R/workspace.R

In hstojic/hfunk: Useful R functions

# ----------------------------------------------------------------------
# An improved listing of objects in the working memory. 
# ----------------------------------------------------------------------
#' A nicely formatted listing of objects in the R working memory. 
#' 
#' An improved listing of the objects available in R memory. Built-in function ls() does not provide information such as how much memory an object takes, which might be important if one is working with large objects. The function relies on internal function .ls_objects().
#'
#' @param n A scalar indicating how many objects to print in the output. By default 20 objects will be printed in decreasing order according to the memory they require.
#' @return A data frame with objects as rows and details about them in five variables: type, size, nicely formatted size, number of rows and number of columns in the object.
#' @seealso \code{\link{ls}}
#' @import assertthat
#' @export
#' @examples
#' # create some objects
#' x <- 1:4
#' y <- runif(10e6)
#'
#' # inspect the objects in the memory and their details
#' lsos()

lsos <- function(n = 20) {
    # basic assertions
    assert_that(is.scalar(n))
    # output
    .ls_objects(order_by = "Size", decreasing = TRUE, head = TRUE, n = n)
}


# ----
# internal function for lsos()
# ----
#' A nicely formatted listing of objects in the R working memory. 
#' 
#' An internal function for providing a summary of objects present in the working environment of the R session. See also: http://stackoverflow.com/questions/1358003/tricks-to-manage-the-available-memory-in-an-r-session.
#'
#' @param pos A scalar indicating the position argument to ls() function.
#' @param pattern A regular expression argument to ls() function.
#' @param order_by Variable name to order the output by.
#' @param decreasing A logical value, when ordered by some variable, will it be in decreasing or increasing order.
#' @param head A logical value, whether to print out only first \code{n} rows.
#' @param n A scalar indicating how many objects to print in the output. 
#' @return A data frame with objects as rows and details about them in five variables: type, size, nicely formatted size, number of rows and number of columns in the object.
#' @importFrom utils capture.output object.size

.ls_objects <- function (pos = 1, pattern, order_by,
                         decreasing = FALSE, head = FALSE, n = 5) {
    
    napply <- function(names, fn) sapply(names, function(x)
                                         fn(get(x, pos = pos)))
    names <- ls(pos = pos, pattern = pattern)
    obj_class <- napply(names, function(x) as.character(class(x))[1])
    obj_mode <- napply(names, mode)
    obj_type <- ifelse(is.na(obj_class), obj_mode, obj_class)
    obj_prettysize <- napply(names, function(x) {
        capture.output(print(object.size(x), units = "auto")) })
    obj_size <- napply(names, object.size)
    obj_dim <- t(napply(names, function(x)
                        as.numeric(dim(x))[1:2]))
    vec <- is.na(obj_dim)[, 1] & (obj_type != "function")
    obj_dim[vec, 1] <- napply(names, length)[vec]
    out <- data.frame(obj_type, obj_size, obj_prettysize, obj_dim)
    names(out) <- c("Type", "Size", "PrettySize", "Rows", "Columns")
    if (!missing(order_by))
        out <- out[order(out[[order_by]], decreasing=decreasing), ]
    if (head) out <- head(out, n)
    return(out)
}