edar: Package for eyetracking data analysis

#' load_data
#' Function for loading data from multiple .asc files into R.
#' @param datadir [character vector] with the path to the folder where the data is located. If of length
#' greater than 1, it is assumed to be a vector of paths to files that are to be read in.
#' @import data.table
#' @import parallel
#' @import purrr
#' @importFrom BBmisc extractSubList
#' @export
load_data <- function(files, trial_seperators = NULL, ncpus = 1) {

    datlist = parallel::mclapply(seq_along(files), function(i) {
        asc = read_ascii(files[i])
        asc = slice_into_trials(asc, trial_seperators)
        for (el in c("raw", "msg", "fix", "sacc", "blink", "raw_notrial"))
            asc[[el]][, subject := sub(".+([0-9]{3}).asc$", "\\1", basename(files[i]))]
        asc
    }, mc.cores = ncpus)
    datlist = purrr::transpose(datlist)
    for (el in c("raw", "msg", "fix", "sacc", "blink", "raw_notrial")) {
        datlist[[el]] = data.table::rbindlist(datlist[[el]])
        setkeyv(datlist[[el]], c("subject", "trial"))
    }
    class(datlist) = c(class(datlist), "edar_data")
    datlist
}

## load_data <- function(datadir, savedir = "", prefix = "", trial_seperators = NULL,
##                       with_images = FALSE, discard_oot_data = FALSE, ncpus = 1, ...) {
##     ## if (missing(prefix))
##     ##     stop("Please provide prefix for naming the data files.")

##   if (length(datadir) > 1 || file.exists(datadir)) {
##     files = datadir 
##     if (! all(file.exists(files)))
##       stop(sprintf("Files %s could not be found.", 
##                    paste(files[!file.exists(files)], collapse = ", ")))
##     print("Reading in the following files: ")
##     print(files)
##     if (interactive())
##       invisible(readline(prompt="Press [enter] to continue"))
##   } else {
##     ## get the ascii files
##     files <- grep(".asc$", dir(datadir, full.names = TRUE), value = TRUE)

##     ## filter with prefix
##     files <-
##       files[grep(paste0("^", prefix), basename(files))]
##     print("Reading in the following files: ")
##     print(files)
##     print("You can restrict the selection by using the 'prefix' argument.")
##     if (interactive())
##       invisible(readline(prompt="Press [enter] to continue"))
##   } 

##   ## extract ids from file names
##   ids <- sub("_[A-z]$", "", gsub("[.][[:alnum:]]+", "", basename(files)))

##   if (with_images) {
##     ## get questions and images
##     images <- get_images(datadir)
##   }

##   ## load and sort out all files
##   system.time(
##     datalist <- parallel::mclapply(seq_along(files), function(i) {
##       id_i <- ids[i]
##       print(paste0("loading data for id: ", id_i, " ... "))

##       if (id_i == "fMmW") {
##         asc <- read_ascii(path_ascii = files[i], save_rdata = "", number_trials = 171,
##                           trial_seperators = trial_seperators)
##         ## add subject indicators
##         asc$raw[ , subject := id_i]
##         if (!is.null(asc$raw_notrial))
##             asc$raw_notrial[ , subject := NULL]
##         asc$msg[ , subject := id_i]
##         asc$fix[ , subject := id_i]
##         asc$sacc[ , subject := id_i]
##         asc$blink[ , subject := id_i]
##       } else {
##         asc <- read_ascii(path_ascii = files[i])

##         ## slice into trials
##         if (! is.null(trial_seperators))
##           asc <- slice_into_trials(asc, trial_seperators)

##         ## add subject indicators
##         asc$raw[ , subject := id_i]
##         if (!is.null(asc$raw_notrial))
##             asc$raw_notrial[ , subject := id_i]
##         asc$msg[ , subject := id_i]
##         asc$fix[ , subject := id_i]
##         asc$sacc[ , subject := id_i]
##         asc$blink[ , subject := id_i]
##       }
##       asc
##     }, mc.cores = ncpus)
##   )

##   if (length(datalist) > 1)
##     ## bind all eye data
##     data <- list(raw = rbindlist(BBmisc::extractSubList(datalist, "raw", simplify = FALSE)),
##                  raw_notrial = rbindlist(BBmisc::extractSubList(datalist, "raw_notrial", simplify = FALSE), fill = TRUE),
##                  msg = rbindlist(BBmisc::extractSubList(datalist, "msg", simplify = FALSE)),
##                  fix = rbindlist(BBmisc::extractSubList(datalist, "fix", simplify = FALSE)),
##                  sacc = rbindlist(BBmisc::extractSubList(datalist, "sacc", simplify = FALSE)),
##                  blink = rbindlist(BBmisc::extractSubList(datalist, "blink", simplify = FALSE)))
##   else
##     data = datalist[[1]]

##   class(data) <- c(class(data), "edar_data")
##   setkeyv(data$raw, c("subject", "trial"))
##   setkeyv(data$raw_notrial, c("subject", "trial"))
##   setkeyv(data$msg, c("subject", "trial"))
##   setkeyv(data$fix, c("subject", "trial"))
##   setkeyv(data$sacc, c("subject", "trial"))
##   setkeyv(data$blink, c("subject", "trial"))

##   ## add resolution to data
##   f1 = read.table(files[1], nrows = 80, sep = "\t", 
##                   col.names = paste0("V", 1:9),
##                   fill = TRUE, header = FALSE, stringsAsFactors = FALSE)
##   names(f1)[1:4] = c("t", "x", "y", "pups")
##   ss = strsplit(f1[grep("DISPLAY_COORDS", f1[,2]), 2], " ")[[1]]
##   coords = as.numeric(ss[(length(ss)-1):length(ss)])
##   data$info$resolution = list(x = coords[1], y = coords[2])

##   ## save data
##   if (savedir != "") {
##     saveRDS(data, file = file.path(savedir, paste0("data_", prefix, "_all.rds")))
##   }
##   data
## }