we were crashing on setCursor for QGWV, but not for QWebView, and is not yet any auto test to catch these =/ this bug is a follow up of bug 30549, which landed a fix/workaround for the crash. ps: it'd be nice if autotest here could check the unsetCursor egde case in ::event() implementation.
seems like a quick one , picking ...
Created attachment 42958 [details] patch 0.1 patch depends on patch in bug 31351.
Comment on attachment 42958 [details] patch 0.1 patch 0.1 is does not cover the right behavior