Generate text/index.md via CI scheduled/push/manual

.ci-scripts/generate-index.py

@@ -0,0 +1,184 @@
+#!/usr/bin/env python3
+"""Generate a markdown index of every RFC and the status of its PR/issue.
+
+Reads the RFC header from each `text/NNNN-*.md`, queries the GitHub GraphQL
+API for the live state of each referenced RFC PR (ponylang/rfcs) and Pony
+issue (ponylang/ponyc), and prints a markdown table to stdout (or `-o FILE`).
+
+Requires the `GITHUB_TOKEN` environment variable. In a GitHub Actions job the
+default workflow token is sufficient; this script only reads public data.
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import os
+import re
+import sys
+import urllib.request
+from pathlib import Path
+
+HEADER_FIELDS = {
+    "- Feature Name:": "feature",
+    "- Start Date:": "start",
+    "- RFC PR:": "pr_url",
+    "- Pony Issue:": "issue_url",
+}
+PR_URL_RE = re.compile(r"github\.com/ponylang/rfcs/pull/(\d+)")
+ISSUE_URL_RE = re.compile(r"github\.com/ponylang/ponyc/issues/(\d+)")
+
+GRAPHQL_URL = "https://api.github.com/graphql"
+
+
+def parse_header(path: Path) -> dict[str, str]:
+    fields = {v: "" for v in HEADER_FIELDS.values()}
+    with path.open() as f:
+        for line in f:
+            line = line.rstrip("\n")
+            if line.startswith("# "):
+                break
+            for prefix, field in HEADER_FIELDS.items():
+                if line.startswith(prefix):
+                    fields[field] = line[len(prefix):].strip()
+                    break
+    return fields
+
+
+def query_github(prs: set[int], issues: set[int], token: str) -> dict:
+    if not prs and not issues:
+        return {"rfcs": {}, "ponyc": {}}
+    pr_aliases = " ".join(
+        f"pr{n}: pullRequest(number: {n}) {{ state merged }}"
+        for n in sorted(prs)
+    )
+    issue_aliases = " ".join(
+        f"i{n}: issue(number: {n}) {{ state stateReason }}"
+        for n in sorted(issues)
+    )
+    query = (
+        "query {"
+        f' rfcs: repository(owner: "ponylang", name: "rfcs") {{ {pr_aliases} }}'
+        f' ponyc: repository(owner: "ponylang", name: "ponyc") {{ {issue_aliases} }}'
+        " }"
+    )
+    req = urllib.request.Request(
+        GRAPHQL_URL,
+        data=json.dumps({"query": query}).encode(),
+        headers={
+            "Authorization": f"bearer {token}",
+            "User-Agent": "ponylang-rfc-index-generator",
+            "Content-Type": "application/json",
+        },
+    )
+    with urllib.request.urlopen(req) as resp:
+        body = json.load(resp)
+    if body.get("errors"):
+        raise RuntimeError(f"GraphQL errors: {body['errors']}")
+    return body["data"]
+
+
+def pr_status(node: dict | None) -> str:
+    if node is None:
+        return "Unknown"
+    if node.get("merged"):
+        return "Merged"
+    return node.get("state", "Unknown").capitalize()
+
+
+def issue_status(node: dict | None) -> str:
+    if node is None:
+        return "Unknown"
+    if node.get("state") == "OPEN":
+        return "Open"
+    reason = node.get("stateReason")
+    return {
+        "COMPLETED": "Closed (completed)",
+        "NOT_PLANNED": "Closed (not planned)",
+        "REOPENED": "Open (reopened)",
+    }.get(reason, "Closed")
+
+
+def cell(num: int | None, url: str, status: str) -> str:
+    if num is None:
+        return ""
+    return f"[#{num}]({url}) — {status}"
+
+
+def collect_rfcs(text_dir: Path) -> list[dict]:
+    rfcs = []
+    for path in sorted(text_dir.glob("[0-9]*.md")):
+        header = parse_header(path)
+        pr_match = PR_URL_RE.search(header["pr_url"])
+        issue_match = ISSUE_URL_RE.search(header["issue_url"])
+        rfcs.append({
+            "num": path.name.split("-", 1)[0],
+            "start": header["start"],
+            "feature": header["feature"],
+            "pr_num": int(pr_match.group(1)) if pr_match else None,
+            "pr_url": header["pr_url"] if pr_match else "",
+            "issue_num": int(issue_match.group(1)) if issue_match else None,
+            "issue_url": header["issue_url"] if issue_match else "",
+        })
+    return rfcs
+
+
+def render(rfcs: list[dict], data: dict) -> str:
+    lines = [
+        "# RFC Index",
+        "",
+        "| RFC # | Start Date | Feature Name | RFC PR | Pony Issue |",
+        "|-------|------------|--------------|--------|------------|",
+    ]
+    for r in rfcs:
+        pr_node = data["rfcs"].get(f"pr{r['pr_num']}") if r["pr_num"] else None
+        issue_node = data["ponyc"].get(f"i{r['issue_num']}") if r["issue_num"] else None
+        lines.append(
+            f"| {r['num']} | {r['start']} | {r['feature']} | "
+            f"{cell(r['pr_num'], r['pr_url'], pr_status(pr_node))} | "
+            f"{cell(r['issue_num'], r['issue_url'], issue_status(issue_node))} |"
+        )
+    lines.append("\nThis file is automatically generated, do not edit")
+    return "\n".join(lines) + "\n"
+
+
+def main() -> int:
+    ap = argparse.ArgumentParser(description=__doc__)
+    ap.add_argument(
+        "text_dir",
+        nargs="?",
+        default="text",
+        type=Path,
+        help="Directory of RFC .md files (default: text)",
+    )
+    ap.add_argument(
+        "-o", "--output",
+        type=Path,
+        help="Output file (default: stdout)",
+    )
+    args = ap.parse_args()
+
+    if not args.text_dir.is_dir():
+        print(f"error: not a directory: {args.text_dir}", file=sys.stderr)
+        return 2
+
+    token = os.environ.get("GITHUB_TOKEN")
+    if not token:
+        print("error: GITHUB_TOKEN not set", file=sys.stderr)
+        return 2
+
+    rfcs = collect_rfcs(args.text_dir)
+    prs = {r["pr_num"] for r in rfcs if r["pr_num"] is not None}
+    issues = {r["issue_num"] for r in rfcs if r["issue_num"] is not None}
+    data = query_github(prs, issues, token)
+    output = render(rfcs, data)
+
+    if args.output:
+        args.output.write_text(output)
+    else:
+        sys.stdout.write(output)
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

.github/workflows/update-rfc-index.yml

@@ -0,0 +1,51 @@
+name: Update RFC index
+
+on:
+  schedule:
+    # Daily at 03:17 UTC. Picks up status changes on tracking issues
+    # in ponylang/ponyc even when no commits land in this repo.
+    - cron: '17 3 * * *'
+  push:
+    branches: [main]
+    paths:
+      - 'text/[0-9]*.md'
+      - '.ci-scripts/generate-index.py'
+      - '.github/workflows/update-rfc-index.yml'
+  workflow_dispatch:
+
+concurrency:
+  group: update-rfc-index
+  cancel-in-progress: false
+
+permissions:
+  contents: write
+
+jobs:
+  update:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6.0.2
+
+      - name: Set up Python
+        uses: actions/setup-python@v6.0.0
+        with:
+          python-version: '3.x'
+
+      - name: Generate index
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          python3 .ci-scripts/generate-index.py > text/index.md
+
+      - name: Commit and push if changed
+        run: |
+          if [[ -z "$(git status --porcelain text/index.md)" ]]; then
+            echo "No changes to text/index.md."
+            exit 0
+          fi
+          git config user.name "github-actions[bot]"
+          git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
+          git add text/index.md
+          git commit -m "Regenerate RFC index"
+          git push