R/convert_aa3.R

In EcoNum/seal.econum: Conversion, correction et analyse des données provenant du seal AA3 d'ECONUM

#' Conversion of aa3 file txt and excell
#'
#' Conversion du fichier fourni par l autoanalysuer seal AA3 au format txt  a l aide d un fichier complementaire au format xlsx
#'
#' @param file_aa3_txt txt file
#' @param file_aa3_xlsx xlsx file
#' @param project name of project
#' @param topic particulary information on the project
#'
#' @return un objet converti
#' @import readr
#' @import readxl
#' @export
#'
#' @examples
#'
#' #todo
convert_aa3 <- function(file_aa3_txt, file_aa3_xlsx, project, topic = NULL) {
  # Import metadata and extract informaation
  header_read <- readr::read_lines(file_aa3_txt, n_max = 13,
    locale = readr::locale(encoding = "LATIN1"))
  header <- strsplit(gsub('"', "", header_read), split = ";+")

  # Rename list elements
  names(header) <- sapply(header, `[[`, 1)

  # Paste Comment
  if (lengths(header["COMM"]) > 2) {
    header$COMM[2] <- paste(header$COMM[-1], collapse = " ")
    header$COMM <- header$COMM[1:2]
  }

  # Check_1
  header_control <- c("ANAL", "RUN ", "DATE", "TIME", "OPER", "COMM", "TYPE",
    "CHAN", "METH", "UNIT", "Base", "Gain", "Lamp")
  length_list_element <- c(2, 2, 2, 2, 2, 2, 4, 4, 4, 4, 4, 4, 4)

  if (sum(names(header) != header_control) != 0 |
    sum(lengths(header) != length_list_element) != 0)
    stop("erreur durant l importation ou l extraction des metadonnees")

  # Change NO3 in NOx
  header$METH <- gsub("NO3", "NOx", header$METH)

  # Extract nutrients that are analysed
  results <- header$METH[-1]

  # Prepare names of variables for raw_data
  stds <- paste(results, "std", sep = "_")
  concs <- paste(results, "conc", sep = "_")
  vals <- paste(results, "values", sep = "_")

  # Change in data NPinorganique.ANL in inorganique
  header$ANAL[2][header$ANAL[2]  == "NPinorganique.ANL"] <- "inorga"
  header$ANAL[2][header$ANAL[2]  == "NPorganique.ANL"] <- "orga"

  # Create a several list with the method information
  method <- list()
  for (i in 2:4) {
    method[[i - 1]] <- list(
      method = header$METH[i],
      unit   = header$UNIT[i],
      base   = header$Base[i],
      gain   = header$Gain[i],
      lamp   = header$Lamp[i])
  }

  bind_rows(method) -> .
    as.data.frame(.) -> method
  row.names(method) <- c("channel_1", "channel_2", "channel_3")

  # Create a list with the meta information
  meta <- list(
    sample      = paste(sub("\\.RUN$", "", header$RUN[2]), header$ANAL[2],sep = "-"),
    project     = project,
    sample_date = as.POSIXct(paste(header$DATE[2], header$TIME[2]), format = "%d/%m/%Y %H:%M:%S"), #lubridate::dmy_hms(paste(header$DATE[2], header$TIME[2])),
    author      = header$OPER[2],
    date        = as.Date(header$DATE[2], format = "%d/%m/%Y"), #lubridate::dmy(header$DATE[2]),
    comment     = header$COMM[2], topic = topic)

  # Extract raw data
  raw_data <- readr::read_delim(file = file_aa3_txt, delim = ";", skip = 13)

  # Select variables in raw_data
  raw_data <- raw_data[, -c(3, 5:7, 18, 19)]

  # Rename the columns in raw_data
  names(raw_data) <- c("sample_id", "peak_number", "sample_type", "date_time",
    stds[1], concs[1], vals[1], stds[2], concs[2], vals[2], stds[3], concs[3],
    vals[3])

  # Recoding type of variable
  raw_data$sample_type <- as.factor(raw_data$sample_type)
  raw_data$date_time <- as.POSIXct(raw_data$date_time, format = "%d/%m/%Y %H:%M:%S")  #lubridate::dmy_hms(raw_data$date_time)

  # Add units informations
  for (i in 1:3) {
    attr(raw_data[[stds[i]]], "units") <- method$unit[[i]]
    attr(raw_data[[concs[i]]], "units") <- method$unit[[i]]
  }

  # Import xlsx file
  add_data <- readxl::read_xlsx(file_aa3_xlsx, sheet = "data")

  raw_data <- dplyr::left_join(raw_data, add_data, by = "sample_id")

  # SAMP contient des valeurs manquantes
  for (i in which(raw_data$sample_type == "SAMP")) {
    if (is.na(raw_data$project[i]) == TRUE) {
      stop("presence de valeurs manquantes dans la colonnes project, le fichier xlsx et txt ne correspondent pas entierement.")
    }
  }

  lm_list <- list()
  for (i in 1:3) {
    # nutrient name
    nutri_name <-  strsplit(vals[i], split = "_", fixed = TRUE)[[1]][1]

    lm_mod <- stats::lm(data = raw_data, stats::as.formula(paste(vals[i],"~", stds[i])))

    # Compute n for the data.frame
    raw_data[raw_data$sample_type == "CALB", ] -> .
    dplyr::select(., stds[i]) -> .
    sum(!is.na(.)) -> n

    lm_list[[i]] <- data.frame(
      std_name  = paste(nutri_name),
      intercept = lm_mod$coefficients[[1]],
      values    = lm_mod$coefficients[[2]],
      r_squared = round(summary(lm_mod)$r.squared,digits = 3),
      n         = n
    )

    names(lm_list)[i] <- paste(nutri_name)
  }
  lm_df <- dplyr::bind_rows(lm_list)

  attr(raw_data, "method") <- method
  attr(raw_data, "calb_lm") <- lm_df

  #attr(raw_data, "metadata") <- meta
  #raw_data <- as.data.frame(raw_data)
  #class(raw_data) <- c("aa3", "data.frame")
  econum::new_econum_data(x = raw_data, metadata = meta,class = "aa3")
  #return(raw_data)
}