github-repo-stats.yml

name: github-repo-stats

on:
    schedule:
        # Run this once per day, towards the end of the day for keeping the most
        # recent data point most meaningful (hours are interpreted in UTC).
        - cron: "0 23 * * *"
    workflow_dispatch: # Allow for running this manually.

permissions:
    contents: read

jobs:
    j1:
        name: github-repo-stats
        runs-on: ubuntu-latest
        steps:
            - name: generate app token
              id: app-token
              uses: actions/create-github-app-token@bcd2ba49218906704ab6c1aa796996da409d3eb1 # v3.2.0
              with:
                  app-id: 3189942
                  private-key: ${{ secrets.POWER_PLATFORM_SKILLS_APP_PRIVATE_KEY }}
                  permission-administration: read
                  permission-contents: write
                  permission-metadata: read

            - name: verify traffic API access
              env:
                  GH_TOKEN: ${{ steps.app-token.outputs.token }}
              run: |
                  set -euo pipefail

                  response_headers="$(mktemp)"
                  trap 'rm -f "${response_headers}"' EXIT
                  status="$(
                      curl -sS -o /dev/null -D "${response_headers}" -w "%{http_code}" \
                          -H "Authorization: Bearer ${GH_TOKEN}" \
                          -H "Accept: application/vnd.github+json" \
                          -H "X-GitHub-Api-Version: 2022-11-28" \
                          "https://api.github.com/repos/${{ github.repository }}/traffic/popular/referrers"
                  )"

                  if [ "${status}" != "200" ] && [ "${status}" != "204" ]; then
                      echo "GitHub traffic API preflight failed with HTTP ${status}."
                      echo "The github-repo-stats action needs an installation token with repository Administration: read access."
                      awk 'BEGIN { IGNORECASE = 1 } /^x-accepted-github-permissions:|^x-oauth-scopes:|^x-github-api-version-selected:|^x-ratelimit-/ { print }' "${response_headers}"
                      exit 1
                  fi

            - name: checkout repository
              uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0

            - name: checkout github-repo-stats action
              uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0
              with:
                  repository: jgehrcke/github-repo-stats
                  ref: RELEASE
                  path: .github/actions/github-repo-stats

            - name: patch github-repo-stats stargazer fetch
              shell: bash
              run: |
                  python3 <<'PY'
                  from pathlib import Path

                  path = Path(".github/actions/github-repo-stats/fetch.py")
                  text = path.read_text()
                  import_line = "from github import Github, Repository  # type: ignore"
                  if import_line not in text:
                      raise RuntimeError("Expected PyGithub import line was not found")
                  text = text.replace(
                      import_line,
                      "from github import Github, Repository, GithubException  # type: ignore",
                  )

                  old_stargazer_fetch = "\n".join(
                      [
                          "    # TODO for addressing the 10ks challenge: save state to disk, and refresh",
                          "    # using reverse order iteration. See for repo in user.get_repos().reversed",
                          "    for count, gazer in enumerate(repo.get_stargazers_with_dates(), 1):",
                          "        # Store `PullRequest` object with integer key in dictionary.",
                          "        gazers.append(gazer)",
                          "        if count % 200 == 0:",
                          '            log.info("%s gazers fetched", count)',
                      ]
                  ) + "\n"
                  if old_stargazer_fetch not in text:
                      raise RuntimeError("Expected REST stargazer fetch block was not found")
                  new_stargazer_fetch = "\n".join(
                      [
                          "    # GitHub restricted the REST stargazers listing in July 2026 to admins and",
                          "    # collaborators. Installation tokens with repository Administration access",
                          "    # can still read the same timestamp data through GraphQL, so keep the",
                          "    # upstream REST path for normal tokens and fall back only for that specific",
                          "    # restriction.",
                          "    # See:",
                          "    # - https://docs.github.com/en/rest/activity/starring#list-stargazers",
                          "    # - https://github.blog/changelog/2026-06-30-upcoming-access-restrictions-to-public-api-endpoints-and-ui-views/",
                          "    try:",
                          "        # TODO for addressing the 10ks challenge: save state to disk, and refresh",
                          "        # using reverse order iteration. See for repo in user.get_repos().reversed",
                          "        for count, gazer in enumerate(repo.get_stargazers_with_dates(), 1):",
                          "            gazers.append(gazer)",
                          "            if count % 200 == 0:",
                          '                log.info("%s gazers fetched", count)',
                          "    except GithubException as exc:",
                          '        if exc.status != 403 or "permission to view the stargazers" not in str(exc):',
                          "            raise",
                          '        log.warning("REST stargazer endpoint denied access; retry via GraphQL")',
                          "        gazers = get_stargazers_with_dates_graphql(repo.full_name)",
                      ]
                  ) + "\n"
                  text = text.replace(old_stargazer_fetch, new_stargazer_fetch)

                  graphql_helper = r'''
                  def get_stargazers_with_dates_graphql(repo_full_name):
                      owner, name = repo_full_name.split("/", 1)
                      query = """
                      query($owner: String!, $name: String!, $cursor: String) {
                        repository(owner: $owner, name: $name) {
                          stargazers(
                            first: 100
                            after: $cursor
                            orderBy: {field: STARRED_AT, direction: ASC}
                          ) {
                            pageInfo {
                              hasNextPage
                              endCursor
                            }
                            edges {
                              starredAt
                            }
                          }
                        }
                      }
                      """

                      class StargazerEvent:
                          def __init__(self, starred_at):
                              # Match PyGithub's REST shape here: get_stargazers_with_dates()
                              # exposes starred_at as a UTC timestamp without tzinfo. The caller
                              # localizes it to UTC before building the pandas index.
                              self.starred_at = starred_at

                      gazers = []
                      cursor = None
                      while True:
                          response = requests.post(
                              "https://api.github.com/graphql",
                              headers={
                                  "Authorization": f"Bearer {os.environ['GHRS_GITHUB_API_TOKEN'].strip()}",
                                  "Content-Type": "application/json",
                              },
                              json={"query": query, "variables": {"owner": owner, "name": name, "cursor": cursor}},
                              timeout=60,
                          )
                          response.raise_for_status()
                          payload = response.json()
                          if payload.get("errors"):
                              raise RuntimeError(f"GitHub GraphQL stargazer fetch failed: {payload['errors']}")

                          stargazers = payload["data"]["repository"]["stargazers"]
                          for edge in stargazers["edges"]:
                              starred_at = datetime.fromisoformat(
                                  edge["starredAt"].replace("Z", "+00:00")
                              ).replace(tzinfo=None)
                              gazers.append(StargazerEvent(starred_at))
                              if len(gazers) % 200 == 0:
                                  log.info("%s gazers fetched", len(gazers))

                          if not stargazers["pageInfo"]["hasNextPage"]:
                              return gazers
                          cursor = stargazers["pageInfo"]["endCursor"]


                  def handle_rate_limit_error(exc):
                  '''
                  helper_insert_point = "\ndef handle_rate_limit_error(exc):\n"
                  if helper_insert_point not in text:
                      raise RuntimeError("Expected helper insertion point was not found")
                  text = text.replace(helper_insert_point, "\n" + graphql_helper.strip() + "\n")
                  path.write_text(text)
                  PY

            - name: run-ghrs
              # Run the checked-out RELEASE action after applying the stargazer fallback patch above.
              uses: ./.github/actions/github-repo-stats
              with:
                  repository: microsoft/power-platform-skills
                  ghtoken: ${{ steps.app-token.outputs.token }}
                  databranch: github-repo-stats
                  ghpagesprefix: https://microsoft.github.io/power-platform-skills