Even though the manifest is invalid according to the spec, shouldn't we
be parsing it the same way as the RI for compatibility reasons?
I wonder how many apps have invalid manifests.
Regards,
Tim
Nathan Beyer wrote:
Careful, the test isn't necessarily invalid - that's quick assumption. It
needs to be verified that the test isn't dyanamically generating a manifest
with Harmony's classlib; it's possible that Harmony's classlib is creating
the invalid manifest, which later causes a failure when reading the
manifest.
-Nathan
On Fri, Apr 18, 2008 at 3:45 AM, Vladimir Beliaev <
[EMAIL PROTECTED]> wrote:
According to the issue quick evaluation the tests are invalid, so noted
revision helped to detect them. The tests are to be excluded now.
Tim, Alexei, thank you for your input.
Thanks
Vladimir
2008/4/18, Alexei Fedotov <[EMAIL
PROTECTED]<https://mail.google.com/mail?view=cm&tf=0&[EMAIL PROTECTED]>
:
Tim,
We are sorting that out.
Thanks.
On Fri, Apr 18, 2008 at 12:21 PM, Tim Ellison <[EMAIL
PROTECTED]<https://mail.google.com/mail?view=cm&tf=0&[EMAIL PROTECTED]>
wrote:
It'll be a couple of days before I can look at it as I'm going away.
I'll
take a look when I get back if nobody beats me to it.
Regards,
Tim
Vladimir Beliaev wrote:
Tim,
I've opened
HARMONY-5747<https://issues.apache.org/jira/browse/HARMONY-5747>
which
reports regression in EUT/update suite due to r641928
[apply patch
HARMONY-4569<https://issues.apache.org/jira/browse/HARMONY-4569>
([classlib][performance]
Ineffecient manifest parsing results in slowdown when debugging java
code)]
Could you take a look at regression & get it resolved, please?
Thanks
Vladimir
--
With best regards,
Alexei
