import test from "node:test";
import assert from "node:assert/strict";
import { getAdapter, normalizeUrl } from "../src/ingest/adapters/index.js";

const RSS = `<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel>
  <title>Test Feed</title>
  <link>https://example.com/blog</link>
  <item>
    <title>First  Post</title>
    <link>https://example.com/posts/first?utm_source=feed</link>
    <description><![CDATA[<p>Some <b>html</b> description</p>]]></description>
  </item>
  <item>
    <title>Relative Link</title>
    <link>/posts/relative</link>
    <description>Relative one</description>
  </item>
  <item>
    <title>Atom style</title>
    <link href="https://example.com/posts/atom" />
    <description>Atom styled link</description>
  </item>
</channel></rss>`;

test("rss adapter resolves each item's own link, not the feed url", async () => {
  // 保留原始 fetch：把 RSS 内容喂进去
  const realFetch = globalThis.fetch;
  globalThis.fetch = async () => new Response(RSS, { status: 200 });
  try {
    const items = await getAdapter("rss").fetch("https://example.com/feed.xml");
    assert.equal(items.length, 3);

    assert.equal(items[0].title, "First Post");
    // 关键：url 必须是条目的链接，而不是 https://example.com/feed.xml
    assert.equal(items[0].url, "https://example.com/posts/first");
    // utm_ 参数被去掉
    assert.ok(!items[0].url.includes("utm_source"));
    assert.match(items[0].description, /Some html description/);

    // 相对链接按 feed 地址解析成绝对地址
    assert.equal(items[1].url, "https://example.com/posts/relative");

    // <link href="..."> 形式也能取到
    assert.equal(items[2].url, "https://example.com/posts/atom");

    assert.ok(items.every((i) => i.url !== "https://example.com/feed.xml"));
  } finally {
    globalThis.fetch = realFetch;
  }
});

test("normalizeUrl drops tracking params and trailing slash", () => {
  assert.equal(normalizeUrl("https://a.com/x/?utm_src=b#frag"), "https://a.com/x");
  assert.equal(normalizeUrl("https://a.com/x"), "https://a.com/x");
});