R/get_language_data.R

In ttimbers/canlang: Canadian language data

#' Get census language data from the Canadian level in a tidy format.
#'
#' @param dataset character vector of the data set name (e.g., "CA16")
#' @param query_vector character vector of vector from which to retrieve the data (e.g., "v_CA16_815")
#' @param region_level character vector of which region to aggregate by. Default is "C" for Canada level.
#'
#' @return a tidy data frame of language data at the Canadian level
#' @export
#'
#' @examples
#' \dontrun{
#' library(canlang)
#' options(cancensus.api_key = "your_api_key")
#' get_language_data("CA16", "v_CA16_815")
#' }
get_language_data <- function(dataset, query_vector, region_level = "C") {
    langs <- cancensus::list_census_vectors(dataset) %>%
        dplyr::filter(vector == query_vector)

    langs_children <- langs %>%
        cancensus::child_census_vectors(TRUE)

    langs_vectors <- dplyr::bind_rows(langs, langs_children) %>%
        dplyr::pull(vector)

    region <- cancensus::list_census_regions(dataset, use_cache = FALSE) %>%
        dplyr::filter(level == region_level) %>%
        cancensus::as_census_region_list()

    if (region_level == "C") {
        language <- cancensus::get_census(dataset = dataset,
                                          regions = region,
                                          vectors = langs_vectors,
                                          level = region_level) %>%
            dplyr::select(-GeoUID, -`Region Name`, -Households, -Type, -`Area (sq km)`, -Population, -Dwellings) %>%
            tidyr::pivot_longer(everything(), names_to = "language", values_to = "count") %>%
            tidyr::separate(language, into = c("junk", "language"), sep = ": ") %>%
            dplyr::arrange(language) %>%
            dplyr::select(-junk) %>%
            dplyr::filter(!language == "Aboriginal languages", !language == "Official languages", !language == "Non-official languages")
    } else {
        language <- cancensus::get_census(dataset = dataset,
                                          regions = region,
                                          vectors = langs_vectors,
                                          level = region_level) %>%
            dplyr::select(-GeoUID, -Households, -Type, -`Area (sq km)`, -Population, -Dwellings, -PR_UID, -C_UID) %>%
            tidyr::pivot_longer(-`Region Name`, names_to = "language", values_to = "count") %>%
            tidyr::separate(language, into = c("junk", "language"), sep = ": ") %>%
            dplyr::arrange(language) %>%
            dplyr::select(-junk) %>%
            dplyr::filter(!language == "Aboriginal languages", !language == "Official languages", !language == "Non-official languages")
    }
}