Get PR stats table from a repo

pr_stats.R

get_prs <- function(repo, owner, base = "dev", state = "closed") {
  query <- list(
    base = base,
    state = state
  )
  query <- paste(purrr::imap_chr(query, \(x, y) paste0(y, "=", x)), collapse = "&")
  prs <- gh::gh(glue::glue("GET /repos/{owner}/{repo}/pulls?{query}"), .limit = Inf)
  prs |>
    purrr::map(\(x) {
      pull_number <- x$number
      files <- gh::gh(glue::glue("/repos/{owner}/{repo}/pulls/{pull_number}/files"), .limit = Inf)
      changes <- files |>
        purrr::map(\(f) {
          tibble::tibble(
            additions = f$additions, 
            deletions = f$deletions,
            changes = f$changes
          )
        }) |>
        dplyr::bind_rows() |>
        dplyr::summarise(dplyr::across(everything(), sum))
      tibble::tibble(
        title = x$title,
        author = x$user$login,
        created_at = x$created_at,
        updated_at = x$updated_at,
        closed_at = x$closed_at,
        merged_at = x$merged_at,
        n_files_changed = length(files),
        changes = changes,
        url = x$url
      )
    }) |>
    dplyr::bind_rows() |>
    tidyr::unnest(changes) |>
    dplyr::arrange(dplyr::desc(n_files_changed))
}

add_stats <- function(data) {
  data |>
    dplyr::mutate(
      merged_at = lubridate::as_datetime(merged_at),
      created_at = lubridate::as_datetime(created_at),
      lead_time_to_change_minutes = lubridate::as.duration(merged_at - created_at) |>
        lubridate::time_length("minutes"),
      lead_time_to_change_days = lubridate::as.duration(merged_at - created_at) |>
        lubridate::time_length("days"),
      avg_changes_per_file = changes / n_files_changed
    )
}

fastests_prs <- function(stats) {
  stats |>
    dplyr::filter(lead_time_to_change_minutes > 0) |>
    dplyr::arrange(lead_time_to_change_minutes) |>
    dplyr::select(title, author, lead_time_to_change_minutes) |>
    dplyr::slice_head(n = 5)
}

lead_time_to_change_days <- function(stats) {
  stats |>
    ggplot2::ggplot(ggplot2::aes(x = lead_time_to_change_days)) +
    ggplot2::geom_histogram(binwidth = 1, color = "black", boundary = 0) +
    ggplot2::scale_x_continuous(breaks = seq(0, 30, 1)) +
    ggplot2::labs(title = "Lead Time to Change", x = "Lead Time to Change (days)", y = "Number of Pull Requests") +
    ggplot2::theme_minimal()
}

n_files_changed <- function(stats) {
  stats |>
    ggplot2::ggplot(ggplot2::aes(x = n_files_changed)) +
    ggplot2::geom_histogram(binwidth = 1, color = "black", boundary = 0) +
    ggplot2::scale_x_continuous(breaks = seq(0, 1000, 10)) +
    ggplot2::labs(title = "Histogram of Number of Files Changed", x = "Number Files Changed", y = "Frequency") +
    ggplot2::theme_minimal()
}