R/read_otg_csv.R

In mackerman44/DASH: DASH: Drone Assisted Stream Habitat

#' @title Read On-The-Ground .csv Data
#'
#' @description Import .csv delimited files containing on-the-ground (OTG) data collected using the
#' DASH protocol. This function only reads in one `otg_type` of data at a time.
#'
#' @inheritParams get_file_nms
#' @inheritParams get_otg_col_specs
#'
#' @import purrr dplyr
#' @importFrom readr read_csv
#' @importFrom magrittr %>%
#' @importFrom compare compare
#' @importFrom stringr str_split
#' @importFrom lubridate mdy
#' @export
#' @return a data frame containing data of \code{otg_type}

read_otg_csv = function(path = ".",
                        otg_type = c("surveyPoint_0.csv",
                                     "CU_1.csv",
                                     "Wood_2.csv",
                                     "Jam_3.csv",
                                     "Undercut_4.csv",
                                     "Discharge_5.csv")) {

  # files list of the otg_type
  file_list = get_file_nms(path) %>%
    dplyr::filter(file_nm == otg_type) %>%
    #dplyr::mutate(file_cnt = row_number()) %>%
    split(list(.$path_nm))

  # grab intended column specifications using get_otg_col_specs()
  otg_col_specs = get_otg_col_specs(otg_type)

  # map_df over file_list
  tmp_df = file_list %>%
    purrr::map_df(.id = "path_nm",
                  .f = function(x) {

                    cat(paste("Importing", otg_type, "file from", x$folder_nm, "survey folder.", "\n"))

                    # get column specifications from file
                    tmp_specs = readr::read_csv(paste0(path, x$path_nm), col_types = readr::cols()) %>%
                      readr::spec()

                    #####
                    # CHECK 1: check if number of columns match
                    chk = identical(length(otg_col_specs$cols), length(tmp_specs$cols))
                    if(chk == FALSE) stop(paste("Fatal Error: Number of columns in", otg_type, "file from", x$folder_nm, "survey folder is unexpected."))

                    #####
                    # CHECK 2: check if column names match
                    chk = identical(names(otg_col_specs$cols), names(tmp_specs$cols))
                    if(chk == FALSE) cat(paste("Column names in", otg_type, "file from", x$folder_nm, "survey folder do not match the expected names defined in get_otg_col_specs(). Check file format. Continuing...", "\n"))

                    # read in the single .csv file of otg_type
                    tmp = try(suppressWarnings(readr::read_csv(paste0(path, x$path_nm),
                                                               col_types = otg_col_specs)))

                    #####
                    # CHECK 3
                    if(chk == FALSE | class(tmp)[1] == "try-error") {
                      cat(paste("Problem reading in", otg_type, "file from", x$folder_nm, "survey. Returning NULL and moving on...", "\n"))
                      return(NULL)
                    } else if(nrow(tmp) == 0) {
                      cat(paste("The", otg_type, "file from", x$folder_nm, "survey contains zero records. Returning NULL and moving on...", "\n"))
                      return(NULL)
                    } else if(chk == TRUE) {
                      cat(paste("Success!", "\n"))
                      return(tmp)
                    }

                  })

  # DELETE ANY ROWS WHERE "GlobalID" is NA...
  if("GlobalID" %in% names(tmp_df)) {
    tmp_df = tmp_df %>%
      dplyr::filter(!is.na(GlobalID))
  }

  if(nrow(tmp_df) == 0) {
    return(NULL)
  } else {
    return(tmp_df)
  }

} # end read_otg_csv()