This test is failing because PurgeCachedWebViewPages() returned before the navigation manager finished restoring session. So when NavigationManager::CanGoForward() is called, |is_restore_session_in_progress_| is still true.

I prototyped a fix by waiting for the native controller of NTP to show up before calling |goForward|: https://chromium-review.googlesource.com/c/chromium/src/+/1252850

This fixes the test which shows that this is indeed a timing issue. But it's not a great fix because the chrome_test_util::WaitForPageToFinishLoading() in PurgeCachedWebViewPages() should have been sufficient.

I'll look more.

I wonder if the load ever starts and finishes inside PurgeCachedWebViewPages?

This issue is marked as a release blocker with no milestone associated. Please add an appropriate milestone.

All release blocking issues should have milestones associated to it, so that the issue can tracked and the fixes can be pushed promptly.

Thanks for your time! To disable nags, add the Disable-Nags label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Re Comment #4: Actually no. PurgeCachedWebViewPages() returns before loading starts with slim nav. I think because slim nav restores history via a web view load, the loading event starts (and stops) in the next run loop cycle. In comparison, the start loading event for LegacyNavigationManager happens in the same run loop cycle, because the load is a native load.

Perhaps we should always spin the run loop once in chrome_test_util::WaitForPageToFinishLoading().

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/90cb72b610ecd6467bc1f565b7bdfb8b76aa2826

commit 90cb72b610ecd6467bc1f565b7bdfb8b76aa2826
Author: Danyao Wang <danyao@google.com>
Date: Fri Nov 02 23:27:38 2018

[Nav Experiment] Support async loading in PurgeCachedWebViewPages().

PurgeCachedWebViews() restores session history and wait for the first
navigation after restore to stop loading before returning. However, if
the first loading event happens asynchronously in the web view,
PurgeCachedWebViews() would return before session restore is finished.
This CL changes it to wait for the navigation finish event to handle
this situation.

This fixes NavigationTestCase//testRestoreHistoryToNTPAndNavigateForward
when #slim-navigation-manager is used.

Bug:  890385 
Cq-Include-Trybots: luci.chromium.try:ios-simulator-cronet;luci.chromium.try:ios-simulator-full-configs
Change-Id: I86913ed64ac0c6982fae48b71f0561f071b4db30
Reviewed-on: https://chromium-review.googlesource.com/c/1252850
Commit-Queue: Danyao Wang <danyao@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Cr-Commit-Position: refs/heads/master@{#605091}
[modify] https://crrev.com/90cb72b610ecd6467bc1f565b7bdfb8b76aa2826/ios/chrome/test/app/chrome_test_util.mm