Refactor away most of ExtensionRendererInfo

chrome/renderer/extensions/chrome_app_bindings.cc

Index: chrome/renderer/extensions/chrome_app_bindings.cc
diff --git a/chrome/renderer/extensions/chrome_app_bindings.cc b/chrome/renderer/extensions/chrome_app_bindings.cc
index 647f9d5ad6124f9826e76b7b81be81ba797c764a..82f38d9caa937f224fe1235ac8f47868f490787f 100644
--- a/chrome/renderer/extensions/chrome_app_bindings.cc
+++ b/chrome/renderer/extensions/chrome_app_bindings.cc
@@ -57,7 +57,8 @@ class ChromeAppExtensionWrapper : public v8::Extension {
         !(url.SchemeIs("http") || url.SchemeIs("https")))
       return v8::Boolean::New(false);
 
-    bool has_web_extent = (ExtensionRendererInfo::GetByURL(url) != NULL);
+    bool has_web_extent =
+        RenderThread::current()->extensions()->GetByURL(url) != NULL;

[Matt Perry, 2011/01/24 20:08:34]

note that RenderThread::current() will be NULL in unit tests.

[Aaron Boodman, 2011/01/25 00:27:33]

Really? RenderThread::current() is backed by lazy TLS. It seems like it will
just create an instance.

[Matt Perry, 2011/01/25 00:45:54]

Oh, well maybe things have changed since I last looked. If unit tests don't
crash, then I guess this is fine.

     return v8::Boolean::New(has_web_extent);
   }