Line 301 of UIView_KIFAdditions causes a bad access error. #1108
Comments
|
This seems like either an Apple bug in Which version of KIF are you using? Some behavior in this method was changed relatively recently: You might be able to comment out that line, as I don't think spinning the runloop there should be necessary. I'd guess though that this will just push the crash to happen elsewhere, but it is worth a shot. |
|
You could also try going back to version 3.7.3 of KIF, before the major changes in how we match against table view cells. If that works, it would be very helpful to know. |
|
It's an odd one because it occasionally appears and I can't get clear steps as to what happened to cause it. |
|
The crash is within UIKit code, so it isn't something that is inherently KIFs fault. It is possible that it is exacerbated by quickly scrolling to many different positions in tableviews to try and find a matching element, though not necessarily. The table view cell matching code changed as a mitigation for issue #1007, where we used to enumerate the table view cells programmatically and started throwing exceptions in iOS 11. This is why I asked about going back to version 3.7.3 or before to see if that avoids the problem. This could SO message could also be related: I'm going to close this issue, as there isn't anything clearly wrong with KIF. If you disagree, please feel free to reactivate. |
|
I ran all the tests again today on a different Mac with the latest pod version of KIF and without the error occurring.
… On Jun 24, 2019, at 6:41 PM, Justin Martin ***@***.***> wrote:
The crash is within UIKit code, so it isn't something that is inherently KIFs fault. It is possible that it is exacerbated by quickly scrolling to many different positions in tableviews to try and find a matching element, though not necessarily. The table view cell matching code changed as a mitigation for issue #1007 <#1007>, where we used to enumerate the table view cells programmatically and started throwing exceptions in iOS 11. This is why I asked about going back to version 3.7.3 or before to see if that avoids the problem.
This could SO message could also be related:
https://stackoverflow.com/questions/39530125/app-crashing-on-runtransitionforcurrentstate-but-no-clue-as-to-why <https://stackoverflow.com/questions/39530125/app-crashing-on-runtransitionforcurrentstate-but-no-clue-as-to-why>
I'm going to close this issue, as there isn't anything clearly wrong with KIF. If you disagree, please feel free to reactivate.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub <#1108?email_source=notifications&email_token=ABGEU2FCJRV4BORT6HJP5I3P4FLTPA5CNFSM4HMI2F7KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYORCVA#issuecomment-505221460>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABGEU2FLYFNYPZQWL3J2M63P4FLTPANCNFSM4HMI2F7A>.
|
When running our tests from Xcode with command U,
calling return [self _predicateSearchWithRequiresMatch:YES mustBeTappable:NO].view;
ends up creating a bad access error on line 301 of UIView-KIFAdditions.
CFRunLoopRunInMode(UIApplicationCurrentRunMode, delay, false);
Lots of cases under Xcode 10.2.1 are reporting EXC_BAD_ACCESS at calls to CFRunLoopRunInMode within UIView-KIFAdditions.
The text was updated successfully, but these errors were encountered: