fix(search): surface iTunes stubs and resolve feeds for delisted shows

Shows that left Apple Podcasts (e.g. Daily Wire's in 2021) come back from the iTunes Search API as metadata-only stub records with feedUrl null. mapItunesResult dropped them, so The Ben Shapiro Show — the #1 hit for 'ben shapiro' — never appeared in search while sibling shows did. - Keep feedUrl-less results (feedUrl "" + directoryUrl pointing at the Apple page) so delisted shows stay findable. - Resolve the real feed from the Apple page at subscribe time (itunes-feed-resolver: anchor on the collection's adamId, forward-scan for the embedded feedUrl; Apple serves page variants where the showOffer block sits thousands of chars after the adamId). - addFeed refuses feedless stubs whose feed can't be resolved instead of adding a broken feed; SearchPage surfaces the failure via toast. - Tests: stub mapping, extractor variants, and an end-to-end subscribe over a local HTTP server.
2026-08-10 22:38:03 -04:00
parent e73e608b9f
commit 0b0637b9dc
8 changed files with 364 additions and 11 deletions
--- a/tests/feedless-subscribe.test.ts
+++ b/tests/feedless-subscribe.test.ts
@@ -0,0 +1,117 @@
+/**
+ * End-to-end subscribe test for feedless directory stubs.
+ *
+ * A delisted show (feedUrl "" + directoryUrl) must resolve its real feed from
+ * the directory page inside addFeed — so subscribing just works. When the
+ * page can't be resolved, addFeed must refuse (return null) instead of adding
+ * a broken feed. Served over a real local HTTP server, mirroring how the
+ * app's other store tests exercise the network path.
+ */
+import { test, expect, beforeAll, afterAll } from "bun:test";
+import { mkdtempSync, rmSync } from "fs";
+import { tmpdir } from "os";
+import { join } from "path";
+
+// Point the config dir at a throwaway directory BEFORE importing the stores
+// (their module-level init reads it).
+const configHome = mkdtempSync(join(tmpdir(), "podtui-feedless-"));
+process.env.XDG_CONFIG_HOME = configHome;
+
+import { useFeedStore } from "../src/stores/feed";
+import type { Podcast } from "../src/types/podcast";
+
+const FEED_ID = "12345";
+const EPISODE_TITLES = ["Ep 2", "Ep 1"];
+
+function pageHtml(feedUrl: string): string {
+  return `<html><body><script>
+{"pageData":{"showOffer":{"title":"Delisted Show","adamId":"${FEED_ID}","feedUrl":"${feedUrl}","showType":"episodic"}}}
+</script></body></html>`;
+}
+
+function feedXml(origin: string): string {
+  const items = EPISODE_TITLES.map(
+    (title, i) => `<item>
+<title>${title}</title>
+<pubDate>2026-08-0${2 - i}T00:00:00Z</pubDate>
+<enclosure url="${origin}/audio-${i}.mp3" length="12345" type="audio/mpeg"/>
+</item>`,
+  ).join("\n");
+  return `<?xml version="1.0" encoding="UTF-8"?>
+<rss version="2.0"><channel>
+<title>Delisted Show</title>
+<description>Feedless stub test</description>
+${items}
+</channel></rss>`;
+}
+
+let server: ReturnType<typeof Bun.serve> | null = null;
+let feedUrl = "";
+let pageUrl = "";
+
+beforeAll(() => {
+  server = Bun.serve({
+    port: 0,
+    fetch(req) {
+      const url = new URL(req.url);
+      if (url.pathname.endsWith(".rss")) {
+        return new Response(feedXml(url.origin), {
+          headers: { "Content-Type": "application/rss+xml" },
+        });
+      }
+      if (url.pathname.startsWith("/show")) {
+        return new Response(pageHtml(feedUrl), {
+          headers: { "Content-Type": "text/html" },
+        });
+      }
+      return new Response("not found", { status: 404 });
+    },
+  });
+  feedUrl = `http://127.0.0.1:${server!.port}/feed.rss`;
+  // The resolver anchors on `/id<digits>` in the directory URL.
+  pageUrl = `http://127.0.0.1:${server!.port}/show/id${FEED_ID}`;
+});
+
+afterAll(() => {
+  server?.stop(true);
+  rmSync(configHome, { recursive: true, force: true });
+});
+
+const makeStub = (feedUrl: string, directoryUrl?: string): Podcast => ({
+  id: "itunes-12345",
+  title: "Delisted Show",
+  description: "Show that left the directory",
+  author: "Some Network",
+  feedUrl,
+  directoryUrl,
+  lastUpdated: new Date(),
+  isSubscribed: false,
+});
+
+test("addFeed resolves a feedless stub's feed from its directory page", async () => {
+  const store = useFeedStore();
+  const feed = await store.addFeed(makeStub("", pageUrl), "itunes");
+  expect(feed).not.toBeNull();
+  expect(feed!.podcast.feedUrl).toBe(feedUrl);
+  // Resolution metadata is dropped from the persisted feed record.
+  expect(feed!.podcast.directoryUrl).toBeUndefined();
+  expect(feed!.episodes.map((e) => e.title)).toEqual(EPISODE_TITLES);
+  // Remove the feed: bun test shares the store singleton across files, and a
+  // leftover feed (whose server dies in afterAll) would reorder other files'
+  // refresh assertions.
+  store.removeFeed(feed!.id);
+});
+
+test("addFeed refuses a stub whose directory page cannot be resolved", async () => {
+  const store = useFeedStore();
+  const unreachable = makeStub("", "http://127.0.0.1:1/nope/id999");
+  const feed = await store.addFeed(unreachable, "itunes");
+  expect(feed).toBeNull();
+});
+
+test("addFeed refuses a stub with no directory page at all", async () => {
+  const store = useFeedStore();
+  const bare = makeStub("", undefined);
+  const feed = await store.addFeed(bare, "itunes");
+  expect(feed).toBeNull();
+});
--- a/tests/itunes-feed-resolver.test.ts
+++ b/tests/itunes-feed-resolver.test.ts
@@ -0,0 +1,70 @@
+/**
+ * Feed-resolution extraction tests.
+ *
+ * The iTunes Search API returns feedUrl null for shows delisted from Apple
+ * Podcasts. Their public Apple page still embeds the real feed URL in JSON
+ * state — alongside feedUrls of RELATED shows — so extraction must anchor on
+ * the show's adamId rather than grabbing the first feedUrl in the document.
+ */
+import { test, expect } from "bun:test";
+import { extractFeedUrlFromPage } from "../src/utils/itunes-feed-resolver";
+
+const MAIN_FEED = "https://rss.pdrl.fm/b32227/feeds.megaphone.fm/BVDWV5370667266";
+const OTHER_FEED = "https://feeds.megaphone.fm/BVDWV7762869899";
+
+/** Synthetic Apple page: related shows first, main showOffer after. */
+const pageWithNoise = `{
+  "shows":[{"showOffer":{"title":"The Matt Walsh Show","adamId":"2950206264","feedUrl":"${OTHER_FEED}","showType":"episodic"}}],
+  "pageData":{"showOffer":{"title":"The Ben Shapiro Show","adamId":"1047335260","feedUrl":"${MAIN_FEED}","showType":"episodic"}}
+}`;
+
+test("extracts the show's feed anchored on its adamId, ignoring related shows", () => {
+  const feed = extractFeedUrlFromPage(
+    pageWithNoise,
+    "https://podcasts.apple.com/us/podcast/the-ben-shapiro-show/id1047335260",
+  );
+  expect(feed).toBe(MAIN_FEED);
+});
+
+test("handles the ?uo=4 suffix Apple appends to directory URLs", () => {
+  const feed = extractFeedUrlFromPage(
+    pageWithNoise,
+    "https://podcasts.apple.com/us/podcast/the-ben-shapiro-show/id1047335260?uo=4",
+  );
+  expect(feed).toBe(MAIN_FEED);
+});
+
+test("returns null when the page has no feedUrl for the requested id", () => {
+  const feed = extractFeedUrlFromPage(
+    pageWithNoise,
+    "https://podcasts.apple.com/us/podcast/some-other-show/id9999999999",
+  );
+  expect(feed).toBeNull();
+});
+
+test("finds the feed when the showOffer sits far after the adamId reference", () => {
+  // Apple serves page variants where thousands of chars separate the first
+  // adamId reference from the showOffer block carrying the feedUrl.
+  const variant = `{"adamId":"1047335260","$kind":"ShowPageIntent"}${"x".repeat(6000)}{"showOffer":{"title":"The Ben Shapiro Show","adamId":"1047335260","feedUrl":"${MAIN_FEED}"}}`;
+  const feed = extractFeedUrlFromPage(
+    variant,
+    "https://podcasts.apple.com/us/podcast/the-ben-shapiro-show/id1047335260",
+  );
+  expect(feed).toBe(MAIN_FEED);
+});
+
+test("falls back to the first feedUrl when the URL carries no id", () => {
+  const feed = extractFeedUrlFromPage(
+    pageWithNoise,
+    "https://podcasts.apple.com/us/podcast/the-ben-shapiro-show",
+  );
+  expect(feed).toBe(OTHER_FEED);
+});
+
+test("returns null when the page contains no feedUrl at all", () => {
+  const feed = extractFeedUrlFromPage(
+    "<html><body>not found</body></html>",
+    "https://podcasts.apple.com/us/podcast/the-ben-shapiro-show/id1047335260",
+  );
+  expect(feed).toBeNull();
+});
--- a/tests/source-searcher.test.ts
+++ b/tests/source-searcher.test.ts
@@ -9,7 +9,7 @@
 * search. This test pins the empty-result contract.
 */
 import { test, expect } from "bun:test";
-import { searchSourceByType } from "../src/utils/source-searcher";
+import { searchSourceByType, mapItunesResult } from "../src/utils/source-searcher";
 import { SourceType } from "../src/types/source";
 import type { PodcastSource } from "../src/types/source";

@@ -29,6 +29,16 @@ const customSource: PodcastSource = {
  enabled: true,
 };

+const itunesSource: PodcastSource = {
+  id: "itunes",
+  name: "Apple Podcasts",
+  type: SourceType.API,
+  baseUrl: "https://itunes.apple.com/search",
+  enabled: true,
+  country: "US",
+  language: "en_us",
+};
+
 test("RSS sources return no directory search results", async () => {
  const results = await searchSourceByType("blocked and reported", rssSource);
  expect(results).toEqual([]);
@@ -38,3 +48,54 @@ test("custom sources return no directory search results", async () => {
  const results = await searchSourceByType("anything", customSource);
  expect(results).toEqual([]);
 });
+
+// ── iTunes stub records (delisted shows) ────────────────────────────────────
+// Shows that left Apple Podcasts (e.g. The Daily Wire's in 2021) remain in
+// the directory as metadata-only records with feedUrl null. They must stay
+// findable — earlier they were dropped entirely, so "ben shapiro" surfaced
+// nothing while the show is the #1 iTunes hit.
+
+test("iTunes results without a feedUrl (delisted shows) are kept, not dropped", () => {
+  const stub = mapItunesResult(
+    {
+      collectionId: 1047335260,
+      collectionName: "The Ben Shapiro Show",
+      artistName: "The Daily Wire",
+      feedUrl: null,
+      collectionViewUrl:
+        "https://podcasts.apple.com/us/podcast/the-ben-shapiro-show/id1047335260",
+    },
+    itunesSource,
+  );
+  expect(stub).not.toBeNull();
+  expect(stub!.title).toBe("The Ben Shapiro Show");
+  // Empty feed marks "unavailable from this directory"; the Apple page URL
+  // is carried for feed resolution at subscribe time.
+  expect(stub!.feedUrl).toBe("");
+  expect(stub!.directoryUrl).toBe(
+    "https://podcasts.apple.com/us/podcast/the-ben-shapiro-show/id1047335260",
+  );
+});
+
+test("iTunes results with a feedUrl keep it and carry no directory fallback", () => {
+  const normal = mapItunesResult(
+    {
+      collectionId: 1487234816,
+      collectionName: "Morning Wire",
+      artistName: "The Daily Wire",
+      feedUrl: "https://feeds.megaphone.fm/BVDWV8747925072",
+    },
+    itunesSource,
+  );
+  expect(normal).not.toBeNull();
+  expect(normal!.feedUrl).toBe("https://feeds.megaphone.fm/BVDWV8747925072");
+  expect(normal!.directoryUrl).toBeUndefined();
+});
+
+test("iTunes results without a collection name stay dropped", () => {
+  const dropped = mapItunesResult(
+    { collectionId: 1, feedUrl: "https://example.com/feed.xml" },
+    itunesSource,
+  );
+  expect(dropped).toBeNull();
+});