import { describe, expect, test } from "vitest";
import { JSDOM } from "jsdom";
import {
extractMetadataFromDocument,
extractPublishedTime,
formatMetadataBlock,
pickBestContentNode,
} from "../utils";
describe("extractMetadataFromDocument", () => {
test("extracts metadata from standard meta tags", () => {
const html = `
Doc Title
`;
const document = new JSDOM(html).window.document;
const metadata = extractMetadataFromDocument(document);
expect(metadata.title).toBe("OG Title");
expect(metadata.byline).toBe("Ada Lovelace");
expect(metadata.siteName).toBe("Example Site");
expect(metadata.publishedTime).toBe("2025-01-01T10:00:00Z");
expect(metadata.lang).toBe("en");
});
});
describe("extractPublishedTime", () => {
test("falls back to time elements", () => {
const html = `
`;
const document = new JSDOM(html).window.document;
expect(extractPublishedTime(document)).toBe("2024-12-24");
});
});
describe("pickBestContentNode", () => {
test("chooses the most substantial content block", () => {
const html = `
Short.
This is the main content with more text than the article.
Another sentence to make it longer.
`;
const document = new JSDOM(html).window.document;
const selected = pickBestContentNode(document);
expect(selected?.selector).toBe("main");
expect(selected?.text).toContain("main content");
});
});
describe("formatMetadataBlock", () => {
test("renders metadata lines", () => {
const block = formatMetadataBlock(
{
title: "Article Title",
byline: "Author Name",
siteName: "Site",
publishedTime: "2024-10-01",
lang: "en",
},
{ url: "https://example.com", contentType: "text/html" },
);
expect(block).toContain("URL: https://example.com");
expect(block).toContain("Content-Type: text/html");
expect(block).toContain("Title: Article Title");
expect(block).toContain("Byline: Author Name");
expect(block).toContain("Site: Site");
expect(block).toContain("Published: 2024-10-01");
expect(block).toContain("Language: en");
});
});