castle-search: read Google results again by resolving its /goto links

Google now wraps every result as google.com/goto?url=<opaque token>. The
self-host filter discarded all of them, so every Google search failed with
'nothing could be parsed'. Accept /goto, resolve it node-side from the
302 Location, and when Google answers its 'Verifying your request'
interstitial instead, keep the wrapped link and put the displayed
breadcrumb in the snippet.

1 files changed, 11 insertions, 2 deletions

diff --git a/castle-search/src/extract.ts b/castle-search/src/extract.ts
index 8b5c57f..653942b 100644
--- a/castle-search/src/extract.ts
+++ b/castle-search/src/extract.ts
@@ -35,7 +35,9 @@
  *   - DuckDuckGo: `//duckduckgo.com/l/?uddg=<percent-encoded target>`
  *   - Bing:       `//bing.com/ck/a?…&u=a1<base64url of target>`
  *
- * Google and Brave link straight out and need no unwrapping.
+ * Brave links straight out. Google wraps results in `/goto?url=<opaque token>`
+ * that only Google can resolve; those are left alone here and resolved
+ * node-side by resolve.ts.
  */
 
 import type { ExtractionConfig } from "./engines.ts";
@@ -45,6 +47,8 @@ export type SearchResult = {
   title: string;
   url: string;
   snippet: string;
+  /** Google only: the site and breadcrumb as displayed, kept for links that cannot be resolved. */
+  display?: string;
 };
 
 /** What the page script returns in a single round trip. */
@@ -261,8 +265,13 @@ export const buildProbeScript = (config: ExtractionConfig): string => String.raw
         ).replace(/Read more$/, "").trim();
       }
 
+      var display = "";
+      var siteEl = anchor.querySelector("span.VuuXrf");
+      var citeEl = anchor.querySelector("cite");
+      display = norm([siteEl && siteEl.innerText, citeEl && citeEl.innerText].filter(Boolean).join(" — "));
+
       seen[url] = true;
-      out.push({ title: title, url: url, snippet: snippet.slice(0, 600) });
+      out.push({ title: title, url: url, snippet: snippet.slice(0, 600), display: display });
     }
     return out;
   };