generate pull request lists

gen_pull_request_csv.rb

# HOW TO USE:
#  ruby gen_pull_request_csv.rb ownername/repo 2024-01-01

require 'json'
require 'csv'

@token = ENV.fetch('GH_TOKEN')

def run_api(path)
  call = <<~CURL
    curl -L \
    -H "Accept: application/vnd.github+json" \
    -H "Authorization: Bearer #{@token}" \
    -H "X-GitHub-Api-Version: 2022-11-28" \
    https://api.github.com#{path} 2> /dev/null
  CURL
  JSON.parse(`#{call}`)
end

def run_graphql(query, variables)
  data = {
    query: query.gsub("\n", '').gsub("\s\+", ' '),
    variables: variables
  }

  call = <<~CURL
    curl -L \
    -H "Accept: application/vnd.github+json" \
    -H "Authorization: Bearer #{@token}" \
    -H "X-GitHub-Api-Version: 2022-11-28" \
    -X POST \
    -d '#{data.to_json}' \
    https://api.github.com/graphql 2> /dev/null
  CURL
  JSON.parse(`#{call}`)
end

def load_all_pull_requests(repo:, created_after:)
  all_items = []
  end_cursor = nil
  c = 1
  search_query = "is:pr is:merged repo:#{repo} created:>=#{created_after}"

  loop do
    query = <<~GQL
      query loadPullRequest($query: String!, $after: String) {
        search(type: ISSUE, query: $query, first: 100, after: $after) {
          nodes {
            __typename
            ... on PullRequest {
              id
              title
              state
              author {
                login
              }
              closedAt
              baseRefName
              headRefName
            }
          }
          pageInfo {
            endCursor
            hasNextPage
          }
        }
      }
    GQL
    sleep 1
    puts "load - attempts #{c}"
    result = run_graphql(query, {
      query: search_query,
      after: end_cursor,
    })
    items = result.dig('data', 'search', 'nodes')
    all_items += items

    page_info = result.dig('data', 'search', 'pageInfo')

    return all_items unless page_info.dig('hasNextPage')
    end_cursor = page_info.dig('endCursor')
    c += 1
  end
end

puts "load pull request with #{ARGV[0..1].join(",")}"
pull_requests = load_all_pull_requests(repo: ARGV[0], created_after: ARGV[1])
  .reject do
    _1.dig('headRefName') == 'release-candidate' \
    || _1.dig('author', 'login') == 'dependabot' \
    || _1.dig('baseRefName') != 'release-candidate'
  end
  .map do |item|
    {
      title: item.dig('title').sub(/^\[review\]\s+/, ''),
      state: item.dig('state'),
      author_login: item.dig('author', 'login'),
      closed_at: item.dig('closedAt')
    }
  end
  .sort_by { _1[:closed_at] }

CSV.open("./gh_pull_requests.csv", "wb") do |csv|
  csv << pull_requests.first.keys

  pull_requests.each do |i|
    csv << i.values
  end
end
puts "out: ./gh_pull_requests.csv"