Add CI check to keep Person/WebSite JSON-LD in sync

.github/workflows/schema-sync-check.yml

@@ -0,0 +1,17 @@
+name: Schema sync check
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  check:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 22
+      - run: node scripts/check-schema-sync.mjs

scripts/check-schema-sync.mjs

@@ -0,0 +1,77 @@
+#!/usr/bin/env node
+import { readFileSync } from 'node:fs';
+import { fileURLToPath } from 'node:url';
+import path from 'node:path';
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const repoRoot = path.resolve(__dirname, '..');
+
+const FILES = ['index.html', 'book.html'];
+const SHARED_IDS = [
+  'https://suedeai.ai/founder#person',
+  'https://suede-ai.github.io/#website',
+];
+
+function extractGraph(fileName) {
+  const html = readFileSync(path.join(repoRoot, fileName), 'utf8');
+  const match = html.match(/<script type="application\/ld\+json">([\s\S]*?)<\/script>/);
+  if (!match) {
+    throw new Error(`No JSON-LD script block found in ${fileName}`);
+  }
+  const data = JSON.parse(match[1]);
+  return data['@graph'] || [];
+}
+
+function findNode(graph, id) {
+  return graph.find((node) => node['@id'] === id);
+}
+
+function canonicalize(value) {
+  if (Array.isArray(value)) {
+    return value.map(canonicalize);
+  }
+  if (value && typeof value === 'object') {
+    const sorted = {};
+    for (const key of Object.keys(value).sort()) {
+      sorted[key] = canonicalize(value[key]);
+    }
+    return sorted;
+  }
+  return value;
+}
+
+const graphs = Object.fromEntries(FILES.map((f) => [f, extractGraph(f)]));
+
+let hasMismatch = false;
+
+for (const id of SHARED_IDS) {
+  const nodesByFile = FILES.map((f) => [f, findNode(graphs[f], id)]);
+  const [[, referenceNode], ...rest] = nodesByFile;
+
+  if (!referenceNode) {
+    console.error(`Node "${id}" is missing from ${FILES[0]}.`);
+    hasMismatch = true;
+    continue;
+  }
+
+  const referenceJson = JSON.stringify(canonicalize(referenceNode));
+
+  for (const [fileName, node] of rest) {
+    if (!node) {
+      console.error(`Node "${id}" is missing from ${fileName}.`);
+      hasMismatch = true;
+      continue;
+    }
+    if (JSON.stringify(canonicalize(node)) !== referenceJson) {
+      console.error(`Node "${id}" differs between ${FILES[0]} and ${fileName}.`);
+      hasMismatch = true;
+    }
+  }
+}
+
+if (hasMismatch) {
+  console.error('\nThe Person/WebSite JSON-LD nodes must stay identical across index.html and book.html.');
+  process.exit(1);
+}
+
+console.log('Schema sync check passed: Person/WebSite JSON-LD nodes match across index.html and book.html.');