R/trud_items.R

In trud: Query the 'NHS TRUD API'

#' Get available NHS TRUD items
#'
#' Scrapes [this
#' page](https://isd.digital.nhs.uk/trud/users/guest/filters/0/categories/1)
#' from the NHS TRUD website for all available items.
#'
#' @return A tibble, with columns `item_number` and `item_name`.
#' @export
#'
#' @examples
#' \dontrun{
#'  trud_items()
#' }
trud_items <- function() {
  # Read web page
  page <-
    rvest::read_html("https://isd.digital.nhs.uk/trud/users/guest/filters/0/categories/1")

  # Find all hyperlinks
  items <- page %>%
    rvest::html_elements(css = ".content , a")

  # Extract text from each element
  items_text <- items %>%  rvest::html_text()

  # Extract the hyperlinks from each element using html_attr()
  items_link <- items %>%  rvest::html_attr("href")

  # Extract item numbers from hyperlinks
  items_number <- items_link %>%
    stringr::str_extract("/items/\\d+") %>%
    stringr::str_remove("/items/")

  # Create a dataframe with the item names and their links/numbers
  df <- tibble::tibble(
    item_name = items_text,
    item_link = items_link,
    item_number = as.integer(items_number)
  )

  # Filter for items
  df %>%
    dplyr::filter(stringr::str_detect(.data[["item_link"]], "trud/users/guest/filters/0/categories/1/items")) %>%
    dplyr::filter(!.data[["item_name"]] %in% c("Releases", "Licences", "Future releases")) %>%
    dplyr::select(tidyselect::all_of(c("item_number", "item_name")))
}