RESOLVED FIXED 212050
Rare crash under -[WKContentView resignFirstResponderForWebView] 
https://bugs.webkit.org/show_bug.cgi?id=212050
Summary Rare crash under -[WKContentView resignFirstResponderForWebView]
Tim Horton
Reported 2020-05-18 15:29:45 PDT
Rare crash under -[WKContentView resignFirstResponderForWebView]
Attachments
Patch (2.92 KB, patch)
2020-05-18 15:30 PDT, Tim Horton 		no flags
DetailsFormatted DiffDiff
View All Add attachment proposed patch, testcase, etc.
Tim Horton
Comment 1 2020-05-18 15:30:19 PDT
Created attachment 399677 [details] Patch
Tim Horton
Comment 2 2020-05-18 15:30:21 PDT
<rdar://problem/60187111>
Wenson Hsieh
Comment 3 2020-05-18 15:32:54 PDT
Comment on attachment 399677 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=399677&action=review > Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm:1412 > + if (strongSelf->_keyWebEventHandler.get() != strongHandler.get()) Nit - are both of these .get() needed?
Megan Gardner
Comment 4 2020-05-18 15:34:53 PDT
Comment on attachment 399677 [details] Patch r=me as well
Tim Horton
Comment 5 2020-05-18 16:02:33 PDT
Comment on attachment 399677 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=399677&action=review >> Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm:1412 >> + if (strongSelf->_keyWebEventHandler.get() != strongHandler.get()) > > Nit - are both of these .get() needed? As we learned extensively on Slack, yes, because this code is being pretty dirty and WeakObjCPtr<id>-ing the block, and no combination of BlockPtr/WeakObjCPtr on either side make enough implicit conversions that you can drop either get (nor both).
EWS
Comment 6 2020-05-18 16:31:47 PDT
Committed r261838: <https://trac.webkit.org/changeset/261838> All reviewed patches have been landed. Closing bug and clearing flags on attachment 399677 [details].
Note You need to log in before you can comment on or make changes to this bug.