In message <49bfd516.6030...@gmail.com>, Tim Ellison writes: > > Sian January wrote: > > Thanks Regis - that fixes the issue for me on Windows XP. I'm happy > > to check it in if another committer agrees. > > Since we see it is a test case failure, and the fix does not affect the > implementation, I suggest that it waits until after the release is declared.
Since we see it is a test case failure, and the fix does not affect the implementation, I'd vote to fix it now. ;-) Junit stops a test method at the first failed assert so any failure could be hiding other "real" bugs, so if a test fix doesn't affect the implementation then I think I'd always be +1 for applying it as soon as possible. Regards, Mark. > > 2009/3/17 Regis <xu.re...@gmail.com>: > >> Sian January wrote: > >>> Ok - I have tried to list all the issues from the other thread below, > >>> but please add any I've missed. > >>> > >>> It would be good to put a name against each of these so that we know > >>> they are being looked into. Any volunteers? > >>> > >>> 1. HttpURLConnectionTest and HttpsURLConnectionTest fail on some > >>> systems - change to use Jetty? > >>> 2. org.apache.harmony.archive.tests.java.util.jar.ManifestTest fails > >>> on Windows XP > >>> 3. RMI test failures on Linux > >>> 4. VM crash in classlib tests on Linux - > >>> org.apache.harmony.luni.tests.java.net.MulticastSocketTest and > >>> org.apache.harmony.xnet.provider.jsse.SSLSocketImplTest > >>> 5. Classlib test failures on Vista in > >>> org.apache.harmony.luni.tests.java.net.MulticastSocketTest > >>> and org.apache.harmony.luni.tests.java.net.SocketTest > >>> 6. hymem.exe crashes on Vista > >>> 7. Various AWT/Swing crashes/ failures > >>> > >>> > >>> 2009/3/16 Regis <xu.re...@gmail.com>: > >>>> Sian January wrote: > >>>>> It should be the end of code freeze today but I'm not sure we're ready > >>>>> to vote on M9 yet without discussing all of the test failures that > >>>>> people have found. > >>>>> > >>>>> It seems like the HttpURLConnection tests are test errors rather than > >>>>> bugs. > >>>>> > >>>>> I also think that if something fails on Windows Vista and not on XP > >>>>> then we probably shouldn't consider it a blocker, though it would be > >>>>> nice to fix it in general. > >>>>> > >>>>> For awt and swing crashes, I think these are common (I don't think > >>>>> Churong runs them as part of the integrity test suite) and have been > >>>>> allowed through in previous milestones, so I don't think these are > >>>>> blockers for M9. > >>>>> > >>>>> Can anyone else offer opinions any of the other issues? > >>>>> > >>>>> Thanks, > >>>>> > >>>>> Sian > >>>>> > >>>>> > >>>>> > >>>> I think all crash issues should be resolved before M9 shipped, even some > >>>> are in previous milestones, it's time to fix them. > >>>> > >>>> -- > >>>> Best Regards, > >>>> Regis. > >>>> > >>> > >>> > >> I have created a JIRA HARMONY-6120 for HttpURLConnection tests, and attach > ed > >> a patch to use Jetty as test server, would anyone help to try it? > >> > >> -- > >> Best Regards, > >> Regis. > >> > > > > > > >
