nc-campaign-finance/_targets.R

# Created by use_targets().
# Follow the comments below to fill in this target script.
# Then follow the manual to check and run the pipeline:
#   https://books.ropensci.org/targets/walkthrough.html#inspect-the-pipeline

# Load packages required to define the pipeline:
library(targets)

# Set target options:
tar_option_set(
  packages = strsplit(desc::desc_get_field("Depends"), ", ")[[1]],
  # For distributed computing in tar_make(), supply a {crew} controller
  # as discussed at https://books.ropensci.org/targets/crew.html.
  controller = crew::crew_controller_local(workers = 2)
)

# Run the R scripts in the R/ folder with your custom functions:
tar_source()

# Replace the target list below with your own:
list(
  tar_target(year, 2016:2023),
  tar_target(report, c("MY", "YE", "Q1", "Q2", "Q3", "Q4")),
  tar_target(
    doc_list,
    get_report_by_year_scrape(year, report),
    pattern = cross(year, report)
  ),
  tar_target(
    committees,
    doc_list |>
      # this is the list of committees and years reported
      # if `n_*` changes, we have to go re-read that year
      group_by(committee_name, sboe_id, year) |>
      summarize(
        n_reports = n(),
        n_amended = sum(amended),
        n_docs = sum(!is.na(report_id)),
      ) |>
      arrange(sboe_id)
  ),
  tar_target(
    report_list,
    doc_list |>
      filter(!is.na(report_id)) |>
      filter(committee_name == "JOHN BELL COMM" | report_id == 188139)
  ),
  tar_target(report_list_report_id, report_list$report_id),
  tar_target(report_list_sboe_id, report_list$sboe_id),
  tar_target(
    receipts,
    get_report_section(report_list_report_id, "receipts", report_list_sboe_id),
    pattern = map(report_list_report_id, report_list_sboe_id),
    format = "parquet"
  )
)