validate-content.yaml

name: Validate Content

on:
  pull_request:
    branches: [master]
    paths:
      - "**.md"
  # push:
  #   branches: [master]
  #   paths:
  #     - "**.md"

# Allow the action to post PR comments
permissions:
  pull-requests: write
  contents: read

jobs:
  validate:
    name: Content Validation
    runs-on: ubuntu-latest

    steps:
      - name: Checkout content repo
        uses: actions/checkout@v4
        with:
          fetch-depth: 0

      # ─────────────────────────────────────────────
      # 1. Collect changed markdown files
      # ─────────────────────────────────────────────
      - name: Get changed markdown files
        id: changed
        run: |
          if [ "${{ github.event_name }}" = "pull_request" ]; then
            FILES=$(git diff --name-only origin/${{ github.base_ref }}...HEAD -- '*.md' | tr '\n' ' ')
          else
            FILES=$(git diff --name-only HEAD~1 HEAD -- '*.md' | tr '\n' ' ')
          fi
          echo "Changed files: $FILES"
          echo "files=$FILES" >> $GITHUB_OUTPUT

      # ─────────────────────────────────────────────
      # 2. Set up tools
      # ─────────────────────────────────────────────
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: "3.11"

      - name: Install dependencies
        run: |
          pip install pyyaml
          npm install -g markdownlint-cli

      # ─────────────────────────────────────────────
      # 3. Run all checks, write errors to report file
      # ─────────────────────────────────────────────
      - name: Run all content checks
        id: checks
        if: always()
        run: |
          python3 << 'PYEOF'
          import os, sys, yaml, glob, re, json

          CHANGED = os.environ.get("CHANGED_FILES", "").split()
          all_md  = glob.glob("**/*.md", recursive=True)
          files   = [f for f in CHANGED if f] if CHANGED else all_md

          report = []

          def add(check, filepath, line, msg):
              report.append({"check": check, "file": filepath, "line": line, "msg": msg})

          # ── Frontmatter ─────────────────────────
          REQUIRED = ["title", "author"]
          for filepath in files:
              if not os.path.isfile(filepath):
                  continue
              content = open(filepath, encoding="utf-8").read()
              if not content.startswith("---"):
                  add("Frontmatter", filepath, 1, "Missing frontmatter block (must start with `---`)")
                  continue
              parts = content.split("---", 2)
              if len(parts) < 3:
                  add("Frontmatter", filepath, 1, "Malformed frontmatter — no closing `---`")
                  continue
              try:
                  fm = yaml.safe_load(parts[1]) or {}
              except yaml.YAMLError as e:
                  add("Frontmatter", filepath, 1, f"Invalid YAML in frontmatter: {e}")
                  continue
              for field in REQUIRED:
                  if field not in fm or not str(fm[field]).strip():
                      add("Frontmatter", filepath, 1, f"Missing required field `{field}`")

          # ── Emoji / keycap sequences ─────────────
          KEYCAP = re.compile(r'[0-9#*]\uFE0F\u20E3')
          for filepath in files:
              if not os.path.isfile(filepath):
                  continue
              for lineno, line in enumerate(open(filepath, encoding="utf-8"), 1):
                  if KEYCAP.search(line):
                      add("Emoji", filepath, lineno,
                          "Unsupported keycap emoji (e.g. 1️⃣) — breaks Quartz OG image generation. Replace with plain text or a supported emoji.")

          # ── Broken internal links ────────────────
          known = set()
          for f in all_md:
              slug = os.path.splitext(f)[0].lstrip("./")
              known.add(slug.lower())
              known.add(os.path.basename(slug).lower())

          WIKILINK = re.compile(r'\[\[([^\]|#]+)(?:[|#][^\]]*)?\]\]')
          MD_LINK  = re.compile(r'\[[^\]]*\]\((?!https?://)([^)#]+?)(?:#[^)]*)?\)')

          for filepath in files:
              if not os.path.isfile(filepath):
                  continue
              for lineno, line in enumerate(open(filepath, encoding="utf-8"), 1):
                  if line.strip().startswith("```") or line.strip().startswith("`"):
                      continue
                  for m in WIKILINK.finditer(line):
                      target = os.path.splitext(m.group(1).strip())[0].lower()
                      if target not in known:
                          add("Links", filepath, lineno, f"Broken wikilink: `[[{m.group(1)}]]`")
                  for m in MD_LINK.finditer(line):
                      target = m.group(1).strip()
                      if target.startswith("mailto:") or target.startswith("/"):
                          continue
                      slug = os.path.splitext(target.lstrip("./"))[0].lower()
                      if slug not in known:
                          add("Links", filepath, lineno, f"Broken markdown link: `({m.group(1)})`")

          # ── Write report ─────────────────────────
          with open("/tmp/check_report.json", "w") as f:
              json.dump(report, f)

          if report:
              sys.exit(1)
          PYEOF
        env:
          CHANGED_FILES: ${{ steps.changed.outputs.files }}

      # ─────────────────────────────────────────────
      # 4. Markdown lint — append to report
      # ─────────────────────────────────────────────
      - name: Markdown lint
        if: always()
        run: |
          FILES="${{ steps.changed.outputs.files }}"
          if [ -z "$FILES" ]; then
            FILES=$(find . -name "*.md" | tr '\n' ' ')
          fi

          # Write output to file — avoids fragile inline variable interpolation into Python
          markdownlint $FILES --disable MD013 MD033 MD041 > /tmp/lint_out.txt 2>&1 || true

          python3 << 'PYEOF'
          import json, re

          lint_raw = open("/tmp/lint_out.txt").read()
          report_path = "/tmp/check_report.json"

          try:
              report = json.load(open(report_path))
          except Exception:
              report = []

          pattern = re.compile(r'^(.+?):(\d+)(?::\d+)?\s+(MD\d+\S+.*)$')
          added = 0
          for line in lint_raw.splitlines():
              m = pattern.match(line.strip())
              if m:
                  report.append({
                      "check": "Markdown Lint",
                      "file": m.group(1),
                      "line": int(m.group(2)),
                      "msg": m.group(3)
                  })
                  added += 1

          with open(report_path, "w") as f:
              json.dump(report, f)

          if added:
              exit(1)
          PYEOF

      # ─────────────────────────────────────────────
      # 5. Post PR comment with all errors
      # ─────────────────────────────────────────────
      - name: Post failure comment on PR
        if: failure() && github.event_name == 'pull_request'
        uses: actions/github-script@v7
        with:
          script: |
            const fs = require('fs');
            let report = [];
            try {
              report = JSON.parse(fs.readFileSync('/tmp/check_report.json', 'utf8'));
            } catch (e) {
              core.warning('Could not read report file: ' + e);
            }

            if (report.length === 0) return;

            const groups = {};
            for (const item of report) {
              if (!groups[item.check]) groups[item.check] = [];
              groups[item.check].push(item);
            }

            const icons = {
              "Frontmatter":   "📋",
              "Emoji":         "🔣",
              "Links":         "🔗",
              "Markdown Lint": "📝",
            };

            let body = `## ❌ Content Validation Failed\n\n`;
            body += `Found **${report.length} issue(s)** in the changed files. Please fix them before merging.\n\n---\n\n`;

            for (const [check, items] of Object.entries(groups)) {
              const icon = icons[check] || "⚠️";
              body += `### ${icon} ${check} — ${items.length} issue${items.length > 1 ? 's' : ''}\n\n`;
              body += `| File | Line | Issue |\n|------|------|-------|\n`;
              for (const item of items) {
                body += `| \`${item.file}\` | ${item.line} | ${item.msg} |\n`;
              }
              body += `\n`;
            }

            body += `---\n> 💡 Fix the issues above, push again, and this check will re-run automatically.`;

            const { data: comments } = await github.rest.issues.listComments({
              owner: context.repo.owner,
              repo: context.repo.repo,
              issue_number: context.issue.number,
            });

            const existing = comments.find(c =>
              c.user.type === 'Bot' && c.body.includes('Content Validation')
            );

            if (existing) {
              await github.rest.issues.updateComment({
                owner: context.repo.owner,
                repo: context.repo.repo,
                comment_id: existing.id,
                body,
              });
            } else {
              await github.rest.issues.createComment({
                owner: context.repo.owner,
                repo: context.repo.repo,
                issue_number: context.issue.number,
                body,
              });
            }

      # ─────────────────────────────────────────────
      # 6. Update comment to green when all fixed
      # ─────────────────────────────────────────────
      - name: Post success comment on PR
        if: success() && github.event_name == 'pull_request'
        uses: actions/github-script@v7
        with:
          script: |
            const { data: comments } = await github.rest.issues.listComments({
              owner: context.repo.owner,
              repo: context.repo.repo,
              issue_number: context.issue.number,
            });

            const existing = comments.find(c =>
              c.user.type === 'Bot' && c.body.includes('Content Validation')
            );

            const body = `## ✅ Content Validation Passed\n\nAll checks passed — frontmatter, links, emoji, and markdown syntax look good! Ready to merge. 🎉`;

            if (existing) {
              await github.rest.issues.updateComment({
                owner: context.repo.owner,
                repo: context.repo.repo,
                comment_id: existing.id,
                body,
              });
            } else {
              await github.rest.issues.createComment({
                owner: context.repo.owner,
                repo: context.repo.repo,
                issue_number: context.issue.number,
                body,
              });
            }