102801 2012-11-20 04:42:41 -0800 [WK2] Viewport meta tag broken after r134801 2012-12-03 06:51:48 -0800 1 1 1 Unclassified WebKit WebKit Qt 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 102971 103889 1 zeno mikhail.pozdnyakov abecsi kenneth mikhail.pozdnyakov tmpsantos oldest_to_newest 771750 0 zeno 2012-11-20 04:42:41 -0800 Loading any page that makes use of the viewport meta tag causes flickering and misplaced textures, when resizing the window. In some cases textures are also being displayed upside down, right after loading the page. (without any resizing) This might be an initialisation issue with a projection matrix or similar. 771809 1 kenneth 2012-11-20 06:30:44 -0800 How did you come to blame this commit? I don't see how it can affect this and everything works perfectly on EFL. Do you have some example pages that triggers this issue, and maybe screenshots? 771810 2 tmpsantos 2012-11-20 06:31:29 -0800 Do you have a test page? Works fine here: http://people.opera.com/andreasb/viewport/ex02.html 771889 3 zeno 2012-11-20 08:22:07 -0800 (In reply to comment #1) > How did you come to blame this commit? I don't see how it can affect this and everything works perfectly on EFL. Do you have some example pages that triggers this issue, and maybe screenshots? Git bisect. 771890 4 zeno 2012-11-20 08:22:29 -0800 (In reply to comment #2) > Do you have a test page? Works fine here: http://people.opera.com/andreasb/viewport/ex02.html http://www.albisser.org/test/viewport.html http://alteredqualia.com/three/examples/webgl_pasta.html 771906 5 tmpsantos 2012-11-20 08:52:12 -0800 (In reply to comment #4) > (In reply to comment #2) > > Do you have a test page? Works fine here: http://people.opera.com/andreasb/viewport/ex02.html > > http://www.albisser.org/test/viewport.html Indeed. This site is broken on Qt and EFL. 771975 6 175242 mikhail.pozdnyakov 2012-11-20 10:26:30 -0800 Created attachment 175242 preliminary patch. 771987 7 175242 kenneth 2012-11-20 10:33:54 -0800 Comment on attachment 175242 preliminary patch. View in context: https://bugs.webkit.org/attachment.cgi?id=175242&action=review > Source/WebKit2/ChangeLog:11 > + Problem turned up as after m_rawAttributes.initialScale was set to '-1', the function > + WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable(); was invoked from > + PageViewportController::didChangeViewportAttributes resetting also min and max scale > + factors to '-1'. Shouldn't we remove that call instead? 771992 8 abecsi 2012-11-20 10:41:30 -0800 (In reply to comment #7) > (From update of attachment 175242 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=175242&action=review > > > Source/WebKit2/ChangeLog:11 > > + Problem turned up as after m_rawAttributes.initialScale was set to '-1', the function > > + WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable(); was invoked from > > + PageViewportController::didChangeViewportAttributes resetting also min and max scale > > + factors to '-1'. > > Shouldn't we remove that call instead? No, that call is needed. I have a complete patch to fix this issue, uploading it soon. 772047 9 abecsi 2012-11-20 11:33:41 -0800 The issue is actually a bit more complicated than I thought. The main problem with leaving the initial scale -1 is that WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable is called by almost all ports (qt, efl, gtk, chromium) and it seems to me, that all callers assume that the scale values in the provided attributes are valid. This was also a reason why I previously introduced a bool flag for this case. AFAICT the solution for this is one of the following: - we change all the call sides to calculate the initial scale if it is <=0 and assert in restrictScaleFactorToInitialScaleIfNotUserScalable for initialScale - we add a fallbackScale parameter to restrictScaleFactorToInitialScaleIfNotUserScalable and use it if initialScale <= 0 - we revert r134801 I think adding a new parameter (second solution) is the proper fix for this. Kenneth, am I right here? 772732 10 kenneth 2012-11-21 03:10:40 -0800 (In reply to comment #9) > The issue is actually a bit more complicated than I thought. > > The main problem with leaving the initial scale -1 is that WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable is called by almost all ports (qt, efl, gtk, chromium) and it seems to me, that all callers assume that the scale values in the provided attributes are valid. This was also a reason why I previously introduced a bool flag for this case. > > AFAICT the solution for this is one of the following: > > - we change all the call sides to calculate the initial scale if it is <=0 and assert in restrictScaleFactorToInitialScaleIfNotUserScalable for initialScale > > - we add a fallbackScale parameter to restrictScaleFactorToInitialScaleIfNotUserScalable and use it if initialScale <= 0 > > - we revert r134801 > > I think adding a new parameter (second solution) is the proper fix for this. > > Kenneth, am I right here? I would also go for number two. Basically restricting the min and max scale factors to the initial one is not part of the viewport meta algorithm, but more of an implementation detail. Maybe just restrict to the minimum scale value then? 772765 11 mikhail.pozdnyakov 2012-11-21 03:37:08 -0800 (In reply to comment #10) > (In reply to comment #9) > > The issue is actually a bit more complicated than I thought. > > > > The main problem with leaving the initial scale -1 is that WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable is called by almost all ports (qt, efl, gtk, chromium) and it seems to me, that all callers assume that the scale values in the provided attributes are valid. This was also a reason why I previously introduced a bool flag for this case. > > > > AFAICT the solution for this is one of the following: > > > > - we change all the call sides to calculate the initial scale if it is <=0 and assert in restrictScaleFactorToInitialScaleIfNotUserScalable for initialScale > > > > - we add a fallbackScale parameter to restrictScaleFactorToInitialScaleIfNotUserScalable and use it if initialScale <= 0 > > > > - we revert r134801 > > > > I think adding a new parameter (second solution) is the proper fix for this. > > > > Kenneth, am I right here? > > I would also go for number two. Basically restricting the min and max scale factors to the initial one is not part of the viewport meta algorithm, but more of an implementation detail. > > Maybe just restrict to the minimum scale value then? Also would like to add that initial scale is set to '-1' only for WK2 at the moment, think the same approach should be spread for all ports. 772766 12 kenneth 2012-11-21 03:40:06 -0800 (In reply to comment #11) > (In reply to comment #10) > > (In reply to comment #9) > > > The issue is actually a bit more complicated than I thought. > > > > > > The main problem with leaving the initial scale -1 is that WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable is called by almost all ports (qt, efl, gtk, chromium) and it seems to me, that all callers assume that the scale values in the provided attributes are valid. This was also a reason why I previously introduced a bool flag for this case. > > > > > > AFAICT the solution for this is one of the following: > > > > > > - we change all the call sides to calculate the initial scale if it is <=0 and assert in restrictScaleFactorToInitialScaleIfNotUserScalable for initialScale > > > > > > - we add a fallbackScale parameter to restrictScaleFactorToInitialScaleIfNotUserScalable and use it if initialScale <= 0 > > > > > > - we revert r134801 > > > > > > I think adding a new parameter (second solution) is the proper fix for this. > > > > > > Kenneth, am I right here? > > > > I would also go for number two. Basically restricting the min and max scale factors to the initial one is not part of the viewport meta algorithm, but more of an implementation detail. > > > > Maybe just restrict to the minimum scale value then? > Also would like to add that initial scale is set to '-1' only for WK2 at the moment, think the same approach should be spread for all ports. There is outcommented code in ViewportArguments class to do this for everyone 772778 13 mikhail.pozdnyakov 2012-11-21 03:55:27 -0800 (In reply to comment #12) > (In reply to comment #11) > > (In reply to comment #10) > > > (In reply to comment #9) > > > > The issue is actually a bit more complicated than I thought. > > > > > > > > The main problem with leaving the initial scale -1 is that WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable is called by almost all ports (qt, efl, gtk, chromium) and it seems to me, that all callers assume that the scale values in the provided attributes are valid. This was also a reason why I previously introduced a bool flag for this case. > > > > > > > > AFAICT the solution for this is one of the following: > > > > > > > > - we change all the call sides to calculate the initial scale if it is <=0 and assert in restrictScaleFactorToInitialScaleIfNotUserScalable for initialScale > > > > > > > > - we add a fallbackScale parameter to restrictScaleFactorToInitialScaleIfNotUserScalable and use it if initialScale <= 0 > > > > > > > > - we revert r134801 > > > > > > > > I think adding a new parameter (second solution) is the proper fix for this. > > > > > > > > Kenneth, am I right here? > > > > > > I would also go for number two. Basically restricting the min and max scale factors to the initial one is not part of the viewport meta algorithm, but more of an implementation detail. > > > > > > Maybe just restrict to the minimum scale value then? > > Also would like to add that initial scale is set to '-1' only for WK2 at the moment, think the same approach should be spread for all ports. > > There is outcommented code in ViewportArguments class to do this for everyone If we do not want to set initial scale to -1 for everyone hence we don't need to touch restrictScaleFactorToInitialScaleIfNotUserScalable, and my "preliminary patch" is acceptable, as it treats WK2 only which is only broken. 772782 14 175242 kenneth 2012-11-21 04:00:49 -0800 Comment on attachment 175242 preliminary patch. View in context: https://bugs.webkit.org/attachment.cgi?id=175242&action=review > Source/WebKit2/UIProcess/PageViewportController.cpp:220 > > m_rawAttributes = newAttributes; > + if (m_rawAttributes.initialScale < 0) > + m_rawAttributes.initialScale = m_minimumScaleToFit; > WebCore::restrictScaleFactorToInitialScaleIfNotUserScalable(m_rawAttributes); > > m_allowsUserScaling = !!m_rawAttributes.userScalable; Are you sure m_minimumScaleToFit is set correctly at this point? I believe this is called before contentSize is changed 772807 15 175413 abecsi 2012-11-21 04:46:56 -0800 Created attachment 175413 Patch 772812 16 175414 abecsi 2012-11-21 04:56:31 -0800 Created attachment 175414 Patch 772825 17 175414 mikhail.pozdnyakov 2012-11-21 05:24:41 -0800 Comment on attachment 175414 Patch isn't it worth adding layout tests? 772855 18 abecsi 2012-11-21 05:56:40 -0800 (In reply to comment #17) > (From update of attachment 175414 [details]) > isn't it worth adding layout tests? Layout tests do not use this code-path on Qt (fit-to-width), but extending the QML unit tests would be possible. 772882 19 kenneth 2012-11-21 06:28:59 -0800 (In reply to comment #18) > (In reply to comment #17) > > (From update of attachment 175414 [details] [details]) > > isn't it worth adding layout tests? > > Layout tests do not use this code-path on Qt (fit-to-width), but extending the QML unit tests would be possible. It should be possible to test that with layout tests now. Thiago Santos added support for that, though I am not sure the patch for Qt landed yet. 772906 20 tmpsantos 2012-11-21 06:44:12 -0800 (In reply to comment #19) > (In reply to comment #18) > > (In reply to comment #17) > > > (From update of attachment 175414 [details] [details] [details]) > > > isn't it worth adding layout tests? > > > > Layout tests do not use this code-path on Qt (fit-to-width), but extending the QML unit tests would be possible. > > It should be possible to test that with layout tests now. Thiago Santos added support for that, though I am not sure the patch for Qt landed yet. Landed on bug 102811. 772926 21 abecsi 2012-11-21 07:04:20 -0800 (In reply to comment #19) > (In reply to comment #18) > > (In reply to comment #17) > > > (From update of attachment 175414 [details] [details] [details]) > > > isn't it worth adding layout tests? > > > > Layout tests do not use this code-path on Qt (fit-to-width), but extending the QML unit tests would be possible. > > It should be possible to test that with layout tests now. Thiago Santos added support for that, though I am not sure the patch for Qt landed yet. Indeed, I missed that change. 772932 22 175414 abecsi 2012-11-21 07:08:49 -0800 Comment on attachment 175414 Patch Clearing flags on attachment: 175414 Committed r135399: <http://trac.webkit.org/changeset/135399> 772933 23 abecsi 2012-11-21 07:08:55 -0800 All reviewed patches have been landed. Closing bug. 772949 24 abecsi 2012-11-21 07:28:36 -0800 (In reply to comment #21) > (In reply to comment #19) > > (In reply to comment #18) > > > (In reply to comment #17) > > > > (From update of attachment 175414 [details] [details] [details] [details]) > > > > isn't it worth adding layout tests? > > > > > > Layout tests do not use this code-path on Qt (fit-to-width), but extending the QML unit tests would be possible. > > > > It should be possible to test that with layout tests now. Thiago Santos added support for that, though I am not sure the patch for Qt landed yet. > > Indeed, I missed that change. Since the preferred event handling behaviour for the flickable webview is not completely established yet, I think using fixed layout for layout testing is still somewhat unreliable on Qt, also because some initialization for Flickable depends on the onComponentComplete signal which is only emitted if the view was instantiated using QML. So if we add layout tests that use the feature we should be careful with the behaviour. As soon as we have free resources we're going to address these issues. 773167 25 tmpsantos 2012-11-21 12:10:56 -0800 This patch caused regression on the EFL bots: crash log for WebKitTestRunner (pid 4198): STDOUT: <empty> STDERR: ASSERTION FAILED: m_rawAttributes.initialScale > 0 STDERR: /home/buildslave-1/webkit-buildslave/efl-linux-64-debug-wk2/build/Source/WebKit2/UIProcess/PageViewportController.cpp(152) : void WebKit::PageViewportController::pageTransitionViewportReady() STDERR: 1 0x7f6af371abfa WebKit::PageViewportController::pageTransitionViewportReady() STDERR: 2 0x7f6af393e405 WebKit::PageClientDefaultImpl::pageTransitionViewportReady() STDERR: 3 0x7f6af37703d5 WebKit::WebPageProxy::pageTransitionViewportReady() STDERR: 4 0x7f6af397672a void CoreIPC::callMemberFunction<WebKit::WebPageProxy, void (WebKit::WebPageProxy::*)()>(CoreIPC::Arguments0 const&, WebKit::WebPageProxy*, void (WebKit::WebPageProxy::*)()) STDERR: 5 0x7f6af3972d36 void CoreIPC::handleMessage<Messages::WebPageProxy::PageTransitionViewportReady, WebKit::WebPageProxy, void (WebKit::WebPageProxy::*)()>(CoreIPC::MessageDecoder&, WebKit::WebPageProxy*, void (WebKit::WebPageProxy::*)()) STDERR: 6 0x7f6af396c149 WebKit::WebPageProxy::didReceiveWebPageProxyMessage(CoreIPC::Connection*, CoreIPC::MessageID, CoreIPC::MessageDecoder&) STDERR: 7 0x7f6af376b779 WebKit::WebPageProxy::didReceiveMessage(CoreIPC::Connection*, CoreIPC::MessageID, CoreIPC::MessageDecoder&) STDERR: 8 0x7f6af37ac895 WebKit::WebProcessProxy::didReceiveMessage(CoreIPC::Connection*, CoreIPC::MessageID, CoreIPC::MessageDecoder&) STDERR: 9 0x7f6af3694d04 CoreIPC::Connection::dispatchMessage(CoreIPC::MessageID, CoreIPC::MessageDecoder&) STDERR: 10 0x7f6af3694e70 CoreIPC::Connection::dispatchMessage(CoreIPC::Connection::Message<CoreIPC::MessageDecoder>&) STDERR: 11 0x7f6af36950bb CoreIPC::Connection::dispatchOneMessage() STDERR: 12 0x7f6af369f25c WTF::FunctionWrapper<void (CoreIPC::Connection::*)()>::operator()(CoreIPC::Connection*) STDERR: 13 0x7f6af369f062 WTF::BoundFunctionImpl<WTF::FunctionWrapper<void (CoreIPC::Connection::*)()>, void (CoreIPC::Connection*)>::operator()() STDERR: 14 0x7f6afa840dca WTF::Function<void ()>::operator()() const STDERR: 15 0x7f6af6d09baa WebCore::RunLoop::performWork() STDERR: 16 0x7f6af77274c7 WebCore::RunLoop::wakeUpEvent(void*, void*, unsigned int) STDERR: 17 0x7f6af2aa8621 STDERR: 18 0x7f6af2aa7571 STDERR: 19 0x7f6af2aa7ab7 ecore_main_loop_begin STDERR: 20 0x43362b WTR::TestController::platformRunUntil(bool&, double) STDERR: 21 0x41e94e WTR::TestController::runUntil(bool&, WTR::TestController::TimeoutDuration) STDERR: 22 0x41de6b WTR::TestController::resetStateToConsistentValues() STDERR: 23 0x425ac2 WTR::TestInvocation::invoke() STDERR: 24 0x41e686 WTR::TestController::runTest(char const*) STDERR: 25 0x41e7bf WTR::TestController::runTestingServerLoop() STDERR: 26 0x41e859 WTR::TestController::run() STDERR: 27 0x41c291 WTR::TestController::TestController(int, char const**) STDERR: 28 0x4337c6 main STDERR: 29 0x7f6af160876d __libc_start_main STDERR: 30 0x41ac09 STDERR: LEAK: 4 WebCoreNode 175242 2012-11-20 10:26:30 -0800 2012-11-21 04:46:51 -0800 preliminary patch. bug102801 text/plain 2545 mikhail.pozdnyakov Y29tbWl0IGEyYTliYTE1ZDA1ZGEyNzJmMWQ4ODQxMDBmZmI0ZjY5ZjVmNzkyYzUKQXV0aG9yOiBN aWtoYWlsIFBvemRueWFrb3YgPG1pa2hhaWwucG96ZG55YWtvdkBpbnRlbC5jb20+CkRhdGU6ICAg VHVlIE5vdiAyMCAyMDoyMDo1MiAyMDEyICswMjAwCgogICAgYnVnMTAyODAxCgpkaWZmIC0tZ2l0 IGEvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nIGIvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCmlu ZGV4IDQyMThjMWYuLmRmMjBmYjEgMTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJLaXQyL0NoYW5nZUxv ZworKysgYi9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cKQEAgLTEsMyArMSwxOSBAQAorMjAxMi0x MS0yMCAgTWlraGFpbCBQb3pkbnlha292ICA8bWlraGFpbC5wb3pkbnlha292QGludGVsLmNvbT4K KworICAgICAgICBbUXRdIFZpZXdwb3J0IG1ldGEgdGFnIGJyb2tlbiBhZnRlciByMTM0ODAxLgor ICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9MTAyODAxCisK KyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgUHJvYmxlbSB0 dXJuZWQgdXAgYXMgYWZ0ZXIgbV9yYXdBdHRyaWJ1dGVzLmluaXRpYWxTY2FsZSB3YXMgc2V0IHRv ICctMScsIHRoZSBmdW5jdGlvbgorICAgICAgICBXZWJDb3JlOjpyZXN0cmljdFNjYWxlRmFjdG9y VG9Jbml0aWFsU2NhbGVJZk5vdFVzZXJTY2FsYWJsZSgpOyB3YXMgaW52b2tlZCBmcm9tCisgICAg ICAgIFBhZ2VWaWV3cG9ydENvbnRyb2xsZXI6OmRpZENoYW5nZVZpZXdwb3J0QXR0cmlidXRlcyBy ZXNldHRpbmcgYWxzbyBtaW4gYW5kIG1heCBzY2FsZQorICAgICAgICBmYWN0b3JzIHRvICctMScu IAorCisgICAgICAgICogVUlQcm9jZXNzL1BhZ2VWaWV3cG9ydENvbnRyb2xsZXIuY3BwOgorICAg ICAgICAoV2ViS2l0OjpQYWdlVmlld3BvcnRDb250cm9sbGVyOjpwYWdlVHJhbnNpdGlvblZpZXdw b3J0UmVhZHkpOgorICAgICAgICAoV2ViS2l0OjpQYWdlVmlld3BvcnRDb250cm9sbGVyOjpkaWRD aGFuZ2VWaWV3cG9ydEF0dHJpYnV0ZXMpOgorCiAyMDEyLTExLTIwICBBbGxhbiBTYW5kZmVsZCBK ZW5zZW4gIDxhbGxhbi5qZW5zZW5AZGlnaWEuY29tPgogCiAgICAgICAgIFtRdF0gUGFnZVZpZXdw b3J0Q29udHJvbGxlckNsaWVudDo6ZGlkQ2hhbmdlVmlld3BvcnRBdHRyaWJ1dGVzIGNhbGxlZCB0 d2ljZQpkaWZmIC0tZ2l0IGEvU291cmNlL1dlYktpdDIvVUlQcm9jZXNzL1BhZ2VWaWV3cG9ydENv bnRyb2xsZXIuY3BwIGIvU291cmNlL1dlYktpdDIvVUlQcm9jZXNzL1BhZ2VWaWV3cG9ydENvbnRy b2xsZXIuY3BwCmluZGV4IDdjYTBhOTMuLmQwYmNiNmUgMTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJL aXQyL1VJUHJvY2Vzcy9QYWdlVmlld3BvcnRDb250cm9sbGVyLmNwcAorKysgYi9Tb3VyY2UvV2Vi S2l0Mi9VSVByb2Nlc3MvUGFnZVZpZXdwb3J0Q29udHJvbGxlci5jcHAKQEAgLTE0NSw4ICsxNDUs OCBAQCB2b2lkIFBhZ2VWaWV3cG9ydENvbnRyb2xsZXI6OnBhZ2VUcmFuc2l0aW9uVmlld3BvcnRS ZWFkeSgpCiB7CiAgICAgaWYgKCFtX3Jhd0F0dHJpYnV0ZXMubGF5b3V0U2l6ZS5pc0VtcHR5KCkp IHsKICAgICAgICAgbV9oYWRVc2VySW50ZXJhY3Rpb24gPSBmYWxzZTsKLSAgICAgICAgZmxvYXQg aW5pdGlhbFNjYWxlID0gKG1fcmF3QXR0cmlidXRlcy5pbml0aWFsU2NhbGUgPCAwKSA/IG1fbWlu aW11bVNjYWxlVG9GaXQgOiBtX3Jhd0F0dHJpYnV0ZXMuaW5pdGlhbFNjYWxlOwotICAgICAgICBh cHBseVNjYWxlQWZ0ZXJSZW5kZXJpbmdDb250ZW50cyhpbm5lckJvdW5kZWRWaWV3cG9ydFNjYWxl KHRvVmlld3BvcnRTY2FsZShpbml0aWFsU2NhbGUpKSk7CisgICAgICAgIEFTU0VSVChtX3Jhd0F0 dHJpYnV0ZXMuaW5pdGlhbFNjYWxlKTsKKyAgICAgICAgYXBwbHlTY2FsZUFmdGVyUmVuZGVyaW5n Q29udGVudHMoaW5uZXJCb3VuZGVkVmlld3BvcnRTY2FsZSh0b1ZpZXdwb3J0U2NhbGUobV9yYXdB dHRyaWJ1dGVzLmluaXRpYWxTY2FsZSkpKTsKICAgICB9CiAKICAgICAvLyBBdCB0aGlzIHBvaW50 IHdlIHNob3VsZCBhbHJlYWR5IGhhdmUgcmVjZWl2ZWQgdGhlIGZpcnN0IHZpZXdwb3J0IGFyZ3Vt ZW50cyBhbmQgdGhlIHJlcXVlc3RlZCBzY3JvbGwKQEAgLTIxMyw2ICsyMTMsOCBAQCB2b2lkIFBh Z2VWaWV3cG9ydENvbnRyb2xsZXI6OmRpZENoYW5nZVZpZXdwb3J0QXR0cmlidXRlcyhjb25zdCBX ZWJDb3JlOjpWaWV3cG9ydAogICAgICAgICByZXR1cm47CiAKICAgICBtX3Jhd0F0dHJpYnV0ZXMg PSBuZXdBdHRyaWJ1dGVzOworICAgIGlmIChtX3Jhd0F0dHJpYnV0ZXMuaW5pdGlhbFNjYWxlIDwg MCkKKyAgICAgICAgbV9yYXdBdHRyaWJ1dGVzLmluaXRpYWxTY2FsZSA9IG1fbWluaW11bVNjYWxl VG9GaXQ7CiAgICAgV2ViQ29yZTo6cmVzdHJpY3RTY2FsZUZhY3RvclRvSW5pdGlhbFNjYWxlSWZO b3RVc2VyU2NhbGFibGUobV9yYXdBdHRyaWJ1dGVzKTsKIAogICAgIG1fYWxsb3dzVXNlclNjYWxp bmcgPSAhIW1fcmF3QXR0cmlidXRlcy51c2VyU2NhbGFibGU7Cg== 175413 2012-11-21 04:46:56 -0800 2012-11-21 04:56:26 -0800 Patch bug-102801-20121121134418.patch text/plain 4708 abecsi U3VidmVyc2lvbiBSZXZpc2lvbjogMTM1Mjk1CmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0Mi9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cKaW5kZXggZGMxMmE5OTMwNzVkMWEy ODUyZmQxODMwMTQ4NWYwYmY5YzYxZjRhYy4uM2YxYjM2MTFkZGY4YmQwNmQ3YjcyNDY5NjFlYzQx ZjU5MzNlOGNmZiAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJLaXQyL0NoYW5nZUxvZwpAQCAtMSwzICsxLDIzIEBACisyMDEyLTExLTIxICBBbmRy YXMgQmVjc2kgIDxhbmRyYXMuYmVjc2lAZGlnaWEuY29tPgorCisgICAgICAgIFtXSzJdIFZpZXdw b3J0IG1ldGEgdGFnIGJyb2tlbiBhZnRlciByMTM0ODAxCisgICAgICAgIGh0dHBzOi8vYnVncy53 ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD0xMDI4MDEKKworICAgICAgICBSZXZpZXdlZCBieSBO T0JPRFkgKE9PUFMhKS4KKworICAgICAgICBJZiB0aGUgaW5pdGlhbC1zY2FsZSBhdHRyaWJ1dGUg aXMgbm90IGV4cGxpY2l0bHkgc3BlY2lmaWVkIGl0IGlzIGxlZnQgLTEgYnkgdGhlCisgICAgICAg IHZpZXdwb3J0IG1ldGEgYWxnb3JpdGhtIGFuZCBpZiB0aGUgY29udGVudCBpcyBub3Qgc2NhbGFi bGUgdGhlIG1pbiBhbmQgbWF4IHNjYWxlcworICAgICAgICBhcmUgcmVzdHJpY3RlZCBieSByZXN0 cmljdFNjYWxlRmFjdG9yVG9Jbml0aWFsU2NhbGVJZk5vdFVzZXJTY2FsYWJsZSgpIHdoaWNoIHJl c3VsdGVkCisgICAgICAgIGluIG5lZ2F0aXZlIHNjYWxlcyBvbiB0aGUgVUkgc2lkZS4KKyAgICAg ICAgTWFrZSBzdXJlIHRoYXQgdGhlIG1pbmltdW0gc2NhbGUgdG8gZml0IHRoZSB2aWV3cG9ydCBp cyBjYWxjdWxhdGVkIGJlZm9yZSByZXN0cmljdGluZworICAgICAgICB0aGUgc2NhbGVzIGFuZCB1 c2UgdGhpcyBzY2FsZSBhcyB0aGUgaW5pdGlhbCBzY2FsZSBpZiBubyBleHBsaWNpdCBpbml0aWFs IHNjYWxlIHdhcyBzZXQuCisKKyAgICAgICAgKiBVSVByb2Nlc3MvUGFnZVZpZXdwb3J0Q29udHJv bGxlci5jcHA6CisgICAgICAgIChXZWJLaXQ6OlBhZ2VWaWV3cG9ydENvbnRyb2xsZXI6OlBhZ2VW aWV3cG9ydENvbnRyb2xsZXIpOgorICAgICAgICAoV2ViS2l0OjpQYWdlVmlld3BvcnRDb250cm9s bGVyOjpwYWdlVHJhbnNpdGlvblZpZXdwb3J0UmVhZHkpOgorICAgICAgICAoV2ViS2l0OjpQYWdl Vmlld3BvcnRDb250cm9sbGVyOjpkaWRDaGFuZ2VWaWV3cG9ydEF0dHJpYnV0ZXMpOgorICAgICAg ICAoV2ViS2l0OjpQYWdlVmlld3BvcnRDb250cm9sbGVyOjp1cGRhdGVNaW5pbXVtU2NhbGVUb0Zp dCk6CisKIDIwMTItMTEtMjAgIEphbWVzIFNpbW9uc2VuICA8c2ltb25qYW1AY2hyb21pdW0ub3Jn PgogCiAgICAgICAgIENvbnNvbGlkYXRlIEZyYW1lTG9hZGVyOjpsb2FkKCkgaW50byBvbmUgZnVu Y3Rpb24gdGFraW5nIGEgRnJhbWVMb2FkUmVxdWVzdApkaWZmIC0tZ2l0IGEvU291cmNlL1dlYktp dDIvVUlQcm9jZXNzL1BhZ2VWaWV3cG9ydENvbnRyb2xsZXIuY3BwIGIvU291cmNlL1dlYktpdDIv VUlQcm9jZXNzL1BhZ2VWaWV3cG9ydENvbnRyb2xsZXIuY3BwCmluZGV4IDdjYTBhOTNiMmZlMjMx YjNlNTQ4YjFjNTE2Njc0YzczMjUxNGQ4NmIuLjM4MGE3ZGJkMTdlYTk5ODVhMzgwMTNkNjMwNjZm Yjc4ODY4MTRjMzUgMTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJLaXQyL1VJUHJvY2Vzcy9QYWdlVmll d3BvcnRDb250cm9sbGVyLmNwcAorKysgYi9Tb3VyY2UvV2ViS2l0Mi9VSVByb2Nlc3MvUGFnZVZp ZXdwb3J0Q29udHJvbGxlci5jcHAKQEAgLTUyLDExICs1MiwxNSBAQCBQYWdlVmlld3BvcnRDb250 cm9sbGVyOjpQYWdlVmlld3BvcnRDb250cm9sbGVyKFdlYktpdDo6V2ViUGFnZVByb3h5KiBwcm94 eSwgUGFnZQogewogICAgIC8vIEluaXRpYWxpemluZyBWaWV3cG9ydCBSYXcgQXR0cmlidXRlcyB0 byBhdm9pZCByYW5kb20gbmVnYXRpdmUgb3IgaW5maW5pdHkgc2NhbGUgZmFjdG9ycwogICAgIC8v IGlmIHRoZXJlIGlzIGEgcmFjZSBjb25kaXRpb24gYmV0d2VlbiB0aGUgZmlyc3QgbGF5b3V0IGFu ZCBzZXR0aW5nIHRoZSB2aWV3cG9ydCBhdHRyaWJ1dGVzIGZvciB0aGUgZmlyc3QgdGltZS4KLSAg ICBtX3Jhd0F0dHJpYnV0ZXMuaW5pdGlhbFNjYWxlID0gMTsKICAgICBtX3Jhd0F0dHJpYnV0ZXMu bWluaW11bVNjYWxlID0gMTsKICAgICBtX3Jhd0F0dHJpYnV0ZXMubWF4aW11bVNjYWxlID0gMTsK ICAgICBtX3Jhd0F0dHJpYnV0ZXMudXNlclNjYWxhYmxlID0gbV9hbGxvd3NVc2VyU2NhbGluZzsK IAorICAgIC8vIFRoZSBpbml0aWFsIHNjYWxlIG1pZ2h0IGJlIGltcGxpY2l0IGFuZCBzZXQgdG8g LTEsIGluIHRoaXMgY2FzZSB3ZSBoYXZlIHRvIGluZmVyIGl0CisgICAgLy8gdXNpbmcgdGhlIHZp ZXdwb3J0IHNpemUgYW5kIHRoZSBmaW5hbCBsYXlvdXQgc2l6ZS4KKyAgICAvLyBUbyBiZSBhYmxl IHRvIGFzc2VydCBmb3IgdmFsaWQgc2NhbGUgd2UgaW5pdGlhbGl6ZSBpdCB0byAtMS4KKyAgICBt X3Jhd0F0dHJpYnV0ZXMuaW5pdGlhbFNjYWxlID0gLTE7CisKICAgICBBU1NFUlQobV9jbGllbnQp OwogICAgIG1fY2xpZW50LT5zZXRDb250cm9sbGVyKHRoaXMpOwogfQpAQCAtMTQ1LDggKzE0OSw4 IEBAIHZvaWQgUGFnZVZpZXdwb3J0Q29udHJvbGxlcjo6cGFnZVRyYW5zaXRpb25WaWV3cG9ydFJl YWR5KCkKIHsKICAgICBpZiAoIW1fcmF3QXR0cmlidXRlcy5sYXlvdXRTaXplLmlzRW1wdHkoKSkg ewogICAgICAgICBtX2hhZFVzZXJJbnRlcmFjdGlvbiA9IGZhbHNlOwotICAgICAgICBmbG9hdCBp bml0aWFsU2NhbGUgPSAobV9yYXdBdHRyaWJ1dGVzLmluaXRpYWxTY2FsZSA8IDApID8gbV9taW5p bXVtU2NhbGVUb0ZpdCA6IG1fcmF3QXR0cmlidXRlcy5pbml0aWFsU2NhbGU7Ci0gICAgICAgIGFw cGx5U2NhbGVBZnRlclJlbmRlcmluZ0NvbnRlbnRzKGlubmVyQm91bmRlZFZpZXdwb3J0U2NhbGUo dG9WaWV3cG9ydFNjYWxlKGluaXRpYWxTY2FsZSkpKTsKKyAgICAgICAgQVNTRVJUKG1fcmF3QXR0 cmlidXRlcy5pbml0aWFsU2NhbGUgPiAwKTsKKyAgICAgICAgYXBwbHlTY2FsZUFmdGVyUmVuZGVy aW5nQ29udGVudHMoaW5uZXJCb3VuZGVkVmlld3BvcnRTY2FsZSh0b1ZpZXdwb3J0U2NhbGUobV9y YXdBdHRyaWJ1dGVzLmluaXRpYWxTY2FsZSkpKTsKICAgICB9CiAKICAgICAvLyBBdCB0aGlzIHBv aW50IHdlIHNob3VsZCBhbHJlYWR5IGhhdmUgcmVjZWl2ZWQgdGhlIGZpcnN0IHZpZXdwb3J0IGFy Z3VtZW50cyBhbmQgdGhlIHJlcXVlc3RlZCBzY3JvbGwKQEAgLTIxMywxMCArMjE3LDE1IEBAIHZv aWQgUGFnZVZpZXdwb3J0Q29udHJvbGxlcjo6ZGlkQ2hhbmdlVmlld3BvcnRBdHRyaWJ1dGVzKGNv bnN0IFdlYkNvcmU6OlZpZXdwb3J0CiAgICAgICAgIHJldHVybjsKIAogICAgIG1fcmF3QXR0cmli dXRlcyA9IG5ld0F0dHJpYnV0ZXM7Ci0gICAgV2ViQ29yZTo6cmVzdHJpY3RTY2FsZUZhY3RvclRv SW5pdGlhbFNjYWxlSWZOb3RVc2VyU2NhbGFibGUobV9yYXdBdHRyaWJ1dGVzKTsKLQogICAgIG1f YWxsb3dzVXNlclNjYWxpbmcgPSAhIW1fcmF3QXR0cmlidXRlcy51c2VyU2NhbGFibGU7Ci0gICAg dXBkYXRlTWluaW11bVNjYWxlVG9GaXQoKTsKKworICAgIC8vIFRoaXMgaXMgYWxzbyByZXNwb25z aWJsZSBmb3Igc2V0dGluZyB0aGUgaW5pdGlhbCBzY2FsZSBpZiBpdCBpcyBpbXBsaWNpdCAoZWcu IGF1dG8pLgorICAgIGlmICghdXBkYXRlTWluaW11bVNjYWxlVG9GaXQoKSkKKyAgICAgICAgcmV0 dXJuOworCisgICAgQVNTRVJUKG1fcmF3QXR0cmlidXRlcy5pbml0aWFsU2NhbGUgPiAwKTsKKwor ICAgIFdlYkNvcmU6OnJlc3RyaWN0U2NhbGVGYWN0b3JUb0luaXRpYWxTY2FsZUlmTm90VXNlclNj YWxhYmxlKG1fcmF3QXR0cmlidXRlcyk7CiAKICAgICBtX2NsaWVudC0+ZGlkQ2hhbmdlVmlld3Bv cnRBdHRyaWJ1dGVzKCk7CiB9CkBAIC0yNjIsMTEgKzI3MSwxNyBAQCB2b2lkIFBhZ2VWaWV3cG9y dENvbnRyb2xsZXI6OmFwcGx5UG9zaXRpb25BZnRlclJlbmRlcmluZ0NvbnRlbnRzKGNvbnN0IEZs b2F0UG9pbgogCiBib29sIFBhZ2VWaWV3cG9ydENvbnRyb2xsZXI6OnVwZGF0ZU1pbmltdW1TY2Fs ZVRvRml0KCkKIHsKLSAgICBpZiAobV92aWV3cG9ydFNpemUuaXNFbXB0eSgpKQorICAgIGlmICht X3ZpZXdwb3J0U2l6ZS5pc0VtcHR5KCkgfHwgbV9jb250ZW50c1NpemUuaXNFbXB0eSgpKQogICAg ICAgICByZXR1cm4gZmFsc2U7CiAKICAgICBmbG9hdCBtaW5pbXVtU2NhbGUgPSBXZWJDb3JlOjpj b21wdXRlTWluaW11bVNjYWxlRmFjdG9yRm9yQ29udGVudENvbnRhaW5lZChtX3Jhd0F0dHJpYnV0 ZXMsIFdlYkNvcmU6OnJvdW5kZWRJbnRTaXplKG1fdmlld3BvcnRTaXplKSwgV2ViQ29yZTo6cm91 bmRlZEludFNpemUobV9jb250ZW50c1NpemUpLCBkZXZpY2VQaXhlbFJhdGlvKCkpOwogCisgICAg aWYgKG1pbmltdW1TY2FsZSA8PSAwKQorICAgICAgICByZXR1cm4gZmFsc2U7CisKKyAgICBpZiAo bV9yYXdBdHRyaWJ1dGVzLmluaXRpYWxTY2FsZSA8IDApCisgICAgICAgIG1fcmF3QXR0cmlidXRl cy5pbml0aWFsU2NhbGUgPSBtaW5pbXVtU2NhbGU7CisKICAgICBpZiAoIWZ1enp5Q29tcGFyZSht aW5pbXVtU2NhbGUsIG1fbWluaW11bVNjYWxlVG9GaXQsIDAuMDAxKSkgewogICAgICAgICBtX21p bmltdW1TY2FsZVRvRml0ID0gbWluaW11bVNjYWxlOwogCg== 175414 2012-11-21 04:56:31 -0800 2012-11-21 07:08:49 -0800 Patch bug-102801-20121121135353.patch text/plain 4693 abecsi U3VidmVyc2lvbiBSZXZpc2lvbjogMTM1Mjk1CmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0Mi9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViS2l0Mi9DaGFuZ2VMb2cKaW5kZXggZGMxMmE5OTMwNzVkMWEy ODUyZmQxODMwMTQ4NWYwYmY5YzYxZjRhYy4uM2YxYjM2MTFkZGY4YmQwNmQ3YjcyNDY5NjFlYzQx ZjU5MzNlOGNmZiAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdDIvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJLaXQyL0NoYW5nZUxvZwpAQCAtMSwzICsxLDIzIEBACisyMDEyLTExLTIxICBBbmRy YXMgQmVjc2kgIDxhbmRyYXMuYmVjc2lAZGlnaWEuY29tPgorCisgICAgICAgIFtXSzJdIFZpZXdw b3J0IG1ldGEgdGFnIGJyb2tlbiBhZnRlciByMTM0ODAxCisgICAgICAgIGh0dHBzOi8vYnVncy53 ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD0xMDI4MDEKKworICAgICAgICBSZXZpZXdlZCBieSBO T0JPRFkgKE9PUFMhKS4KKworICAgICAgICBJZiB0aGUgaW5pdGlhbC1zY2FsZSBhdHRyaWJ1dGUg aXMgbm90IGV4cGxpY2l0bHkgc3BlY2lmaWVkIGl0IGlzIGxlZnQgLTEgYnkgdGhlCisgICAgICAg IHZpZXdwb3J0IG1ldGEgYWxnb3JpdGhtIGFuZCBpZiB0aGUgY29udGVudCBpcyBub3Qgc2NhbGFi bGUgdGhlIG1pbiBhbmQgbWF4IHNjYWxlcworICAgICAgICBhcmUgcmVzdHJpY3RlZCBieSByZXN0 cmljdFNjYWxlRmFjdG9yVG9Jbml0aWFsU2NhbGVJZk5vdFVzZXJTY2FsYWJsZSgpIHdoaWNoIHJl c3VsdGVkCisgICAgICAgIGluIG5lZ2F0aXZlIHNjYWxlcyBvbiB0aGUgVUkgc2lkZS4KKyAgICAg ICAgTWFrZSBzdXJlIHRoYXQgdGhlIG1pbmltdW0gc2NhbGUgdG8gZml0IHRoZSB2aWV3cG9ydCBp cyBjYWxjdWxhdGVkIGJlZm9yZSByZXN0cmljdGluZworICAgICAgICB0aGUgc2NhbGVzIGFuZCB1 c2UgdGhpcyBzY2FsZSBhcyB0aGUgaW5pdGlhbCBzY2FsZSBpZiBubyBleHBsaWNpdCBpbml0aWFs IHNjYWxlIHdhcyBzZXQuCisKKyAgICAgICAgKiBVSVByb2Nlc3MvUGFnZVZpZXdwb3J0Q29udHJv bGxlci5jcHA6CisgICAgICAgIChXZWJLaXQ6OlBhZ2VWaWV3cG9ydENvbnRyb2xsZXI6OlBhZ2VW aWV3cG9ydENvbnRyb2xsZXIpOgorICAgICAgICAoV2ViS2l0OjpQYWdlVmlld3BvcnRDb250cm9s bGVyOjpwYWdlVHJhbnNpdGlvblZpZXdwb3J0UmVhZHkpOgorICAgICAgICAoV2ViS2l0OjpQYWdl Vmlld3BvcnRDb250cm9sbGVyOjpkaWRDaGFuZ2VWaWV3cG9ydEF0dHJpYnV0ZXMpOgorICAgICAg ICAoV2ViS2l0OjpQYWdlVmlld3BvcnRDb250cm9sbGVyOjp1cGRhdGVNaW5pbXVtU2NhbGVUb0Zp dCk6CisKIDIwMTItMTEtMjAgIEphbWVzIFNpbW9uc2VuICA8c2ltb25qYW1AY2hyb21pdW0ub3Jn PgogCiAgICAgICAgIENvbnNvbGlkYXRlIEZyYW1lTG9hZGVyOjpsb2FkKCkgaW50byBvbmUgZnVu Y3Rpb24gdGFraW5nIGEgRnJhbWVMb2FkUmVxdWVzdApkaWZmIC0tZ2l0IGEvU291cmNlL1dlYktp dDIvVUlQcm9jZXNzL1BhZ2VWaWV3cG9ydENvbnRyb2xsZXIuY3BwIGIvU291cmNlL1dlYktpdDIv VUlQcm9jZXNzL1BhZ2VWaWV3cG9ydENvbnRyb2xsZXIuY3BwCmluZGV4IDdjYTBhOTNiMmZlMjMx YjNlNTQ4YjFjNTE2Njc0YzczMjUxNGQ4NmIuLjM1ZjM2MWNhZDk4YTA4ODk0MzQwOGNkYmVhZmM2 NGFjMDk3ODg4MTQgMTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJLaXQyL1VJUHJvY2Vzcy9QYWdlVmll d3BvcnRDb250cm9sbGVyLmNwcAorKysgYi9Tb3VyY2UvV2ViS2l0Mi9VSVByb2Nlc3MvUGFnZVZp ZXdwb3J0Q29udHJvbGxlci5jcHAKQEAgLTUyLDExICs1MiwxNSBAQCBQYWdlVmlld3BvcnRDb250 cm9sbGVyOjpQYWdlVmlld3BvcnRDb250cm9sbGVyKFdlYktpdDo6V2ViUGFnZVByb3h5KiBwcm94 eSwgUGFnZQogewogICAgIC8vIEluaXRpYWxpemluZyBWaWV3cG9ydCBSYXcgQXR0cmlidXRlcyB0 byBhdm9pZCByYW5kb20gbmVnYXRpdmUgb3IgaW5maW5pdHkgc2NhbGUgZmFjdG9ycwogICAgIC8v IGlmIHRoZXJlIGlzIGEgcmFjZSBjb25kaXRpb24gYmV0d2VlbiB0aGUgZmlyc3QgbGF5b3V0IGFu ZCBzZXR0aW5nIHRoZSB2aWV3cG9ydCBhdHRyaWJ1dGVzIGZvciB0aGUgZmlyc3QgdGltZS4KLSAg ICBtX3Jhd0F0dHJpYnV0ZXMuaW5pdGlhbFNjYWxlID0gMTsKICAgICBtX3Jhd0F0dHJpYnV0ZXMu bWluaW11bVNjYWxlID0gMTsKICAgICBtX3Jhd0F0dHJpYnV0ZXMubWF4aW11bVNjYWxlID0gMTsK ICAgICBtX3Jhd0F0dHJpYnV0ZXMudXNlclNjYWxhYmxlID0gbV9hbGxvd3NVc2VyU2NhbGluZzsK IAorICAgIC8vIFRoZSBpbml0aWFsIHNjYWxlIG1pZ2h0IGJlIGltcGxpY2l0IGFuZCBzZXQgdG8g LTEsIGluIHRoaXMgY2FzZSB3ZSBoYXZlIHRvIGluZmVyIGl0CisgICAgLy8gdXNpbmcgdGhlIHZp ZXdwb3J0IHNpemUgYW5kIHRoZSBmaW5hbCBsYXlvdXQgc2l6ZS4KKyAgICAvLyBUbyBiZSBhYmxl IHRvIGFzc2VydCBmb3IgdmFsaWQgc2NhbGUgd2UgaW5pdGlhbGl6ZSBpdCB0byAtMS4KKyAgICBt X3Jhd0F0dHJpYnV0ZXMuaW5pdGlhbFNjYWxlID0gLTE7CisKICAgICBBU1NFUlQobV9jbGllbnQp OwogICAgIG1fY2xpZW50LT5zZXRDb250cm9sbGVyKHRoaXMpOwogfQpAQCAtMTQ1LDggKzE0OSw4 IEBAIHZvaWQgUGFnZVZpZXdwb3J0Q29udHJvbGxlcjo6cGFnZVRyYW5zaXRpb25WaWV3cG9ydFJl YWR5KCkKIHsKICAgICBpZiAoIW1fcmF3QXR0cmlidXRlcy5sYXlvdXRTaXplLmlzRW1wdHkoKSkg ewogICAgICAgICBtX2hhZFVzZXJJbnRlcmFjdGlvbiA9IGZhbHNlOwotICAgICAgICBmbG9hdCBp bml0aWFsU2NhbGUgPSAobV9yYXdBdHRyaWJ1dGVzLmluaXRpYWxTY2FsZSA8IDApID8gbV9taW5p bXVtU2NhbGVUb0ZpdCA6IG1fcmF3QXR0cmlidXRlcy5pbml0aWFsU2NhbGU7Ci0gICAgICAgIGFw cGx5U2NhbGVBZnRlclJlbmRlcmluZ0NvbnRlbnRzKGlubmVyQm91bmRlZFZpZXdwb3J0U2NhbGUo dG9WaWV3cG9ydFNjYWxlKGluaXRpYWxTY2FsZSkpKTsKKyAgICAgICAgQVNTRVJUKG1fcmF3QXR0 cmlidXRlcy5pbml0aWFsU2NhbGUgPiAwKTsKKyAgICAgICAgYXBwbHlTY2FsZUFmdGVyUmVuZGVy aW5nQ29udGVudHMoaW5uZXJCb3VuZGVkVmlld3BvcnRTY2FsZSh0b1ZpZXdwb3J0U2NhbGUobV9y YXdBdHRyaWJ1dGVzLmluaXRpYWxTY2FsZSkpKTsKICAgICB9CiAKICAgICAvLyBBdCB0aGlzIHBv aW50IHdlIHNob3VsZCBhbHJlYWR5IGhhdmUgcmVjZWl2ZWQgdGhlIGZpcnN0IHZpZXdwb3J0IGFy Z3VtZW50cyBhbmQgdGhlIHJlcXVlc3RlZCBzY3JvbGwKQEAgLTIxMywxMCArMjE3LDE4IEBAIHZv aWQgUGFnZVZpZXdwb3J0Q29udHJvbGxlcjo6ZGlkQ2hhbmdlVmlld3BvcnRBdHRyaWJ1dGVzKGNv bnN0IFdlYkNvcmU6OlZpZXdwb3J0CiAgICAgICAgIHJldHVybjsKIAogICAgIG1fcmF3QXR0cmli dXRlcyA9IG5ld0F0dHJpYnV0ZXM7Ci0gICAgV2ViQ29yZTo6cmVzdHJpY3RTY2FsZUZhY3RvclRv SW5pdGlhbFNjYWxlSWZOb3RVc2VyU2NhbGFibGUobV9yYXdBdHRyaWJ1dGVzKTsKLQogICAgIG1f YWxsb3dzVXNlclNjYWxpbmcgPSAhIW1fcmF3QXR0cmlidXRlcy51c2VyU2NhbGFibGU7Ci0gICAg dXBkYXRlTWluaW11bVNjYWxlVG9GaXQoKTsKKworICAgIGlmICghdXBkYXRlTWluaW11bVNjYWxl VG9GaXQoKSkKKyAgICAgICAgcmV0dXJuOworCisgICAgQVNTRVJUKG1fbWluaW11bVNjYWxlVG9G aXQgPiAwKTsKKworICAgIC8vIFNldCB0aGUgaW5pdGlhbCBzY2FsZSBpZiBpdCB3YXMgbm90IHNw ZWNpZmllZCBpbiB0aGUgdmlld3BvcnQgbWV0YSB0YWcuCisgICAgaWYgKG1fcmF3QXR0cmlidXRl cy5pbml0aWFsU2NhbGUgPCAwKQorICAgICAgICBtX3Jhd0F0dHJpYnV0ZXMuaW5pdGlhbFNjYWxl ID0gbV9taW5pbXVtU2NhbGVUb0ZpdDsKKworICAgIFdlYkNvcmU6OnJlc3RyaWN0U2NhbGVGYWN0 b3JUb0luaXRpYWxTY2FsZUlmTm90VXNlclNjYWxhYmxlKG1fcmF3QXR0cmlidXRlcyk7CiAKICAg ICBtX2NsaWVudC0+ZGlkQ2hhbmdlVmlld3BvcnRBdHRyaWJ1dGVzKCk7CiB9CkBAIC0yNjIsMTEg KzI3NCwxNCBAQCB2b2lkIFBhZ2VWaWV3cG9ydENvbnRyb2xsZXI6OmFwcGx5UG9zaXRpb25BZnRl clJlbmRlcmluZ0NvbnRlbnRzKGNvbnN0IEZsb2F0UG9pbgogCiBib29sIFBhZ2VWaWV3cG9ydENv bnRyb2xsZXI6OnVwZGF0ZU1pbmltdW1TY2FsZVRvRml0KCkKIHsKLSAgICBpZiAobV92aWV3cG9y dFNpemUuaXNFbXB0eSgpKQorICAgIGlmIChtX3ZpZXdwb3J0U2l6ZS5pc0VtcHR5KCkgfHwgbV9j b250ZW50c1NpemUuaXNFbXB0eSgpKQogICAgICAgICByZXR1cm4gZmFsc2U7CiAKICAgICBmbG9h dCBtaW5pbXVtU2NhbGUgPSBXZWJDb3JlOjpjb21wdXRlTWluaW11bVNjYWxlRmFjdG9yRm9yQ29u dGVudENvbnRhaW5lZChtX3Jhd0F0dHJpYnV0ZXMsIFdlYkNvcmU6OnJvdW5kZWRJbnRTaXplKG1f dmlld3BvcnRTaXplKSwgV2ViQ29yZTo6cm91bmRlZEludFNpemUobV9jb250ZW50c1NpemUpLCBk ZXZpY2VQaXhlbFJhdGlvKCkpOwogCisgICAgaWYgKG1pbmltdW1TY2FsZSA8PSAwKQorICAgICAg ICByZXR1cm4gZmFsc2U7CisKICAgICBpZiAoIWZ1enp5Q29tcGFyZShtaW5pbXVtU2NhbGUsIG1f bWluaW11bVNjYWxlVG9GaXQsIDAuMDAxKSkgewogICAgICAgICBtX21pbmltdW1TY2FsZVRvRml0 ID0gbWluaW11bVNjYWxlOwogCg==