Ambassador maintenance (duplicates + decisions + cleanup) #2
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: "Ambassador maintenance (duplicates + decisions + cleanup)" | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| private_repo: | |
| description: "Private repo containing the CSVs (owner/name)" | |
| required: true | |
| default: pytorch-fdn/ambassador-program-management | |
| type: string | |
| private_ref: | |
| description: "Branch/tag/sha of the private repo" | |
| required: true | |
| default: main | |
| type: string | |
| duplicates_csv_path: | |
| description: "Path to duplicates CSV" | |
| required: true | |
| default: .github/ISSUE_TEMPLATE/duplicates.csv | |
| type: string | |
| decision_csv_path: | |
| description: "Path to decision CSV (has Submission ID, Decision, Candidate)" | |
| required: true | |
| default: .github/ISSUE_TEMPLATE/dmsap.csv | |
| type: string | |
| dry_run: | |
| description: "Preview without writing comments/labels" | |
| required: true | |
| default: false | |
| type: boolean | |
| permissions: | |
| contents: read | |
| issues: write | |
| concurrency: | |
| group: ambassador-maintenance | |
| cancel-in-progress: true | |
| jobs: | |
| apply: | |
| runs-on: ubuntu-latest | |
| steps: | |
| - name: Checkout PUBLIC repo | |
| uses: actions/checkout@v4 | |
| - name: Checkout PRIVATE repo | |
| uses: actions/checkout@v4 | |
| with: | |
| repository: ${{ inputs.private_repo }} | |
| ref: ${{ inputs.private_ref }} | |
| token: ${{ secrets.PRIVATE_REPO_TOKEN }} | |
| path: private-src | |
| - name: Set up Python | |
| uses: actions/setup-python@v5 | |
| with: | |
| python-version: "3.11" | |
| - name: Install dependencies | |
| run: pip install requests | |
| - name: Process duplicates, post decisions by row, clean pending-review | |
| env: | |
| GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} | |
| REPO_FULL: ${{ github.repository }} | |
| DUP_CSV_ABS: ${{ github.workspace }}/private-src/${{ inputs.duplicates_csv_path }} | |
| DEC_CSV_ABS: ${{ github.workspace }}/private-src/${{ inputs.decision_csv_path }} | |
| DRY_RUN: ${{ inputs.dry_run }} | |
| run: | | |
| python - <<'PY' | |
| import os, csv, re, sys, urllib.parse | |
| import requests | |
| from requests.adapters import HTTPAdapter, Retry | |
| token = os.getenv("GITHUB_TOKEN") | |
| repo_full = os.getenv("REPO_FULL") | |
| dup_csv = os.getenv("DUP_CSV_ABS") | |
| dec_csv = os.getenv("DEC_CSV_ABS") | |
| dry_run = (os.getenv("DRY_RUN","false").lower() == "true") | |
| if not token or not repo_full: | |
| print("Missing GITHUB_TOKEN or REPO_FULL", file=sys.stderr); sys.exit(1) | |
| owner, repo = repo_full.split("/", 1) | |
| # resilient session | |
| s = requests.Session() | |
| retries = Retry(total=8, backoff_factor=0.7, | |
| status_forcelist=[429,500,502,503,504], | |
| allowed_methods=["GET","POST","PATCH","DELETE","HEAD","OPTIONS"]) | |
| s.mount("https://", HTTPAdapter(max_retries=retries)) | |
| s.headers.update({"Authorization": f"Bearer {token}", | |
| "Accept": "application/vnd.github+json", | |
| "X-GitHub-Api-Version": "2022-11-28"}) | |
| def norm_key(x): return re.sub(r"[^a-z0-9]+","",(x or "").strip().lower()) | |
| def ensure_label(name, color): | |
| enc = urllib.parse.quote(name, safe="") | |
| r = s.get(f"https://api.github.com/repos/{owner}/{repo}/labels/{enc}") | |
| if r.status_code==200: return | |
| if r.status_code==404: | |
| if dry_run: print(f"DRY-RUN: would create label '{name}' #{color}"); return | |
| r = s.post(f"https://api.github.com/repos/{owner}/{repo}/labels", | |
| json={"name":name,"color":color}) | |
| if r.status_code not in (200,201): | |
| print(f"Warn: create label {name} failed: {r.status_code} {r.text}") | |
| else: | |
| print(f"Warn: get label {name} failed: {r.status_code} {r.text}") | |
| def get_issue(num): return s.get(f"https://api.github.com/repos/{owner}/{repo}/issues/{num}") | |
| def get_issue_labels(num): | |
| r = s.get(f"https://api.github.com/repos/{owner}/{repo}/issues/{num}/labels?per_page=100") | |
| if r.status_code==200: return {x["name"] for x in r.json()} | |
| print(f"Warn: get labels #{num} failed: {r.status_code} {r.text}"); return set() | |
| def add_labels(num, labels): | |
| if not labels: return | |
| if dry_run: print(f"DRY-RUN: would add labels to #{num}: {sorted(labels)}"); return | |
| r = s.post(f"https://api.github.com/repos/{owner}/{repo}/issues/{num}/labels", | |
| json={"labels": list(labels)}) | |
| if r.status_code not in (200,201): | |
| print(f"Warn: add labels #{num} failed: {r.status_code} {r.text}") | |
| def remove_label(num, label): | |
| if dry_run: print(f"DRY-RUN: would remove label '{label}' from #{num}"); return | |
| enc = urllib.parse.quote(label, safe="") | |
| r = s.delete(f"https://api.github.com/repos/{owner}/{repo}/issues/{num}/labels/{enc}") | |
| if r.status_code not in (200,204): | |
| print(f"Warn: remove label #{num} failed: {r.status_code} {r.text}") | |
| def add_comment_once(num, body, marker): | |
| # avoid duplicate comments using hidden marker | |
| url = f"https://api.github.com/repos/{owner}/{repo}/issues/{num}/comments?per_page=100" | |
| while url: | |
| r = s.get(url) | |
| if r.status_code!=200: print(f"Warn: list comments #{num} failed: {r.status_code} {r.text}"); break | |
| if any(marker in (c.get("body") or "") for c in r.json()): | |
| print(f"Skip duplicate comment on #{num}"); return | |
| url = r.links.get("next",{}).get("url") | |
| if dry_run: print(f"DRY-RUN: would comment on #{num}:\n{body}\n---"); return | |
| r = s.post(f"https://api.github.com/repos/{owner}/{repo}/issues/{num}/comments", json={"body": body}) | |
| if r.status_code not in (200,201): print(f"Warn: comment #{num} failed: {r.status_code} {r.text}") | |
| def list_issues_with_label(label): | |
| url = f"https://api.github.com/repos/{owner}/{repo}/issues?state=all&per_page=100&labels={urllib.parse.quote(label, safe='')}" | |
| while url: | |
| r = s.get(url) | |
| if r.status_code!=200: print(f"Warn: list issues for '{label}' failed: {r.status_code} {r.text}"); return | |
| for it in r.json(): | |
| if "pull_request" in it: continue | |
| yield it | |
| url = r.links.get("next",{}).get("url") | |
| # labels & messages | |
| ACCEPT_LABEL, REJECT_LABEL, DUP_LABEL, PENDING_LABEL = "Accepted","Rejected","Duplicate","pending-review" | |
| ensure_label(ACCEPT_LABEL, "2ea44f") | |
| ensure_label(REJECT_LABEL, "d73a4a") | |
| ensure_label(DUP_LABEL, "cfd3d7") | |
| ACCEPT_MARKER = "<!-- ambassador-decision:accepted-2025 -->" | |
| REJECT_MARKER = "<!-- ambassador-decision:rejected-2025 -->" | |
| ACCEPT_MSG = ( | |
| "Hello {name},\n\n" | |
| "Congratulations! 🎉 We're excited to inform you that you have been selected as a 2025 PyTorch Ambassador. " | |
| "Your contributions, impact, and commitment to the PyTorch community stood out during the review process.\n\n" | |
| "You should also have received an invitation to join our private program management repository, " | |
| "where onboarding details and next steps will be shared. If you do not see the invite, please comment here and we will assist you.\n\n" | |
| "Best regards,\n" | |
| "The PyTorch PMO\n\n" + ACCEPT_MARKER | |
| ) | |
| REJECT_MSG = ( | |
| "Hello {name},\n\n" | |
| "Thank you for applying to the 2025 PyTorch Ambassador Program and for the valuable contributions you make in the PyTorch community. " | |
| "After careful consideration, we regret to inform you that your application was not selected in this cycle.\n\n" | |
| "We encourage you to apply again in the future — the next application cycle will open in early 2026. " | |
| "Please keep an eye on our website for updates on the timeline: https://pytorch.org/programs/ambassadors/\n\n" | |
| "We truly appreciate the time, effort, and passion you bring to the PyTorch community, and we hope to see your application again in the future.\n\n" | |
| "If you have any questions in the meantime, please feel free to reach out to us at [email protected].\n\n" | |
| "Best regards,\n" | |
| "The PyTorch PMO\n\n" + REJECT_MARKER | |
| ) | |
| # --- Part 1: Label duplicates from duplicates.csv (no closing) --- | |
| dup_counts={"total":0,"labeled":0,"missing":0,"skipped":0} | |
| try: | |
| with open(dup_csv, newline="", encoding="utf-8") as f: | |
| rdr = csv.DictReader(f) | |
| hdr = {norm_key(h): h for h in (rdr.fieldnames or [])} | |
| id_key = hdr.get("issue") or hdr.get("submissionid") or hdr.get("issuenumber") or hdr.get("id") | |
| if not id_key: | |
| print("duplicates.csv must have an issue id column (e.g., 'Issue' or 'Submission ID').", file=sys.stderr); sys.exit(1) | |
| for idx,row in enumerate(rdr, start=2): | |
| dup_counts["total"] += 1 | |
| raw = (row.get(id_key) or "").strip() | |
| if not raw: dup_counts["skipped"] += 1; continue | |
| try: num = int(float(raw)) | |
| except: print(f"Row {idx}: invalid ID '{raw}'; skipping."); dup_counts["skipped"] += 1; continue | |
| r = get_issue(num) | |
| if r.status_code==404: print(f"Issue #{num} not found; skipping."); dup_counts["missing"] += 1; continue | |
| if r.status_code!=200: print(f"Warn: GET issue #{num} failed: {r.status_code} {r.text}"); dup_counts["skipped"] += 1; continue | |
| existing = get_issue_labels(num) | |
| to_add = {"Duplicate"} - existing | |
| add_labels(num, to_add) | |
| if to_add: dup_counts["labeled"] += 1 | |
| except FileNotFoundError: | |
| print(f"duplicates.csv not found at: {dup_csv}", file=sys.stderr); sys.exit(1) | |
| # --- Part 2: Iterate decision.csv rows and post per-row messages with Candidate name --- | |
| # Expects columns: Submission ID, Decision, Candidate | |
| dec_counts={"rows":0,"accept_commented":0,"reject_commented":0,"skipped":0} | |
| try: | |
| with open(dec_csv, newline="", encoding="utf-8") as f: | |
| rdr = csv.DictReader(f) | |
| hdr = {norm_key(h): h for h in (rdr.fieldnames or [])} | |
| sid_key = hdr.get("submissionid") or hdr.get("issue") or hdr.get("issuenumber") or hdr.get("id") | |
| decision_key = hdr.get("decision") | |
| name_key = hdr.get("candidate") | |
| if not sid_key or not decision_key or not name_key: | |
| print("decision.csv must contain 'Submission ID', 'Decision', and 'Candidate' columns.", file=sys.stderr); sys.exit(1) | |
| for idx,row in enumerate(rdr, start=2): | |
| dec_counts["rows"] += 1 | |
| raw_id = (row.get(sid_key) or "").strip() | |
| decision_raw = (row.get(decision_key) or "").strip().lower() | |
| candidate = (row.get(name_key) or "").strip() or "there" | |
| if not raw_id: | |
| dec_counts["skipped"] += 1; continue | |
| try: | |
| num = int(float(raw_id)) | |
| except: | |
| print(f"Row {idx}: invalid Submission ID '{raw_id}'; skipping.") | |
| dec_counts["skipped"] += 1; continue | |
| # Only comment if the label matches (safety) and avoid duplicates via marker | |
| if decision_raw in ("accept","accepted","approve","approved"): | |
| # optional: ensure Accepted label exists on the issue before commenting | |
| labels = get_issue_labels(num) | |
| if "Accepted" in labels: | |
| add_comment_once(num, ACCEPT_MSG.format(name=candidate), ACCEPT_MARKER) | |
| dec_counts["accept_commented"] += 1 | |
| else: | |
| print(f"Row {idx}: issue #{num} lacks 'Accepted' label; skipping comment.") | |
| dec_counts["skipped"] += 1 | |
| elif decision_raw in ("reject","rejected","decline","declined"): | |
| labels = get_issue_labels(num) | |
| if "Rejected" in labels: | |
| add_comment_once(num, REJECT_MSG.format(name=candidate), REJECT_MARKER) | |
| dec_counts["reject_commented"] += 1 | |
| else: | |
| print(f"Row {idx}: issue #{num} lacks 'Rejected' label; skipping comment.") | |
| dec_counts["skipped"] += 1 | |
| else: | |
| print(f"Row {idx}: unknown decision '{decision_raw}'; skipping.") | |
| dec_counts["skipped"] += 1 | |
| except FileNotFoundError: | |
| print(f"decision.csv not found at: {dec_csv}", file=sys.stderr); sys.exit(1) | |
| # --- Part 3: Remove 'pending-review' everywhere --- | |
| removed = 0 | |
| for issue in list_issues_with_label(PENDING_LABEL) or []: | |
| remove_label(issue["number"], PENDING_LABEL) | |
| removed += 1 | |
| print(f"Duplicates: labeled={dup_counts['labeled']} / total_rows={dup_counts['total']}, missing={dup_counts['missing']}, skipped={dup_counts['skipped']}") | |
| print(f"Decisions: accept_commented={dec_counts['accept_commented']}, reject_commented={dec_counts['reject_commented']}, skipped={dec_counts['skipped']} (rows={dec_counts['rows']})") | |
| print(f"pending-review removed from {removed} issues.") | |
| if dry_run: print("DRY-RUN complete (no changes were made).") | |
| PY |