nc-campaign-finance/process/R/process_report_list.R

process_report_list <- function(report_list_raw, report_amended_score) {
  # select the correct report_id to use for each report group

  report_keep <-
    report_amended_score |>
    group_by(sboe_id, year, doc_name) |>
    slice_max(amended_score) |>
    # follow up to be certain there's only one report per group
    slice_max(report_id)

  report_keep_count <- report_keep |> count()
  if (any(report_keep_count$n > 1)) {
    stop("Have not successfully selected a single report per committee and report group.")
  }

  report_keep <- report_keep |> ungroup() |> select(report_id, matches("received|date"))

  report_list_raw |>
    select(-matches("received|date")) |>
    right_join(report_keep, by = "report_id") |>
    select(-committee_name, -doc_type, -tar_group, image_id = img_link) |>
    mutate(
      across(doc_name, as_report_factor),
      image_id = sub(".+?(\\d+)$", "\\1", image_id)
    )
}