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

write_committee_parquet <- function(report_list = tar_read(report_list)) {
  cover_path <- here::here("../data/cover")
  cover <- arrow::open_dataset(cover_path, partitioning = "sboe_id")

  latest_report_by_committee <- report_list |> slice_max(end_date, by = "sboe_id")

  committees <-
    cover |>
    semi_join(latest_report_by_committee, by = "report_id") |>
    collect() |>
    select(sboe_id:committee_type, fund_type, fund_name)

  arrow::write_parquet(committees, here::here("data/committees/part-0.parquet"))
}