#!/usr/bin/env node
/**
 * Generate public/sitemap.xml and public/feed.xml from the real route tree
 * and the blog manifest.
 *
 * Runs automatically before `next build` (see the "prebuild" npm script), so
 * the deployed sitemap can never drift from the pages that actually exist.
 * Fails loudly if the blog manifest cannot be parsed, rather than silently
 * emitting a partial sitemap.
 */
import fs from "node:fs";
import path from "node:path";
import { fileURLToPath } from "node:url";

const root = path.resolve(path.dirname(fileURLToPath(import.meta.url)), "..");
const APP_DIR = path.join(root, "src", "app");
const MANIFEST = path.join(root, "src", "content", "blog", "manifest.ts");
const PUBLIC_DIR = path.join(root, "public");
const ORIGIN = "https://dashcaddy.net";
const SITE_DESCRIPTION =
  "Self-hosting deep-dives, product updates, and practical guides to running your own infrastructure.";

// Pages that must never be advertised to crawlers.
const EXCLUDE = new Set(["/success/", "/404/", "/500/"]);

function walk(dir) {
  const found = [];
  for (const entry of fs.readdirSync(dir, { withFileTypes: true })) {
    const full = path.join(dir, entry.name);
    if (entry.isDirectory()) found.push(...walk(full));
    else if (entry.name === "page.tsx") found.push(full);
  }
  return found;
}

function routeFor(file) {
  const rel = path.relative(APP_DIR, path.dirname(file));
  const parts = rel
    .split(path.sep)
    .filter((part) => part && part !== "." && !/^\(.*\)$/.test(part));
  // Dynamic segments ([slug]) are expanded from data sources, not enumerated.
  if (parts.some((part) => part.startsWith("["))) return null;
  return "/" + parts.join("/") + (parts.length ? "/" : "");
}

function meta(route) {
  if (route === "/") return { priority: "1.0", changefreq: "weekly" };
  if (route === "/blog/" || route === "/docs/" || route === "/features/")
    return { priority: "0.9", changefreq: "weekly" };
  if (route === "/pricing/" || route === "/docs/catalog/")
    return { priority: "0.8", changefreq: "weekly" };
  if (route.startsWith("/docs/catalog/")) return { priority: "0.6", changefreq: "monthly" };
  if (route.startsWith("/docs/")) return { priority: "0.7", changefreq: "monthly" };
  if (route === "/about/") return { priority: "0.6", changefreq: "monthly" };
  if (route.startsWith("/policies/")) return { priority: "0.3", changefreq: "yearly" };
  return { priority: "0.5", changefreq: "monthly" };
}

function xmlEscape(value) {
  return value
    .replace(/&/g, "&")
    .replace(/</g, "&lt;")
    .replace(/>/g, "&gt;")
    .replace(/"/g, "&quot;");
}

// --- Blog manifest ---------------------------------------------------------
const manifestSrc = fs.readFileSync(MANIFEST, "utf8");

const POSTS = [
  ...manifestSrc.matchAll(
    /\{\s*slug:\s*"([^"]+)",\s*title:\s*"((?:[^"\\]|\\.)*)",\s*description:\s*"((?:[^"\\]|\\.)*)",\s*date:\s*"([^"]+)"/g
  ),
].map((m) => ({ slug: m[1], title: m[2], description: m[3], date: m[4] }));

const manifestSlugCount = [...manifestSrc.matchAll(/slug:\s*"([^"]+)"/g)].length;
if (POSTS.length === 0 || POSTS.length !== manifestSlugCount) {
  throw new Error(
    `generate-sitemap: parsed ${POSTS.length} posts but manifest declares ${manifestSlugCount} slugs — refusing to emit a partial sitemap`
  );
}

// --- Sitemap ---------------------------------------------------------------
const staticRoutes = [...new Set(walk(APP_DIR).map(routeFor).filter(Boolean))]
  .filter((route) => !EXCLUDE.has(route))
  .sort();

const urls = [
  ...staticRoutes.map((route) => ({ loc: ORIGIN + route, ...meta(route) })),
  ...POSTS.map((post) => ({
    loc: `${ORIGIN}/blog/${post.slug}/`,
    lastmod: post.date,
    changefreq: "monthly",
    priority: "0.8",
  })),
].sort((a, b) => (a.loc < b.loc ? -1 : 1));

const sitemap =
  `<?xml version="1.0" encoding="UTF-8"?>\n` +
  `<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">\n` +
  urls
    .map(
      (u) =>
        `  <url>\n    <loc>${u.loc}</loc>` +
        (u.lastmod ? `\n    <lastmod>${u.lastmod}</lastmod>` : "") +
        `\n    <changefreq>${u.changefreq}</changefreq>\n    <priority>${u.priority}</priority>\n  </url>`
    )
    .join("\n") +
  `\n</urlset>\n`;

// --- RSS feed --------------------------------------------------------------
const sortedPosts = [...POSTS].sort((a, b) => (a.date < b.date ? 1 : -1));

const feed =
  `<?xml version="1.0" encoding="UTF-8"?>\n` +
  `<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">\n  <channel>\n` +
  `    <title>DashCaddy Blog</title>\n` +
  `    <link>${ORIGIN}/blog/</link>\n` +
  `    <description>${xmlEscape(SITE_DESCRIPTION)}</description>\n` +
  `    <language>en-us</language>\n` +
  `    <atom:link href="${ORIGIN}/feed.xml" rel="self" type="application/rss+xml" />\n` +
  sortedPosts
    .map(
      (p) =>
        `    <item>\n` +
        `      <title>${xmlEscape(p.title)}</title>\n` +
        `      <link>${ORIGIN}/blog/${p.slug}/</link>\n` +
        `      <guid isPermaLink="true">${ORIGIN}/blog/${p.slug}/</guid>\n` +
        `      <pubDate>${new Date(`${p.date}T00:00:00Z`).toUTCString()}</pubDate>\n` +
        `      <description>${xmlEscape(p.description)}</description>\n` +
        `    </item>`
    )
    .join("\n") +
  `\n  </channel>\n</rss>\n`;

fs.mkdirSync(PUBLIC_DIR, { recursive: true });
fs.writeFileSync(path.join(PUBLIC_DIR, "sitemap.xml"), sitemap);
fs.writeFileSync(path.join(PUBLIC_DIR, "feed.xml"), feed);

console.log(
  `generate-sitemap: ${urls.length} sitemap URLs (${staticRoutes.length} pages + ${POSTS.length} posts), ${sortedPosts.length} feed items`
);