[ios] Make sure that |didFinishNavigation| is signalled for fast goBack.

ios/web/web_state/ui/crw_wk_web_view_web_controller.mm

Index: ios/web/web_state/ui/crw_wk_web_view_web_controller.mm
diff --git a/ios/web/web_state/ui/crw_wk_web_view_web_controller.mm b/ios/web/web_state/ui/crw_wk_web_view_web_controller.mm
index 2c2b1e320076d2a769b891cf9fb8d4b848cc3cbd..1c00628d9853f09edfedb5e09488509fc9d84318 100644
--- a/ios/web/web_state/ui/crw_wk_web_view_web_controller.mm
+++ b/ios/web/web_state/ui/crw_wk_web_view_web_controller.mm
@@ -1616,6 +1616,12 @@ WKWebViewErrorSource WKWebViewErrorSourceFromError(NSError* error) {
     [self addActivityIndicatorTask];
   } else {
     [self clearActivityIndicatorTasks];
+    if ([self currentBackForwardListItemHolder]->navigation_type() ==
+        WKNavigationTypeBackForward) {
+      // A fast back/forward may not call |webView:didFinishNavigation:|, so

[Jackie Quinn, 2015/12/22 23:41:35]

Is is "may not" or "will not"? Any chance we'd run into timing issues here if
didFinishNavigation is called too early? I know calling didFinishNavigation
multiple times is fine, just wondering if calling it early may mess something
up. 

[Eugene But (OOO till 7-30), 2015/12/22 23:56:29]

It is actually "may not" (I think depending on caching). I don't see how calling
|didFinishNavigation| can mess up things, but I can't prove that :(

This place is the last chance to call |didFinishNavigation| for BF navigation
and should probably be done here anyways.

[Jackie Quinn, 2015/12/23 00:13:03]

That was my suspicion... So long as it's not called before didCommit, I think
it's fine, and I don't think there's any way for it to be called before
didCommit. 

[Eugene But (OOO till 7-30), 2015/12/23 00:55:57]

You are right it is not called before |didCommit|.

+      // finishing the navigation should be signalled explicitely.

[Jackie Quinn, 2015/12/22 23:41:35]

s/explicitely/explicitly

[Eugene But (OOO till 7-30), 2015/12/22 23:56:29]

Done.

+      [self didFinishNavigation];
+    }
   }
 }