Now with legal policies and services pages

scripts/generate-sitemap.mjs

@@ -0,0 +1,99 @@
+#!/usr/bin/env node
+/**
+ * Post-build: generate sitemap.xml from prerendered pages in dist/.
+ * Scans for index.html (root and under each path), excludes 410 paths.
+ * Run after vite build and critters, before copy-410-paths so 410 dirs don't exist yet.
+ */
+
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+import { PATHS as PATHS_410 } from './410-paths.mjs';
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const DIST = path.join(__dirname, '..', 'dist');
+
+/** Must match src/lib/seo.ts SEO_DEFAULTS.baseUrl */
+const BASE_URL = 'https://mifi.ventures';
+
+const EXCLUDE_FILES = new Set(['404.html', '410.html']);
+const excludeSet = new Set(PATHS_410);
+
+/**
+ * adapter-static emits path.html or path/index.html. Walk dist and collect
+ * every .html that represents a page (exclude 404/410 and 410-gone paths).
+ */
+function findPages(dir, basePath = '') {
+    const entries = fs.readdirSync(dir, { withFileTypes: true });
+    const pages = [];
+
+    for (const e of entries) {
+        const rel = basePath ? `${basePath}/${e.name}` : e.name;
+
+        if (e.isDirectory()) {
+            const indexPath = path.join(dir, e.name, 'index.html');
+            if (fs.existsSync(indexPath)) {
+                if (!excludeSet.has(rel)) {
+                    pages.push({ path: rel, indexPath });
+                }
+            } else {
+                pages.push(...findPages(path.join(dir, e.name), rel));
+            }
+        } else if (e.name.endsWith('.html') && !EXCLUDE_FILES.has(e.name)) {
+            const urlPath = e.name === 'index.html'
+                ? basePath
+                : (basePath ? `${basePath}/${e.name.slice(0, -5)}` : e.name.slice(0, -5));
+            if (!excludeSet.has(urlPath)) {
+                pages.push({ path: urlPath, indexPath: path.join(dir, e.name) });
+            }
+        }
+    }
+
+    return pages;
+}
+
+function escapeXml(s) {
+    return s
+        .replace(/&/g, '&')
+        .replace(/</g, '&lt;')
+        .replace(/>/g, '&gt;')
+        .replace(/"/g, '&quot;')
+        .replace(/'/g, '&apos;');
+}
+
+function main() {
+    const indexHtml = path.join(DIST, 'index.html');
+    if (!fs.existsSync(indexHtml)) {
+        console.error('dist/index.html not found. Run build first.');
+        process.exit(1);
+    }
+
+    const pages = findPages(DIST).sort((a, b) => {
+        if (a.path === '') return -1;
+        if (b.path === '') return 1;
+        return a.path.localeCompare(b.path);
+    });
+    const urlElements = [];
+
+    for (const { path: pagePath, indexPath } of pages) {
+        const loc = pagePath ? `${BASE_URL}/${pagePath}` : BASE_URL;
+        const stat = fs.statSync(indexPath);
+        const lastmod = stat.mtime.toISOString().slice(0, 10);
+
+        urlElements.push(
+            `  <url>\n    <loc>${escapeXml(loc)}</loc>\n    <lastmod>${lastmod}</lastmod>\n  </url>`
+        );
+    }
+
+    const xml = `<?xml version="1.0" encoding="UTF-8"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+${urlElements.join('\n')}
+</urlset>
+`;
+
+    fs.writeFileSync(path.join(DIST, 'sitemap.xml'), xml, 'utf8');
+    console.log('✓ sitemap.xml generated with', pages.length, 'URLs.');
+}
+
+main();