Fix adopting a video element with controls

Source/core/html/HTMLMediaElement.cpp

 /*
  * Copyright (C) 2007, 2008, 2009, 2010, 2011, 2012, 2013 Apple Inc. All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
  * are met:
  * 1. Redistributions of source code must retain the above copyright
  *    notice, this list of conditions and the following disclaimer.
  * 2. Redistributions in binary form must reproduce the above copyright
  *    notice, this list of conditions and the following disclaimer in the
@@ skipping 809 matching lines @@
 
     // FIXME: Investigate whether these can be moved into m_networkState != NETWORK_EMPTY block above
     // so they are closer to the relevant spec steps.
     m_lastSeekTime = 0;
     m_duration = std::numeric_limits<double>::quiet_NaN();
 
     // The spec doesn't say to block the load event until we actually run the asynchronous section
     // algorithm, but do it now because we won't start that until after the timer fires and the
     // event may have already fired by then.
     setShouldDelayLoadEvent(true);
-
-    configureMediaControls();
 }
 
 void HTMLMediaElement::loadInternal()
 {
+    configureMediaControls();

[philipj_slow, 2014/09/16 11:59:39]

All existing tests and the one you've added pass if one just removes
configureMediaControls() above but does not add this one, i.e. this line is not
really tested.

A smaller fix is to replace configureMediaControls() with if
(hasMediaControls()) mediaControls()->reset(). AFAICT, the purpose of
configureMediaControls() in prepareForLoad() is to update the controls since the
currentTime, duration and other things may have changed, and if there are no
controls there's no point in creating them at this point I believe. When moving
to a new document, HTMLMediaElement::didNotifySubtreeInsertionsToDocument() is
going to call configureMediaControls() and otherwise parseAttribute() should do
it.

I suggest the more conservative fix, along with a test that would fail if the
code were removed entirely, to prove that it's actually needed there.

[aberent, 2014/09/16 14:45:40]

Having failed to create a test in a bit more than your suggested 30 minutes I
will go for the conservative fix without the test.

     // HTMLMediaElement::textTracksAreReady will need "... the text tracks whose mode was not in the
     // disabled state when the element's resource selection algorithm last started".
     m_textTracksWhenResourceSelectionBegan.clear();
     if (m_textTracks) {
         for (unsigned i = 0; i < m_textTracks->length(); ++i) {
             TextTrack* track = m_textTracks->item(i);
             if (track->mode() != TextTrack::disabledKeyword())
                 m_textTracksWhenResourceSelectionBegan.append(track);
         }
     }
@@ skipping 3137 matching lines @@
 
 #if ENABLE(WEB_AUDIO)
 void HTMLMediaElement::clearWeakMembers(Visitor* visitor)
 {
     if (!visitor->isAlive(m_audioSourceNode) && audioSourceProvider())
         audioSourceProvider()->setClient(0);
 }
 #endif
 
 }