarise_dt}.Additional parameters to be passed to
#'    parameter '.func' in \code{summarise_vars}.
#' @param by unquoted name of grouping variable of list of unquoted names of
#'   grouping variables. For details see \link[data.table]{data.table}
#' @param when An object which can be coerced to logical mode
#' @param .cols Columns to be summarised.
#' @param .func Function to be run within each column, should return a value or vectors with same length.
#' @details \code{summarise_vars} could complete summarise on specific columns.
#' @return data.table
#' @seealso \code{\link[dplyr]{summarise}}
#' @examples
#' iris %>% summarise_dt(avg = mean(Sepal.Length))
#' iris %>% summarise_dt(avg = mean(Sepal.Length),by = Species)
#' mtcars %>% summarise_dt(avg = mean(hp),by = .(cyl,vs))
#'
#' # the data.table way
#' mtcars %>% summarise_dt(cyl_n = .N, by = .(cyl, vs)) # `.` is short for list
#'
#' iris %>% summarise_vars(is.numeric,min)
#' iris %>% summarise_vars(-is.factor,min)
#' iris %>% summarise_vars(1:4,min)
#'
#' iris %>% summarise_vars(is.numeric,min,by ="Species")
#' mtcars %>% summarise_vars(is.numeric,mean,by = c("vs", "am"))
#'
#' # use multiple functions on multiple columns
#' iris %>%
#'   summarise_vars(is.numeric,.func = list(mean,sd,median))
#' iris %>%
#'   summarise_vars(is.numeric,.func = list(mean,sd,median),by = Species)
#'

globalVariables("res")

#' @rdname summarise_dt
#' @export

summarise_dt = function(.data,...,by = NULL){
  dt = as_dt(.data)
  eval.parent(substitute(dt[,.(...),by = by]))
}

#' @rdname summarise_dt
#' @export

summarize_dt = summarise_dt

#' @rdname summarise_dt
#' @export

summarise_when = function(.data,when,...,by = NULL){
  dt = as_dt(.data)
  eval.parent(substitute(dt[when,.(...),by = by]))
}

#' @rdname summarise_dt
#' @export

summarize_when = summarise_when

#' @rdname summarise_dt
#' @export

summarise_vars = function (.data, .cols = NULL, .func, ...,by) {
  dt = as_dt(.data)
  deparse(substitute(.cols)) -> .cols
  deparse(substitute(by)) -> .by
  if (.cols == "NULL")
    sel_name = names(dt[0])
  else{
    eval(
      parse(
        text =
          str_glue("select_dt(dt[0],{.cols}) %>% names() -> sel_name")))
  }

  if(!is.list(.func)){
    eval(parse(text = str_glue(
      "res = dt[,lapply(.SD, .func, ...), by = {.by},.SDcols = sel_name]")))
    res[, unique(names(res)), with = FALSE]
  }else{
    func_names = sapply(substitute(.func),deparse) %>% .[-1]
    lapply(func_names,function(func){
      eval(parse(text = str_glue(
        "res = dt[,lapply(.SD, get(func), ...),
        by = {.by},.SDcols = sel_name][,fun_name:=func]")))
    }) %>% rbindlist()
  }

}

#' @rdname summarise_dt
#' @export

summarize_vars = summarise_vars