disclosure-bureau/web/app/sitemap.ts
Luiz Gustavo 8283237f87
Some checks failed
CI / Web — typecheck + lint + build (push) Failing after 34s
CI / Scripts — Python smoke (push) Failing after 6s
CI / Web — npm audit (push) Failing after 40s
CI / Retrieval — golden set (Recall@5 + MRR) (push) Failing after 4s
W5.4 followup: hero illustration on /c/[slug] + sitemap fix
Hero illustration:
  - Painterly 16:9 editorial illustration generated via Nano Banana Pro
    for the featured case (green-fireballs-narrative): late-1940s
    desert night, vivid emerald fireball over silhouetted Sandia mesas,
    1948-era state-police sedan parked on US 66 shoulder with an
    officer in period uniform looking up, faint green glow on his
    face. Sandia Base 5 Miles roadsign. New Yorker-cover painterly
    register, NOT photorealistic, NOT sci-fi.
  - Stored at /data/disclosure/processing/case-art/<slug>.png, served
    through the existing /api/static/processing/ route. 2.7MB at 2K.
  - components/featured-case.tsx: prefers the illustration over the
    declassified-page thumbnail when present. Tags it "Editorial
    illustration" / "Ilustração editorial" so the reader knows it's
    not a photograph.
  - app/c/[slug]/page.tsx: full-bleed editorial hero at the top of
    the article when an illustration exists for the slug. Title sits
    on the image with gradient overlay; "Ilustração editorial" chip
    in the top-right corner labels the art honestly. When no
    illustration exists the page falls back to the plain title header.

Sitemap fix:
  - Added export const dynamic = "force-dynamic" + revalidate = 3600
    to app/sitemap.ts. Without these Next.js statically generated the
    sitemap at build time, when the DB and case-files volume were
    unreachable from the build container — which is why production
    was serving only the 9 static URLs instead of ~3000.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 16:16:20 -03:00

128 lines
4.5 KiB
TypeScript

/**
* sitemap.xml — dynamic, regenerated per request.
*
* Aggregates:
* - Static pages (home, bureau, sub-pages)
* - Every declassified document (/d/<id>)
* - Every case report (/c/<slug>)
* - Every entity with an AI summary (/e/<class>/<id>) — these get a
* lastModified from summary_generated_at, which helps crawlers
* re-index when we re-enrich.
* - Every top-mentioned entity even without summary (cap at 500 per
* class so the sitemap doesn't balloon past Google's 50k limit).
*
* Per Next.js the file must export a default function returning a flat
* MetadataRoute.Sitemap array. ChangeFreq/priority are honoured by most
* crawlers as hints.
*/
import type { MetadataRoute } from "next";
import { listDocuments } from "@/lib/wiki";
import { pgQuery } from "@/lib/retrieval/db";
import { readdir } from "node:fs/promises";
import path from "node:path";
// Without these, Next.js statically generates the sitemap at build time
// when the DB is unreachable from the build container — which is why we
// were getting only 9 static URLs in production.
export const dynamic = "force-dynamic";
export const revalidate = 3600;
const SITE_URL = process.env.NEXT_PUBLIC_SITE_URL ?? "https://disclosure.top";
const CASE_ROOT = process.env.CASE_ROOT ?? "/data/ufo/case";
type Url = MetadataRoute.Sitemap[number];
const ENTITY_FOLDER_BY_CLASS: Record<string, string> = {
event: "events",
person: "people",
uap_object: "uap-objects",
location: "locations",
organization: "organizations",
};
export default async function sitemap(): Promise<MetadataRoute.Sitemap> {
const out: Url[] = [];
const now = new Date();
// 1. Top-level pages
const STATIC_PAGES = [
{ url: "/", priority: 1.0, changeFrequency: "daily" as const },
{ url: "/bureau", priority: 0.9, changeFrequency: "weekly" as const },
{ url: "/sightings", priority: 0.9, changeFrequency: "weekly" as const },
{ url: "/witnesses", priority: 0.8, changeFrequency: "weekly" as const },
{ url: "/objects", priority: 0.8, changeFrequency: "weekly" as const },
{ url: "/locations", priority: 0.8, changeFrequency: "weekly" as const },
{ url: "/operations", priority: 0.8, changeFrequency: "weekly" as const },
{ url: "/documents", priority: 0.8, changeFrequency: "weekly" as const },
{ url: "/search", priority: 0.5, changeFrequency: "monthly" as const },
];
for (const p of STATIC_PAGES) {
out.push({
url: `${SITE_URL}${p.url}`,
lastModified: now,
changeFrequency: p.changeFrequency,
priority: p.priority,
});
}
// 2. Documents
try {
const docIds = await listDocuments();
for (const id of docIds) {
out.push({
url: `${SITE_URL}/d/${id}`,
lastModified: now,
changeFrequency: "monthly",
priority: 0.7,
});
}
} catch { /* fs failure — skip docs */ }
// 3. Case reports — read filesystem for /c/[slug]
try {
const dir = path.join(CASE_ROOT, "reports");
const files = await readdir(dir);
for (const f of files.filter((x) => x.endsWith(".md"))) {
out.push({
url: `${SITE_URL}/c/${f.replace(/\.md$/, "")}`,
lastModified: now,
changeFrequency: "monthly",
priority: 0.95,
});
}
} catch { /* no case files yet */ }
// 4. Entities — surface those with summaries first (high priority), plus
// the top by mention count up to 500/class. Cap per class avoids blowing
// past sitemap size limits (Google: 50k urls, 50MB).
for (const [klass, folder] of Object.entries(ENTITY_FOLDER_BY_CLASS)) {
try {
const rows = await pgQuery<{
entity_id: string;
summary_generated_at: string | null;
summary_status: string | null;
total_mentions: number;
}>(
`SELECT entity_id, summary_generated_at, summary_status, total_mentions
FROM public.entities
WHERE entity_class = $1
AND total_mentions >= 1
ORDER BY (summary_status IN ('ai_generated','curated')) DESC,
total_mentions DESC, entity_id ASC
LIMIT 500`,
[klass],
);
for (const r of rows) {
const hasSummary = r.summary_status === "ai_generated" || r.summary_status === "curated";
out.push({
url: `${SITE_URL}/e/${folder}/${r.entity_id}`,
lastModified: r.summary_generated_at ? new Date(r.summary_generated_at) : now,
changeFrequency: "monthly",
priority: hasSummary ? 0.7 : 0.4,
});
}
} catch { /* db unavailable for this class — skip */ }
}
return out;
}