97875 2012-09-28 00:52:28 -0700 EventHandler should not use shadowAncestorNode() 2019-10-04 22:48:15 -0700 1 1 1 Unclassified WebKit DOM 528+ (Nightly build) Unspecified Unspecified RESOLVED CONFIGURATION CHANGED InRadar P2 Normal --- 91821 1 shinyak webkit-unassigned dbates dglazkov morrita rniwa tkent tonikitoo webkit-bug-importer oldest_to_newest 730325 0 shinyak 2012-09-28 00:52:28 -0700 shadowAncestorNode() is used in handleMouseReleaseEvent(), but it seems to cause some error in nested shadow subtree. 734345 1 167028 tasak 2012-10-03 21:13:42 -0700 Created attachment 167028 Patch 734907 2 167028 dbates 2012-10-04 11:57:43 -0700 Comment on attachment 167028 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=167028&action=review > Source/WebCore/ChangeLog:12 > + Since Node::shadowAncestorNode is deprecated, it should not be used. > + So replaced the method with Node::shadowHost and Node::isInShadowTree. > + > + No new tests, because this patch is just for refactoring. > + All existing tests have already covered the EventHandler's behavior. The description for this change differs from the description in the Bugzilla bug report. In particular, the Bugzilla bug description implies that there is some kind of error that is caused by the use of shadowAncestorNode() in EventHandler::handleMouseReleaseEvent(). I take it that the Bugzilla bug description is inaccurate? 737306 3 shinyak 2012-10-08 18:27:55 -0700 It would be good to have a test for this. When I clicked in selected text in shadow dom, the selection is not released, while selection in light dom is released. 737309 4 shinyak 2012-10-08 18:29:57 -0700 Since we have implemented event retargeting, it might be OK if we don't use shadowHost() here. I'm not sure though... 737415 5 shinyak 2012-10-08 22:10:14 -0700 When we add preventDefault in Shadow DOM, we cannot release selection. 737420 6 167691 shinyak 2012-10-08 22:13:11 -0700 Created attachment 167691 Testcase Select 'foobar' and click the selection. 737459 7 shinyak 2012-10-08 23:36:00 -0700 Oops, that might be related to nothing... 737463 8 shinyak 2012-10-08 23:38:38 -0700 (In reply to comment #4) > Since we have implemented event retargeting, it might be OK if we don't use shadowHost() here. I'm not sure though... This will cause an error for input[type="range"], according to http://trac.webkit.org/changeset/122224 781614 9 dglazkov 2012-12-03 10:38:06 -0800 Sakamoto-san, any updates on the patch? It's sitting in the queue, but I haven't seen any responses to the comments. 781615 10 167028 dglazkov 2012-12-03 10:38:22 -0800 Comment on attachment 167028 Patch Needs test? 1577182 11 rniwa 2019-10-04 22:47:15 -0700 shadowAncestorNode no longer exists. 1577183 12 webkit-bug-importer 2019-10-04 22:48:15 -0700 <rdar://problem/56003995> 167028 2012-10-03 21:13:42 -0700 2012-12-03 10:38:21 -0800 Patch bug-97875-20121004131253.patch text/plain 2340 tasak U3VidmVyc2lvbiBSZXZpc2lvbjogMTMwMzUyCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXggYzJmYTNiNjhiMjIwN2I5 MTc1ODY5ZDBmMzI4YTE4ZTM1NTEwNTBhZC4uMThmMmZmNWI1NzNkNmUzMGYzZWYyOTgxNDE0Zjk3 NjU5NDdiZTZlMyAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJDb3JlL0NoYW5nZUxvZwpAQCAtMSwzICsxLDIyIEBACisyMDEyLTEwLTAzICBUYWth c2hpIFNha2Ftb3RvICA8dGFzYWtAZ29vZ2xlLmNvbT4KKworICAgICAgICBFdmVudEhhbmRsZXIg c2hvdWxkIG5vdCB1c2Ugc2hhZG93QW5jZXN0b3JOb2RlKCkKKyAgICAgICAgaHR0cHM6Ly9idWdz LndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTk3ODc1CisKKyAgICAgICAgUmV2aWV3ZWQgYnkg Tk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgU2luY2UgTm9kZTo6c2hhZG93QW5jZXN0b3JOb2Rl IGlzIGRlcHJlY2F0ZWQsIGl0IHNob3VsZCBub3QgYmUgdXNlZC4KKyAgICAgICAgU28gcmVwbGFj ZWQgdGhlIG1ldGhvZCB3aXRoIE5vZGU6OnNoYWRvd0hvc3QgYW5kIE5vZGU6OmlzSW5TaGFkb3dU cmVlLgorCisgICAgICAgIE5vIG5ldyB0ZXN0cywgYmVjYXVzZSB0aGlzIHBhdGNoIGlzIGp1c3Qg Zm9yIHJlZmFjdG9yaW5nLgorICAgICAgICBBbGwgZXhpc3RpbmcgdGVzdHMgaGF2ZSBhbHJlYWR5 IGNvdmVyZWQgdGhlIEV2ZW50SGFuZGxlcidzIGJlaGF2aW9yLgorCisgICAgICAgICogcGFnZS9F dmVudEhhbmRsZXIuY3BwOgorICAgICAgICAoV2ViQ29yZTo6RXZlbnRIYW5kbGVyOjpoYW5kbGVN b3VzZVJlbGVhc2VFdmVudCk6CisgICAgICAgIFJlcGxhY2VkIE5vZGU6OnNoYWRvd0FuY2VzdHJv Tm9kZSB3aXRoIE5vZGU6OmlzSW5TaGFkb3dUcmVlIGFuZAorICAgICAgICBOb2RlOjpzaGFkb3dI b3N0LiBJZiBpc0luU2hhZG93VHJlZSByZXR1cm5zIHRydWUsIHVzZXMgc2hhZG93SG9zdC4KKyAg ICAgICAgT3RoZXJ3aXNlLCBqdXN0IHVzZSB0aGUgc2FtZSB2YWx1ZS4KKwogMjAxMi0xMC0wMyAg QWRhbSBCYXJ0aCAgPGFiYXJ0aEB3ZWJraXQub3JnPgogCiAgICAgICAgIE1lYXN1cmUgdGhlIHVz YWdlIG9mIFdlYlNRTERhdGFiYXNlCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9wYWdlL0V2 ZW50SGFuZGxlci5jcHAgYi9Tb3VyY2UvV2ViQ29yZS9wYWdlL0V2ZW50SGFuZGxlci5jcHAKaW5k ZXggZmUyMzMwZWEyZGZlNDYyNmU3YWM4MGQ4MzUxZjY5MWZiNjg2MTNlOS4uYzBiZDljMWFmZGVm NGZkYzJlZDdmMGUwOTE0NjY0MTE2NGNiODJmNSAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUv cGFnZS9FdmVudEhhbmRsZXIuY3BwCisrKyBiL1NvdXJjZS9XZWJDb3JlL3BhZ2UvRXZlbnRIYW5k bGVyLmNwcApAQCAtMTg4OSw5ICsxODg5LDkgQEAgYm9vbCBFdmVudEhhbmRsZXI6OmhhbmRsZU1v dXNlUmVsZWFzZUV2ZW50KGNvbnN0IFBsYXRmb3JtTW91c2VFdmVudCYgbW91c2VFdmVudCkKICAg ICBib29sIHN3YWxsb3dNb3VzZVVwRXZlbnQgPSBkaXNwYXRjaE1vdXNlRXZlbnQoZXZlbnROYW1l cygpLm1vdXNldXBFdmVudCwgbWV2LnRhcmdldE5vZGUoKSwgdHJ1ZSwgbV9jbGlja0NvdW50LCBt b3VzZUV2ZW50LCBmYWxzZSk7CiAKICAgICBOb2RlKiBjbGlja1RhcmdldCA9IG1ldi50YXJnZXRO b2RlKCk7Ci0gICAgaWYgKGNsaWNrVGFyZ2V0KQotICAgICAgICBjbGlja1RhcmdldCA9IGNsaWNr VGFyZ2V0LT5zaGFkb3dBbmNlc3Rvck5vZGUoKTsKLSAgICBOb2RlKiBhZGp1c3RlZENsaWNrTm9k ZSA9IG1fY2xpY2tOb2RlID8gbV9jbGlja05vZGUtPnNoYWRvd0FuY2VzdG9yTm9kZSgpIDogMDsK KyAgICBpZiAoY2xpY2tUYXJnZXQgJiYgY2xpY2tUYXJnZXQtPmlzSW5TaGFkb3dUcmVlKCkpCisg ICAgICAgIGNsaWNrVGFyZ2V0ID0gY2xpY2tUYXJnZXQtPnNoYWRvd0hvc3QoKTsKKyAgICBOb2Rl KiBhZGp1c3RlZENsaWNrTm9kZSA9IG1fY2xpY2tOb2RlICYmIG1fY2xpY2tOb2RlLT5pc0luU2hh ZG93VHJlZSgpID8gbV9jbGlja05vZGUtPnNoYWRvd0hvc3QoKSA6IG1fY2xpY2tOb2RlLmdldCgp OwogCiAgICAgYm9vbCBzd2FsbG93Q2xpY2tFdmVudCA9IG1fY2xpY2tDb3VudCA+IDAgJiYgbW91 c2VFdmVudC5idXR0b24oKSAhPSBSaWdodEJ1dHRvbiAmJiBjbGlja1RhcmdldCA9PSBhZGp1c3Rl ZENsaWNrTm9kZSAmJiBkaXNwYXRjaE1vdXNlRXZlbnQoZXZlbnROYW1lcygpLmNsaWNrRXZlbnQs IG1ldi50YXJnZXROb2RlKCksIHRydWUsIG1fY2xpY2tDb3VudCwgbW91c2VFdmVudCwgdHJ1ZSk7 CiAK 167691 2012-10-08 22:13:11 -0700 2012-10-08 22:13:11 -0700 Testcase mouse-release-in-nested-shadow.html text/html 705 shinyak PCFET0NUWVBFIGh0bWw+CjxodG1sPgo8aGVhZD4KICAgIDxzY3JpcHQgc3JjPSJyZXNvdXJjZXMv cG9seWZpbGwuanMiPjwvc2NyaXB0Pgo8L2hlYWQ+Cjxib2R5PgoKPGRpdiBpZD0iY29udGFpbmVy Ij4KICAgIDxkaXYgaWQ9Imhvc3QxIj48L2Rpdj4KPC9kaXY+Cgo8aHIgLz4KCjxkaXY+CiAgICA8 ZGl2IGlkPSJ0YXJnZXQyIj5mb288L2Rpdj5iYXIKPC9kaXY+Cgo8c2NyaXB0Pgp2YXIgc2hhZG93 Um9vdDEgPSBuZXcgV2ViS2l0U2hhZG93Um9vdChob3N0MSk7CnNoYWRvd1Jvb3QxLmlubmVySFRN TCA9ICc8ZGl2IGlkPSJob3N0MiI+PC9kaXY+JwoKdmFyIHNoYWRvd1Jvb3QyID0gbmV3IFdlYktp dFNoYWRvd1Jvb3Qoc2hhZG93Um9vdDEuZ2V0RWxlbWVudEJ5SWQoJ2hvc3QyJykpOwpzaGFkb3dS b290Mi5pbm5lckhUTUwgPSAnPGRpdiBpZD0idGFyZ2V0MSI+Zm9vPC9kaXY+YmFyJzsKCnZhciB0 YXJnZXQxID0gc2hhZG93Um9vdDIuZ2V0RWxlbWVudEJ5SWQoJ3RhcmdldDEnKTsKCnRhcmdldDEu YWRkRXZlbnRMaXN0ZW5lcignbW91c2V1cCcsIGZ1bmN0aW9uKCkgewogICAgZXZlbnQucHJldmVu dERlZmF1bHQoKTsKfSwgdHJ1ZSk7CnRhcmdldDIuYWRkRXZlbnRMaXN0ZW5lcignbW91c2V1cCcs IGZ1bmN0aW9uKCkgewogICAgZXZlbnQucHJldmVudERlYWZ1bHQoKTsKfSwgdHJ1ZSk7Cjwvc2Ny aXB0Pgo8L2JvZHk+CjwvaHRtbD4K