R/pibmunic.R

In datazoom.amazonia: Simplify Access to Data from the Amazon Region

#' @title PIB MUNICIPAL - Municipal GDP
#'
#' @description Loads information on gross domestic product at current prices, taxes, net of subsidies, on products at current prices and gross value added at current prices, total and by economic activity, and respective shares.
#'
#' @param dataset A dataset name ("pibmunic") with Municipal GDP information. You can also use SIDRA codes (See \url{https://sidra.ibge.gov.br/pesquisa/pib-munic/tabelas})
#' @inheritParams load_baci
#' @param geo_level A \code{string} that defines the geographic level of the data. Can be one of "country", "state" or "municipality".
#'
#' @return A \code{tibble}.
#'
#' @export
#'
#' @examples \dontrun{
#' # download treated municipal GDP data at the state level for 2010 to 2012
#' data <- load_pibmunic(
#'   raw_data = FALSE,
#'   geo_level = "state",
#'   time_period = 2010:2012
#' )
#' }
load_pibmunic <- function(dataset = "pibmunic", raw_data = FALSE,
                          geo_level, time_period,
                          language = "eng") {
  ##############################
  ## Binding Global Variables ##
  ##############################

  sidra_code <- available_time <- legal_amazon <- municipio_codigo <- ano <- NULL
  ano_codigo <- geo_id <- nivel_territorial <- nivel_territorial_codigo <- NULL
  unidade_de_medida <- unidade_de_medida_codigo <- valor <- variavel <- variavel_codigo <- NULL

  #############################
  ## Define Basic Parameters ##
  #############################

  param <- list()
  param$source <- "pibmunic"
  param$dataset <- dataset
  param$geo_level <- geo_level
  param$time_period <- time_period
  param$language <- language
  param$raw_data <- raw_data

  # check if dataset, geo_level, and time_period are supported

  check_params(param)

  if (!is.numeric(param$dataset)) {
    param$code <- datasets_link() %>%
      dplyr::filter(dataset == param$dataset) %>%
      dplyr::select(sidra_code) %>%
      unlist() %>%
      as.numeric()
  } else {
    param$code <- param$dataset
  }

  ##############
  ## Download ##
  ##############

  # We need to show year that is being downloaded as well
  # Heavy Datasets may take several minutes

  dat <- as.list(as.character(param$time_period)) %>%
    purrr::map(function(year_num) {
      # suppressMessages(
      sidra_download(
        sidra_code = param$code,
        year = year_num,
        geo_level = param$geo_level
      )
      # )
    }) %>%
    dplyr::bind_rows() %>%
    tibble::as_tibble()

  ## Return Raw Data

  if (param$raw_data) {
    return(dat)
  }

  ######################
  ## Data Enginnering ##
  ######################

  dat <- dat %>%
    janitor::clean_names() %>%
    dplyr::mutate_all(function(var) {
      stringi::stri_trans_general(str = var, id = "Latin-ASCII")
    })

  dat <- dat %>%
    dplyr::select(-c(nivel_territorial_codigo, nivel_territorial, ano_codigo)) %>%
    dplyr::mutate(valor = as.numeric(valor))

  ## Only Keep Valid Observations

  dat <- dat %>%
    dplyr::filter(!is.na(valor))

  #########################################
  ## Create Geographical Unit Identifier ##
  #########################################

  if (geo_level == "country") {
    dat$geo_id <- dat$brasil
    dat <- dplyr::select(dat, -"brasil_codigo", -"brasil")
  }

  if (geo_level == "state") {
    dat$geo_id <- dat$unidade_da_federacao_codigo
    dat <- dplyr::select(dat, -"unidade_da_federacao_codigo", -"unidade_da_federacao")
  }
  if (geo_level == "municipality") {
    dat$geo_id <- dat$municipio_codigo
    dat <- dplyr::select(dat, -"municipio", -"municipio_codigo")
  }

  ################################
  ## Harmonizing Variable Names ##
  ################################

  dat <- dat %>%
    dplyr::select(-unidade_de_medida, -unidade_de_medida_codigo)


  dat <- dat %>%
    dplyr::arrange(variavel_codigo, variavel) %>%
    tidyr::pivot_wider(
      id_cols = c(ano, geo_id),
      names_from = variavel:variavel_codigo,
      values_from = valor,
      names_sep = "_V",
      values_fn = sum,
      values_fill = NA
    ) %>%
    janitor::clean_names()

  if (language == "eng") {
    dat <- dat %>%
      dplyr::rename(year = ano)
  }

  ###############
  ## Labelling ##
  ###############

  labelled <- function(x, label) {
    Hmisc::label(x) <- label
    x
  }

  label_data_eng <- function(df, cols, dic) {
    label_value <- as.character(dic[dic$var_code == cols, "var_eng"])

    df <- df %>%
      dplyr::mutate_at(
        dplyr::vars(tidyr::matches(cols)),
        ~ labelled(., as.character(dic[dic$var_code == cols, "var_eng"]))
      )

    return(df)
  }

  label_data_pt <- function(df, cols, dic) {
    label_value <- as.character(dic[dic$var_code == cols, "var_pt"])

    df <- df %>%
      dplyr::mutate_at(
        dplyr::vars(tidyr::matches(cols)),
        ~ labelled(., as.character(dic[dic$var_code == cols, "var_pt"]))
      )

    return(df)
  }

  ## Load Dictionary

  dic <- load_dictionary(param$dataset)

  types <- as.character(dic$var_code)
  types <- types[types != "0"] ## Remove 0


  if (language == "eng") {
    for (i in seq_along(types)) {
      dat <- label_data_eng(dat, cols = types[i], dic = dic)
    }
  }

  if (language == "pt") {
    for (i in seq_along(types)) {
      dat <- label_data_pt(dat, cols = types[i], dic = dic)
    }
  }

  ##########################
  ## Returning Data Frame ##
  ##########################

  return(dat)
}