Adam Roben wrote:
On Apr 2, 2010, at 9:25 PM, Russ Schnapp wrote:
I've built WebKit for Windows, and am trying to make sure it's all working
(though I have to say I'm daunted by all the non-open-source third party DLLs
that have to be present in order for it to run!).
While running WebKitAPITest, the MainFrameAfterClose test is failing. The
destruction of WebView.m_page (triggered by WebView::close) is somehow also
causing the deallocation of m_mainFrame.
Is this a latent bug, discovered only because I happen to be testing with
Visual Studio's Debug configuration (which fills deallocated blocks with
0xfeeefeee)?
This is a known issue: <https://bugs.webkit.org/show_bug.cgi?id=32868>. This
bug is mentioned in the source code for that test.
Ah. I saw the reference in the code, but didn't realize it was for an
outstanding bug.
I just noticed one other failure: NoCloseOrDestroyViewWindow reports
"webViewCount() == currentWebViewCount - 1 (.\tests\WebViewDestruction.cpp:92)"
Apparently, the webViewCount() didn't decrement from 1 to 0 after deleting the
reference to webView.
This is unexpected. The original issue was reported as
<https://bugs.webkit.org/show_bug.cgi?id=33162>. Would you be willing to file a
new bug report?
-Adam
Sure, I'm glad to help. I just wanted to make sure I wasn't somehow
building the library or the test incorrectly.
_______________________________________________
webkit-help mailing list
[email protected]
http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
