R/get_hub_locations.R

In reichlab/covidHubUtils: Utility functions for the COVID-19 forecast hub

#' Create data/hub_locations.rda for loading data from `zoltar` and `covidData`
#'
#' @importFrom readr read_csv
#' @return data frame with columns `fips`, `location_name`, `population`, `geo_type`,
#' `geo_value`, `abbreviation`
#'
get_hub_locations <- function() {
  hub_locations <- readr::read_csv("https://raw.githubusercontent.com/reichlab/covid19-forecast-hub/master/data-locations/locations.csv") %>%
    # Add columns used for reading files in covidcast package
    # US has "state" as geo_type
    dplyr::mutate(
      geo_type = ifelse(!is.na(abbreviation),
        "state",
        "county"
      ),
      geo_value = ifelse(!is.na(abbreviation),
        tolower(abbreviation),
        location
      )
    ) %>%
    # Add a column for states/'US' that each location belongs to
    dplyr::mutate(state_abbreviation = substr(location, start = 1, stop = 2)) %>%
    dplyr::mutate(state_abbreviation = unlist(
      lapply(state_abbreviation, function(x) {
        ifelse(x %in% location, abbreviation[location == x], "NA")
      })
    )) %>%
    dplyr::rename(
      abbreviation = state_abbreviation,
      state_abbreviation = abbreviation,
      fips = location
    ) %>%
    dplyr::select(-c("state_abbreviation"))

  # Rename US to United States in location_name
  hub_locations[hub_locations$fips == "US", ]$location_name <- "United States"

  # Append state abbreviation to county names
  hub_locations <- hub_locations %>%
    dplyr::mutate(
      full_location_name =
        ifelse(geo_type == "county",
          paste(location_name, abbreviation, sep = ", "),
          location_name
        )
    )

  save(hub_locations, file = "data/hub_locations.rda")
}