castle-search: resolve Google links lazily, in Chrome, not eagerly over plain HTTPHEADmaster

The Node-side resolver sent one curl-like request per result with no
cookies, JS or browser; Google answers that with its 'Verifying your
request' countdown after a handful, and the testing itself triggered it.
A click or direct navigation in Chrome redirects at once (measured).

Search now leaves Google links wrapped and prints the breadcrumb Google
shows for each. castle_cdp_resolve follows one wrapped link in a Chrome tab
when the agent picks it, accepts only google.com/goto?url=<token>, and
reports the verification page instead of waiting it out.

2 files changed, 41 insertions, 50 deletions

diff --git a/castle-search/test/google-links.test.ts b/castle-search/test/google-links.test.ts
new file mode 100644
index 0000000..f31f112
--- /dev/null
+++ b/castle-search/test/google-links.test.ts
@@ -0,0 +1,41 @@
+import assert from "node:assert/strict";
+import { test } from "node:test";
+
+import { formatResults } from "../src/format.ts";
+import { isGoogleGoto } from "../src/google-links.ts";
+import { getEngine } from "../src/engines.ts";
+import { resolveLink } from "../src/search.ts";
+
+test("only well-formed google /goto links count", () => {
+  assert.equal(isGoogleGoto("https://www.google.com/goto?url=CAESabc_-9"), true);
+  assert.equal(isGoogleGoto("https://www.google.co.uk/goto?url=CAESabc"), true);
+  assert.equal(isGoogleGoto("https://www.google.com/search?q=x"), false);
+  assert.equal(isGoogleGoto("https://example.com/goto?url=CAES"), false);
+  // Not a general navigator: nothing may ride along after the token.
+  assert.equal(isGoogleGoto("https://www.google.com/goto?url=CAES&x=https://evil.example"), false);
+  assert.equal(isGoogleGoto("https://www.google.com.evil.example/goto?url=CAES"), false);
+});
+
+test("the google self-host filter lets /goto through but still blocks google's own pages", () => {
+  const re = new RegExp(getEngine("google").extraction.selfHostPattern);
+  assert.equal(re.test("https://www.google.com/goto?url=CAES"), false);
+  assert.equal(re.test("https://www.google.com/search?q=x"), true);
+  assert.equal(re.test("https://accounts.google.com/ServiceLogin"), false);
+});
+
+test("resolveLink refuses anything that is not a google /goto link, before touching the browser", async () => {
+  await assert.rejects(() => resolveLink("https://example.com/", AbortSignal.timeout(1000)), /only follows Google result links/);
+  await assert.rejects(() => resolveLink("file:///etc/passwd", AbortSignal.timeout(1000)), /only follows Google result links/);
+});
+
+test("a wrapped result says so, shows Google's breadcrumb, and names the tool to resolve it", () => {
+  const text = formatResults({
+    query: "q", searchUrl: "u", finalUrl: "u", endpoint: "e", engine: "Google", engineSource: "s",
+    results: [
+      { title: "Wrapped", url: "https://www.google.com/goto?url=CAES", snippet: "snip", display: "GitHub — https://github.com › a › b" },
+      { title: "Direct", url: "https://example.com/", snippet: "" },
+    ],
+  });
+  assert.match(text, /\[Google-wrapped link, real URL not yet known\. Google shows it as: GitHub — https:\/\/github\.com › a › b\. Pass it to castle_cdp_resolve/);
+  assert.equal((text.match(/Google-wrapped/g) ?? []).length, 1);
+});
diff --git a/castle-search/test/resolve.test.ts b/castle-search/test/resolve.test.ts
deleted file mode 100644
index bcd4ef4..0000000
--- a/castle-search/test/resolve.test.ts
+++ /dev/null
@@ -1,50 +0,0 @@
-import assert from "node:assert/strict";
-import { test } from "node:test";
-
-import { isGoogleGoto, resolveGoogleLinks } from "../src/resolve.ts";
-import { getEngine } from "../src/engines.ts";
-
-test("google /goto links are recognised, other google links are not", () => {
-  assert.equal(isGoogleGoto("https://www.google.com/goto?url=CAES"), true);
-  assert.equal(isGoogleGoto("https://www.google.co.uk/goto?url=CAES"), true);
-  assert.equal(isGoogleGoto("https://www.google.com/search?q=x"), false);
-  assert.equal(isGoogleGoto("https://example.com/goto?url=x"), false);
-});
-
-test("the google self-host filter lets /goto through but still blocks google's own pages", () => {
-  const re = new RegExp(getEngine("google").extraction.selfHostPattern);
-  assert.equal(re.test("https://www.google.com/goto?url=CAES"), false);
-  assert.equal(re.test("https://www.google.com/search?q=x"), true);
-  assert.equal(re.test("https://accounts.google.com/ServiceLogin"), false);
-});
-
-test("non-google results pass through untouched", async () => {
-  const input = [{ title: "t", url: "https://example.com/", snippet: "" }];
-  assert.deepEqual(await resolveGoogleLinks(input), input);
-});
-
-test("a link that cannot be resolved keeps the page's breadcrumb in its snippet", async () => {
-  const wrapped = "https://www.google.com/goto?url=CAES";
-  const real = globalThis.fetch;
-  // Google's interstitial: 200 with no Location header.
-  globalThis.fetch = (async () => new Response("<html>Verifying your request</html>", { status: 200 })) as typeof fetch;
-  try {
-    const [r] = await resolveGoogleLinks([{ title: "t", url: wrapped, snippet: "s", display: "GitHub — https://github.com › a › b" }]);
-    assert.equal(r?.url, wrapped);
-    assert.match(r?.snippet ?? "", /^\[link unresolved; Google shows it as: GitHub — https:\/\/github\.com › a › b\] s$/);
-  } finally {
-    globalThis.fetch = real;
-  }
-});
-
-test("a link that resolves is replaced and the snippet is left alone", async () => {
-  const real = globalThis.fetch;
-  globalThis.fetch = (async () => new Response(null, { status: 302, headers: { location: "https://github.com/a/b" } })) as typeof fetch;
-  try {
-    const [r] = await resolveGoogleLinks([{ title: "t", url: "https://www.google.com/goto?url=CAES", snippet: "s", display: "x" }]);
-    assert.equal(r?.url, "https://github.com/a/b");
-    assert.equal(r?.snippet, "s");
-  } finally {
-    globalThis.fetch = real;
-  }
-});