R/get_investigators.R

In jesse-smith/covidassign: COVID-19 Case Assignment for SCHD

#' Join Investigators Listed in Investigation Schedule & Case Assignment Project
#'
#' `get_investigators()` joins the investigators scheduled to work on
#' `date` with the list of investigators from the \strong{Case Assignment}
#' REDcap project. By default, it performs an inner join, since only
#' investigators in both lists are available for case assignment.
#'
#' While the default behavior is an inner join, `get_investigators()` can
#' also perform anti-joins with either the scheduled or REDcap investigator list
#' as the primary table. This is useful for debugging and checking that all
#' scheduled investigators are able to have cases assigned.
#'
#' @inheritParams sched_investigators
#'
#' @inheritParams download_redcap_investigators
#'
#' @param type The type or group of investigators to return
#'
#' @param quiet Should update messages be suppressed?
#'
#' @return A `tibble` with one row per investigator and an `investigator`
#'   column. If `type = "inner"`, an `id` column is also included.
#'
#' @export
get_investigators <- function(
  date = Sys.Date(),
  type = c("all", "general", "school"),
  scheduled_only = TRUE,
  api_token = Sys.getenv("redcap_NCA_token"),
  quiet = TRUE
) {

  type <- rlang::arg_match(type)[[1L]]

  if (!quiet) rlang::inform("Loading investigators...")

  inv_scheduled <- purrr::when(
    suppressMessages(
      sched_investigators(date = date, scheduled_only = scheduled_only)
    ),
    type == "general" ~ dplyr::filter(
      ., stringr::str_detect(.data[["team"]], "(?i)^[a-z]$")
    ),
    type == "school" ~ dplyr::filter(
      ., stringr::str_detect(.data[["team"]], "(?i)^school$")
    ),
    ~ .
  )

  inv_redcap <- suppressMessages(
    download_redcap_investigators(api_token = api_token)
  )

  if (!quiet) rlang::inform("Returning investigators in both lists...")

  dplyr::inner_join(
    inv_scheduled,
    inv_redcap,
    by = "investigator"
  ) %>%
    dplyr::select("id", "team", "investigator") %>%
    dplyr::arrange(.data[["investigator"]], .data[["id"]]) %>%
    dplyr::distinct(.data[["investigator"]], .keep_all = TRUE)
}