clean up

Author: sbidari

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: hubhelpr
 Title: Helper functions for CDC hubs maintainance
 Version: 0.0.0.9000
-Authors@R: 
-    person("Subekshya", "Bidari", 
+Authors@R:
+    person("Subekshya", "Bidari",
     email = "[email protected]",
     role = c("aut", "cre"))
 Description: This package provides functions to help with the maintainance of CDC hubs.

NAMESPACE

@@ -1,2 +1,3 @@
 # Generated by roxygen2: do not edit by hand
 
+export(update_hub_target_data)

R/update_hub_target_data.R

@@ -12,20 +12,27 @@ nssp_col_names <- list(
 ##' This function pulls, formats and save NHSN and NSSP data for use in the hub.
 ##'
 ##' @param base_hub_path Path to the base hub directory.
-##' @param disease Disease name (e.g., "covid", "rsv").
+##' @param disease Disease name ("covid" or "rsv").
 ##' @param excluded_locations Vector of location codes to exclude from the output.
-##' @param first_full_weekending_date First week-ending date to include.
+##' Default value is c("78", "69", "66", "60"), corresponding to "US Virgin Islands",
+##' "Guam", "American Samoa", and "Northern Mariana Islands".
+##' @param nhsn_first_weekending_date First week-ending date to include for
+##' the NHSN dataset. Default value is "2024-11-09".
 ##' @param legacy_file Logical. Whether to write legacy CSV output (default: FALSE).
 ##'
-##' @returns Writes target data files to target-data directory in the hub.
+##' @return Writes `time-series.parquet` and optionally legacy csv target data files
+##' to the target-data directory in the hub.
 ##' @export
 update_hub_target_data <- function(
   base_hub_path,
   disease,
-  first_full_weekending_date = lubridate::as_date("2024-11-09"),
-  excluded_locations = c("78", "0", "69", "66", "60"),
+  nhsn_first_weekending_date = lubridate::as_date("2024-11-09"),
+  excluded_locations = c("78", "69", "66", "60"),
   legacy_file = FALSE
 ) {
+  if (!disease %in% c("covid", "rsv")) {
+    stop("'disease' must be either 'covid' or 'rsv'")
+  }
   today <- lubridate::today()
   output_dirpath <- fs::path(base_hub_path, "target-data")
   fs::dir_create(output_dirpath)
@@ -36,14 +43,14 @@ update_hub_target_data <- function(
   nhsn_data <- forecasttools::pull_data_cdc_gov_dataset(
     dataset = "nhsn_hrd_prelim",
     columns = nhsn_col_name,
-    start_date = first_full_weekending_date
+    start_date = nhsn_first_weekending_date
   ) |>
     dplyr::rename(
       observation = nhsn_col_name,
       date = "weekendingdate"
     ) |>
     dplyr::mutate(
-      date = as.Date(.data$date),
+      date = lubridate::as_date(.data$date),
       observation = as.numeric(.data$observation),
       jurisdiction = stringr::str_replace(.data$jurisdiction, "USA", "US")
     ) |>
@@ -82,7 +89,7 @@ update_hub_target_data <- function(
     locations = "All"
   ) |>
     dplyr::mutate(
-      date = as.Date(.data$week_end),
+      date = lubridate::as_date(.data$week_end),
       observation = as.numeric(.data[[nssp_col_name]]) / 100,
       location = forecasttools::us_location_recode(
         .data$geography,

air.toml

@@ -6,4 +6,4 @@ line-ending = "auto"
 persistent-line-breaks = true
 exclude = []
 default-exclude = true
-skip = []
+skip = []

man/update_hub_target_data.Rd

@@ -1,32 +1,26 @@
-library(fs)
-library(dplyr)
-library(testthat)
-library(hubhelpr)
-library(forecasttools)
-
-run_update_hub_target_data_test <- function(
+update_hub_target_data_test <- function(
   disease,
   nhsn_col,
   nssp_col,
   nhsn_target,
   nssp_target
 ) {
   base_hub_path <- tempfile(paste0("base_hub_", disease, "_"))
-  dir_create(base_hub_path)
-  output_file <- path(base_hub_path, "target-data/time-series.parquet")
+  fs::dir_create(base_hub_path)
+  output_file <- fs::path(base_hub_path, "target-data/time-series.parquet")
   fs::dir_create(fs::path(base_hub_path, "target-data"))
 
   mock_target_ts <- tibble::tibble(
-    date = as.Date(character()),
+    date = lubridate::as_date(character()),
     observation = numeric(),
     location = character(),
-    as_of = as.Date(character()),
+    as_of = lubridate::as_date(character()),
     target = character()
   )
   forecasttools::write_tabular_file(mock_target_ts, output_file)
 
   mock_nhsn <- tibble::tibble(
-    weekendingdate = as.Date(c(
+    weekendingdate = lubridate::as_date(c(
       "2024-11-09",
       "2024-11-16",
       "2024-11-09",
@@ -37,7 +31,7 @@ run_update_hub_target_data_test <- function(
     totalconfrsvnewadm = c(3, 4, 3, 2)
   )
   mock_nssp <- tibble::tibble(
-    week_end = as.Date(c("2024-11-09", "2024-11-09")),
+    week_end = lubridate::as_date(c("2024-11-09", "2024-11-09")),
     county = c("All", "All"),
     geography = c("Alabama", "Alaska"),
     percent_visits_covid = c(1.0, 0.2),
@@ -52,17 +46,16 @@ run_update_hub_target_data_test <- function(
       } else if (dataset == "nssp_prop_ed_visits") {
         mock_nssp |> dplyr::select(week_end, geography, all_of(columns))
       } else {
-        stop("Unknown dataset")
+        stop("Dataset type not supported by the Hubs")
       }
     },
     ns = "forecasttools"
   )
 
   expect_silent(
-    update_hub_target_data(
+    hubhelpr::update_hub_target_data(
       base_hub_path = base_hub_path,
-      disease = disease,
-      first_full_weekending_date = as.Date("2024-11-09")
+      disease = disease
     )
   )
 
@@ -78,23 +71,30 @@ run_update_hub_target_data_test <- function(
       dplyr::select(date, observation, location, target) |>
       dplyr::arrange(date, location),
     tibble::tibble(
-      date = as.Date(c(mock_nhsn$weekendingdate, mock_nssp$week_end)),
+      date = lubridate::as_date(c(
+        mock_nhsn$weekendingdate,
+        mock_nssp$week_end
+      )),
       observation = c(
         as.numeric(mock_nhsn[[nhsn_col]]),
         as.numeric(mock_nssp[[nssp_col]]) / 100
       ),
       location = c(
-        us_location_recode(mock_nhsn$jurisdiction, "abbr", "code"),
-        us_location_recode(mock_nssp$geography, "name", "code")
+        forecasttools::us_location_recode(
+          mock_nhsn$jurisdiction,
+          "abbr",
+          "code"
+        ),
+        forecasttools::us_location_recode(mock_nssp$geography, "name", "code")
       ),
       target = c(rep(nhsn_target, 4), rep(nssp_target, 2))
     ) |>
       dplyr::arrange(date, location)
   )
 }
 
-test_that("update_hub_target_data returns exactly the mocked NHSN data for covid", {
-  run_update_hub_target_data_test(
+test_that("update_hub_target_data returns expected data for covid", {
+  update_hub_target_data_test(
     disease = "covid",
     nhsn_col = "totalconfc19newadm",
     nssp_col = "percent_visits_covid",
@@ -103,12 +103,22 @@ test_that("update_hub_target_data returns exactly the mocked NHSN data for covid
   )
 })
 
-test_that("update_hub_target_data returns exactly the mocked NHSN data for rsv", {
-  run_update_hub_target_data_test(
+test_that("update_hub_target_data returns expected data for rsv", {
+  update_hub_target_data_test(
     disease = "rsv",
     nhsn_col = "totalconfrsvnewadm",
     nssp_col = "percent_visits_rsv",
     nhsn_target = "wk inc rsv hosp",
     nssp_target = "wk inc rsv prop ed visits"
   )
 })
+
+test_that("update_hub_target_data errors for unsupported disease", {
+  expect_error(
+    update_hub_target_data(
+      base_hub_path = tempdir(),
+      disease = "flu"
+    ),
+    "'disease' must be either 'covid' or 'rsv'"
+  )
+})