R/readResultCSV.R

In hydrosolutions/riversCentralAsia: Accompanying R Package to Applied Modeling of Hydrological Systems in Central Asia Course

#' Reads RS Minerve result file and returns data.
#'
#' @param filename Path to file to be read. It is any csv file that is written
#'   by saving data from the Selection and plots tab in RSMinerve to a csv file.
#'   Depending on the systems locale, the csv may use comas or semi-colons as
#'   separators. The decimal separator must be a point, commas are not
#'   recognized.
#' @param tz Time zone character passed to as.POSIXct. Defaults to "UTC".
#' @return data Tibble with data from file. Null if file not found or cant be read.
#' @note An example RSMinerve results csv file is available under \url{http://raw.githubusercontent.com/hydrosolutions/RSMinerveR/main/tests/testthat/test_translateCSVtoDST.csv}. If the function generates funny numbers, please check the csv file. The decimal separators must be \code{.} and not \code{,}. If you find commas please change the settings of your PC (Control Panel, Clock and Region, Region, Additional settings, select \code{.} for the Decimal symbol, press "Apply" and close the Control Panel windows. and RSMinerve will write \code{,}.
#' @family RS Minerve IO
#' @export
#'
readResultCSV <- function(filename, tz = "UTC") {

  if (!utils::file_test("-f", filename)) {
    cat("ERROR: File", filename, "not found.\n")
    return(NULL)
  }

  # Test delimiter of csv file
  test <- readr::read_csv(filename, skip = 2, col_names = F)
  if (dim(test)[2] == 1) {

    # Not tab delimited, try semi-colon
    data <- readr::read_delim(filename, skip = 1, col_names = T,
                              delim = ";")

  } else {

    data <- readr::read_csv(filename, skip = 1, col_names = T)

  }

  colnames(data)[1] <- "date"
  data$date <- as.POSIXct(data$date, format = "%d.%m.%Y %H:%M:%S", tz = tz)
  data <- data |>
    dplyr::mutate(dplyr::across(-date, ~base::as.numeric(.))) |>
    tidyr::pivot_longer(-date, names_to = "model", values_to = "value") |>
    tidyr::separate(model, into = c("model", "variable"), sep = "( - )",
                    extra = "merge") |>
    tidyr::separate(variable, into = c("variable", "unit"),
                    sep = "\\s\\(") |>
    dplyr::mutate(unit = gsub("\\)", "", unit))
  return(data)
}