diff --git a/app/katalog/[id]/layout.tsx b/app/katalog/[id]/layout.tsx new file mode 100644 index 0000000..caa00c8 --- /dev/null +++ b/app/katalog/[id]/layout.tsx @@ -0,0 +1,39 @@ +import type { Metadata } from "next"; +import { findDataset } from "@/app/katalog/datasets"; + +const BASE = "https://0pt.app"; + +// A dataset detail is a real, indexable search target: it has its own title, +// description and content, and is linked from /katalog. Without its own +// metadata it inherits the parent /katalog canonical and search engines fold +// every dataset into the catalog page. Route-level generateMetadata is the +// existing Next mechanism for per-dataset metadata; the detail page itself +// stays a client component. +export async function generateMetadata({ + params, +}: { + params: Promise<{ id: string }>; +}): Promise { + const { id } = await params; + const dataset = findDataset(id); + + if (!dataset) { + // An unknown id renders the client-side "dataset does not exist" state. + // Make the policy explicit: not indexable, canonical back to the catalog. + return { + title: "Dataset not found — 0pt", + alternates: { canonical: `${BASE}/katalog` }, + robots: { index: false, follow: true }, + }; + } + + return { + title: `${dataset.title.en} — 0pt`, + description: `${dataset.blurb.en} ${dataset.count.en} across ${dataset.markets}. Provenance per field and transparent pricing.`, + alternates: { canonical: `${BASE}/katalog/${dataset.id}` }, + }; +} + +export default function DatasetDetailLayout({ children }: { children: React.ReactNode }) { + return children; +} diff --git a/app/katalog/[id]/page.tsx b/app/katalog/[id]/page.tsx index 43ab725..ffdd236 100644 --- a/app/katalog/[id]/page.tsx +++ b/app/katalog/[id]/page.tsx @@ -11,119 +11,7 @@ import { LoadingSkeleton, EmptyState, ErrorState } from "@/components/states"; import { useReveal } from "@/hooks/use-reveal"; import { FIELDS, SAMPLE_ROWS, MARKETS, formatSEK, type ProvLevel } from "@/lib/data"; import type { Loc } from "@/lib/data"; - -// ───────────────────────────────────────────────────────────────────────────── -// Inline catalog. No detail API yet — keyed by slugified id with a safe fallback. -// ───────────────────────────────────────────────────────────────────────────── - -interface Dataset { - id: string; - markets: string; - marketCodes: string[]; - title: Loc; - blurb: Loc; - count: Loc; - rows: number; - freshness: Loc; - level: ProvLevel; - source: string; - cat: Loc; - price: number; - artifact: "table" | "map" | "schema" | "chart"; -} - -const CATALOG: Dataset[] = [ - { - id: "nordic-companies", - markets: "SE · NO · DK · FI", - marketCodes: ["SE", "NO", "DK", "FI"], - title: { en: "Nordic company data", sv: "Nordisk bolagsdata" }, - blurb: { en: "Official registers. Per field.", sv: "Officiella register. Per fält." }, - count: { en: "4.9M records", sv: "4,9M poster" }, - rows: 4900000, - freshness: { en: "daily", sv: "daglig" }, - level: 3, - source: "GLEIF · Brreg", - cat: { en: "Companies", sv: "Bolag" }, - price: 18400, - artifact: "table", - }, - { - id: "tech-decisionmakers", - markets: "SE · NO · DK · FI", - marketCodes: ["SE", "NO", "DK", "FI"], - title: { en: "Decision-makers in tech", sv: "Beslutsfattare i tech" }, - blurb: { en: "Role and path. Source per field.", sv: "Roll och väg. Källa per fält." }, - count: { en: "182k records", sv: "182k poster" }, - rows: 182000, - freshness: { en: "weekly", sv: "veckovis" }, - level: 1, - source: "public web", - cat: { en: "Decision-makers", sv: "Beslutsfattare" }, - price: 9600, - artifact: "schema", - }, - { - id: "vehicles-fleet", - markets: "SE", - marketCodes: ["SE"], - title: { en: "Vehicles & fleet", sv: "Fordon & flotta" }, - blurb: { en: "Technical data. VehIQ synergy.", sv: "Tekniska data. VehIQ-synergin." }, - count: { en: "6.1M records", sv: "6,1M poster" }, - rows: 6100000, - freshness: { en: "daily", sv: "daglig" }, - level: 2, - source: "register", - cat: { en: "Vehicles", sv: "Fordon" }, - price: 12200, - artifact: "map", - }, - { - id: "financial-series", - markets: "SE · NO · DK · FI", - marketCodes: ["SE", "NO", "DK", "FI"], - title: { en: "Financial series", sv: "Finansiella serier" }, - blurb: { en: "Revenue, result, KPIs.", sv: "Omsättning, resultat, nyckeltal." }, - count: { en: "12 years", sv: "12 år" }, - rows: 240000, - freshness: { en: "annual", sv: "årlig" }, - level: 2, - source: "register", - cat: { en: "Financials", sv: "Finansiellt" }, - price: 14800, - artifact: "chart", - }, - { - id: "buying-signals", - markets: "SE · NO · DK · FI", - marketCodes: ["SE", "NO", "DK", "FI"], - title: { en: "Buying signals", sv: "Köpsignaler" }, - blurb: { en: "Ads, news, tech changes. Intent per company.", sv: "Annonser, nyheter, teknikbyten. Intent per bolag." }, - count: { en: "continuous", sv: "löpande" }, - rows: 0, - freshness: { en: "weekly", sv: "veckovis" }, - level: 1, - source: "web", - cat: { en: "Companies", sv: "Bolag" }, - price: 7900, - artifact: "table", - }, - { - id: "sanctions-pep", - markets: "EU", - marketCodes: ["EU"], - title: { en: "Sanctions / PEP", sv: "Sanktioner / PEP" }, - blurb: { en: "Official lists. Automated screening.", sv: "Officiella listor. Automatisk screening." }, - count: { en: "global", sv: "global" }, - rows: 0, - freshness: { en: "daily", sv: "daglig" }, - level: 3, - source: "OpenSanctions", - cat: { en: "Compliance", sv: "Efterlevnad" }, - price: 5900, - artifact: "table", - }, -]; +import { findDataset, type Dataset } from "@/app/katalog/datasets"; // Stable tab keys — never change, display labels are localized separately type TabKey = "overview" | "samples" | "quality" | "schema" | "coverage" | "price"; @@ -136,12 +24,6 @@ const TABS: { key: TabKey; label: Loc }[] = [ { key: "price", label: { en: "Price", sv: "Pris" } }, ]; -function findDataset(raw: string | undefined): Dataset | undefined { - if (!raw) return undefined; - const slug = decodeURIComponent(raw).toLowerCase(); - return CATALOG.find((d) => d.id === slug); -} - const TAB_KEYS = TABS.map((t) => t.key); function isTabKey(value: string | null): value is TabKey { return value !== null && (TAB_KEYS as string[]).includes(value); diff --git a/app/katalog/datasets.ts b/app/katalog/datasets.ts new file mode 100644 index 0000000..a1091e8 --- /dev/null +++ b/app/katalog/datasets.ts @@ -0,0 +1,124 @@ +import type { Loc, ProvLevel } from "@/lib/data"; + +// Single source of truth for the ready-made datasets shown under /katalog. +// Kept framework-neutral (no "use client") so both the client detail page and +// the route's server-side generateMetadata resolve the same ids and titles. + +// ───────────────────────────────────────────────────────────────────────────── +// Static dataset seed (no detail API yet) — keyed by slugified id with a safe fallback. +// ───────────────────────────────────────────────────────────────────────────── + +export interface Dataset { + id: string; + markets: string; + marketCodes: string[]; + title: Loc; + blurb: Loc; + count: Loc; + rows: number; + freshness: Loc; + level: ProvLevel; + source: string; + cat: Loc; + price: number; + artifact: "table" | "map" | "schema" | "chart"; +} + +export const CATALOG: Dataset[] = [ + { + id: "nordic-companies", + markets: "SE · NO · DK · FI", + marketCodes: ["SE", "NO", "DK", "FI"], + title: { en: "Nordic company data", sv: "Nordisk bolagsdata" }, + blurb: { en: "Official registers. Per field.", sv: "Officiella register. Per fält." }, + count: { en: "4.9M records", sv: "4,9M poster" }, + rows: 4900000, + freshness: { en: "daily", sv: "daglig" }, + level: 3, + source: "GLEIF · Brreg", + cat: { en: "Companies", sv: "Bolag" }, + price: 18400, + artifact: "table", + }, + { + id: "tech-decisionmakers", + markets: "SE · NO · DK · FI", + marketCodes: ["SE", "NO", "DK", "FI"], + title: { en: "Decision-makers in tech", sv: "Beslutsfattare i tech" }, + blurb: { en: "Role and path. Source per field.", sv: "Roll och väg. Källa per fält." }, + count: { en: "182k records", sv: "182k poster" }, + rows: 182000, + freshness: { en: "weekly", sv: "veckovis" }, + level: 1, + source: "public web", + cat: { en: "Decision-makers", sv: "Beslutsfattare" }, + price: 9600, + artifact: "schema", + }, + { + id: "vehicles-fleet", + markets: "SE", + marketCodes: ["SE"], + title: { en: "Vehicles & fleet", sv: "Fordon & flotta" }, + blurb: { en: "Technical data. VehIQ synergy.", sv: "Tekniska data. VehIQ-synergin." }, + count: { en: "6.1M records", sv: "6,1M poster" }, + rows: 6100000, + freshness: { en: "daily", sv: "daglig" }, + level: 2, + source: "register", + cat: { en: "Vehicles", sv: "Fordon" }, + price: 12200, + artifact: "map", + }, + { + id: "financial-series", + markets: "SE · NO · DK · FI", + marketCodes: ["SE", "NO", "DK", "FI"], + title: { en: "Financial series", sv: "Finansiella serier" }, + blurb: { en: "Revenue, result, KPIs.", sv: "Omsättning, resultat, nyckeltal." }, + count: { en: "12 years", sv: "12 år" }, + rows: 240000, + freshness: { en: "annual", sv: "årlig" }, + level: 2, + source: "register", + cat: { en: "Financials", sv: "Finansiellt" }, + price: 14800, + artifact: "chart", + }, + { + id: "buying-signals", + markets: "SE · NO · DK · FI", + marketCodes: ["SE", "NO", "DK", "FI"], + title: { en: "Buying signals", sv: "Köpsignaler" }, + blurb: { en: "Ads, news, tech changes. Intent per company.", sv: "Annonser, nyheter, teknikbyten. Intent per bolag." }, + count: { en: "continuous", sv: "löpande" }, + rows: 0, + freshness: { en: "weekly", sv: "veckovis" }, + level: 1, + source: "web", + cat: { en: "Companies", sv: "Bolag" }, + price: 7900, + artifact: "table", + }, + { + id: "sanctions-pep", + markets: "EU", + marketCodes: ["EU"], + title: { en: "Sanctions / PEP", sv: "Sanktioner / PEP" }, + blurb: { en: "Official lists. Automated screening.", sv: "Officiella listor. Automatisk screening." }, + count: { en: "global", sv: "global" }, + rows: 0, + freshness: { en: "daily", sv: "daglig" }, + level: 3, + source: "OpenSanctions", + cat: { en: "Compliance", sv: "Efterlevnad" }, + price: 5900, + artifact: "table", + }, +]; + +export function findDataset(raw: string | undefined): Dataset | undefined { + if (!raw) return undefined; + const slug = decodeURIComponent(raw).toLowerCase(); + return CATALOG.find((d) => d.id === slug); +} diff --git a/app/sitemap.ts b/app/sitemap.ts index aa2f2a3..ee6c2c5 100644 --- a/app/sitemap.ts +++ b/app/sitemap.ts @@ -1,7 +1,10 @@ import type { MetadataRoute } from "next"; const BASE = "https://0pt.app"; -const ROUTES = ["", "/tjanster", "/katalog", "/kallor", "/bestall", "/bestall/beskriv", "/bestall/forskott"]; +// /bestall, /bestall/beskriv and /bestall/forskott are deliberately noindex +// (app/bestall/layout.tsx) — the order flow is transactional, not a landing +// page. Keep the sitemap in sync with that indexation policy. +const ROUTES = ["", "/tjanster", "/katalog", "/kallor"]; export default function sitemap(): MetadataRoute.Sitemap { return ROUTES.map((r) => ({ diff --git a/tests/seo/sitemap-and-dataset-metadata.test.ts b/tests/seo/sitemap-and-dataset-metadata.test.ts new file mode 100644 index 0000000..f3d3769 --- /dev/null +++ b/tests/seo/sitemap-and-dataset-metadata.test.ts @@ -0,0 +1,121 @@ +/** + * SEO regression guard: sitemap vs. indexation policy, and per-dataset + * canonicals on /katalog/[id]. + * + * PR #622 made /bestall*, noindex via app/bestall/layout.tsx but left those + * steps in app/sitemap.ts, and the new /katalog layout canonical leaked down + * to every dataset detail. These tests exercise the public surfaces together: + * home (sitemap root), catalog, two datasets, an order step, sitemap + robots. + */ +import { describe, it, expect } from "vitest"; +import type { Metadata } from "next"; +import sitemap from "@/app/sitemap"; +import robots from "@/app/robots"; +import { metadata as catalogMetadata } from "@/app/katalog/layout"; +import { metadata as bestallMetadata } from "@/app/bestall/layout"; +import { generateMetadata as datasetGenerateMetadata } from "@/app/katalog/[id]/layout"; +import { CATALOG } from "@/app/katalog/datasets"; + +const BASE = "https://0pt.app"; + +function canonicalOf(m: Metadata): unknown { + return m.alternates?.canonical; +} + +// Next's Metadata.robots is `string | Robots`; the root layout sets +// index:true, so an unset value means the detail stays indexable. +function isIndexed(m: Metadata): boolean { + const r = m.robots; + if (r === undefined || r === null) return true; + if (typeof r === "string") return r !== "noindex" && r !== "none"; + return r.index !== false; +} + +function datasetMeta(id: string): Promise { + // Next 16 passes dynamic route params as a Promise to generateMetadata. + return datasetGenerateMetadata({ params: Promise.resolve({ id }) }); +} + +describe("sitemap — public pages only", () => { + const urls = () => sitemap().map((entry) => entry.url); + + it("lists home, tjanster, katalog and kallor — and nothing else", () => { + expect(urls()).toEqual([ + `${BASE}`, + `${BASE}/tjanster`, + `${BASE}/katalog`, + `${BASE}/kallor`, + ]); + }); + + it("never lists the noindex order flow (/bestall*)", () => { + expect(urls().filter((url) => url.startsWith(`${BASE}/bestall`))).toEqual([]); + }); + + it("keeps home as the top-priority entry", () => { + const home = sitemap().find((entry) => entry.url === `${BASE}`); + expect(home).toBeDefined(); + expect(home?.priority).toBe(1); + for (const entry of sitemap().filter((e) => e.url !== `${BASE}`)) { + expect(entry.priority).toBeLessThan(1); + } + }); +}); + +describe("robots + order-step indexation", () => { + it("allows public crawling, disallows auth surfaces and advertises the sitemap", () => { + const rules = robots().rules as Array<{ + userAgent: string; + allow?: string; + disallow?: string | string[]; + }>; + expect(rules[0].userAgent).toBe("*"); + expect(rules[0].allow).toBe("/"); + expect(rules[0].disallow).toEqual(["/portal/", "/admin/", "/order/", "/api/"]); + expect(robots().sitemap).toBe(`${BASE}/sitemap.xml`); + }); + + it("the order step is noindex/follow and stays out of the sitemap", () => { + expect(bestallMetadata.robots).toMatchObject({ index: false, follow: true }); + expect(sitemap().map((e) => e.url).some((u) => u.includes("/bestall"))).toBe(false); + }); +}); + +describe("catalog + dataset detail canonicals", () => { + it("the catalog index keeps its own canonical", () => { + expect(canonicalOf(catalogMetadata)).toBe(`${BASE}/katalog`); + }); + + it("every indexable dataset detail carries its own canonical, title and description", async () => { + expect(CATALOG).toHaveLength(6); + for (const dataset of CATALOG) { + const meta = await datasetMeta(dataset.id); + expect(canonicalOf(meta)).toBe(`${BASE}/katalog/${dataset.id}`); + expect(canonicalOf(meta)).not.toBe(canonicalOf(catalogMetadata)); + expect(String(meta.title)).toContain(dataset.title.en); + expect(String(meta.description ?? "")).toContain(dataset.blurb.en); + // Root layout declares index:true; details must not silently opt out. + expect(isIndexed(meta)).toBe(true); + } + }); + + it("two different datasets resolve to distinct canonicals and titles", async () => { + const a = await datasetMeta("nordic-companies"); + const b = await datasetMeta("tech-decisionmakers"); + expect(canonicalOf(a)).toBe(`${BASE}/katalog/nordic-companies`); + expect(canonicalOf(b)).toBe(`${BASE}/katalog/tech-decisionmakers`); + expect(canonicalOf(a)).not.toBe(canonicalOf(b)); + expect(a.title).not.toBe(b.title); + }); + + it("canonical uses the stable catalogue id even for a non-normalized request", async () => { + const meta = await datasetMeta("Nordic-Companies"); + expect(canonicalOf(meta)).toBe(`${BASE}/katalog/nordic-companies`); + }); + + it("an unknown dataset id is explicitly noindex and canonicals to the catalog", async () => { + const meta = await datasetMeta("does-not-exist"); + expect(meta.robots).toMatchObject({ index: false, follow: true }); + expect(canonicalOf(meta)).toBe(`${BASE}/katalog`); + }); +});