http://trac.webkit.org/changeset/118105 added a new test: http/tests/cookies/js-get-and-set-http-only-cookie.php which is failing on Mac WK1 and WK2. http://build.webkit.org/results/Lion%20Release%20(Tests)/r118111%20(8569)/results.html --- /Volumes/Data/slave/lion-intel-release-tests/build/layout-test-results/http/tests/cookies/js-get-and-set-http-only-cookie-expected.txt +++ /Volumes/Data/slave/lion-intel-release-tests/build/layout-test-results/http/tests/cookies/js-get-and-set-http-only-cookie-actual.txt @@ -5,7 +5,7 @@ Check that we can't get or set httpOnly Cookies by JavaScript. PASS We can't get httpOnly cookies by JavaScript. -PASS We can't set httpOnly cookies by JavaScript. +FAIL We shouldn't set httpOnly cookies by JavaScript. PASS successfullyParsed is true TEST COMPLETE
<rdar://problem/11512237>
Checked in failing results http://trac.webkit.org/projects/webkit/changeset/118122
Same fail on GTK and Qt too.
I skipped it on Qt - r118132.
It seems this test is also causing various failures (even crashes in GTK builders) in http/tests/cookies/multiple-cookies.html, which is the test that is run after this test. http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20-%20webkit.org&tests=http%2Ftests%2Fcookies%2Fmultiple-cookies.html I'll skip the test for Gtk.
This test can have side effects. Skipped in http://trac.webkit.org/changeset/118305
(In reply to comment #5) > It seems this test is also causing various failures (even crashes in GTK builders) in http/tests/cookies/multiple-cookies.html, which is the test that is run after this test. > > http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20-%20webkit.org&tests=http%2Ftests%2Fcookies%2Fmultiple-cookies.html > > I'll skip the test for Gtk. Do you know why multiple-cookies.html failed? I don't think my test case can cause this failing.
(In reply to comment #7) > (In reply to comment #5) > > It seems this test is also causing various failures (even crashes in GTK builders) in http/tests/cookies/multiple-cookies.html, which is the test that is run after this test. > > > > http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20-%20webkit.org&tests=http%2Ftests%2Fcookies%2Fmultiple-cookies.html > > > > I'll skip the test for Gtk. > > Do you know why multiple-cookies.html failed? > > I don't think my test case can cause this failing. The stderr output of the crash was: 'Creating pipes for GWakeup: Too many open files' This perhaps signals a platform-specific problem, i.e. a bug in either the libsoup HTTP backend or libsoup itself.
The failure on mac is because of an actual bug in cookie jar.
To be more precise in WebCore/platform/mac/CookieJar.mm, it is not checked that a new cookie wouldn't override an existing httpOnly cookie.
I will work on a fix for the SOUP-specific code in Bug 88760. This looks like a valid problem as we allow setting/overwriting httpOnly cookies from JavaScript.
Skipped on Qt by http://trac.webkit.org/changeset/119964/trunk/LayoutTests/platform/qt/Skipped Please unskip it with the proper fix.
This test was removed in http://trac.webkit.org/changeset/118166 and reintroduced as http/tests/cookies/js-get-and-set-http-only-cookie.html, which does not appear to be suffering from the same problems. I am going to fix up the Skipped lists.
Actually, it looks like the Qt Skipped list entry was for the .html version introduced in http://trac.webkit.org/changeset/119947. Updating the title to reflect that.
Actually, this still fails on Mac, <rdar://problem/12427027>.
I checked this test but it still fails on Qt.
=== Bulk closing of Qt bugs === If you believe that this bug report is still relevant for a non-Qt port of webkit.org, please re-open it and remove [Qt] from the summary. If you believe that this is still an important QtWebKit bug, please fill a new report at https://bugreports.qt-project.org and add a link to this issue. See http://qt-project.org/wiki/ReportingBugsInQt for additional guidelines.