harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alexei Zakharov (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HARMONY-975) [classlib][beans] fixes for DefaultPersistenceDelegateTest failures (1)
Date Wed, 02 Aug 2006 07:27:15 GMT
    [ http://issues.apache.org/jira/browse/HARMONY-975?page=comments#action_12425161 ] 
Alexei Zakharov commented on HARMONY-975:


I think I need to describe what I am doing for beans module a little bit. Currently we have
a lot of failed tests for beans - they came from the separate contribution and were initially
designed for testing of some third-party implementation of beans. The most common reasons
of these failures are:
1. The test checks some RI-specific behavior that is not specified by java beans spec. It
was decided (in list) and also was stated here [1] that we need to follow the RI behavior
if it is natural and doesn't violate the common sense.
2. The test checks some third-party implementation specific behavior. Such tests should be
marked as invalid and rewritten to follow the RI behavior or removed if this is not possible.
3. The test checks the behavior specified by the java beans spec our implementation currently
violates. We need to conform with the spec.

Four of five tests mentioned above fall to (1) category. These tests check our implementation
of DefaultPersistenceDelegate for exception throwning compatibility with RI.  RI behavior
is reasonable here. Moreover, our implemention's old behavior is illogical - we just catch
all exceptions and do System.out.println(e.toString()) and nothing more. We don't provide
any error information to the user application. So for DefaultPersistenceDeleage I decide to
change the exception handling model. With the attached patch it conforms to the RI behavior.

Does this make sense?


[1] http://incubator.apache.org/harmony/subcomponents/classlibrary/compat.html

> [classlib][beans] fixes for DefaultPersistenceDelegateTest failures (1)
> -----------------------------------------------------------------------
>                 Key: HARMONY-975
>                 URL: http://issues.apache.org/jira/browse/HARMONY-975
>             Project: Harmony
>          Issue Type: Improvement
>          Components: Classlib
>         Environment: winXP
>            Reporter: Alexei Zakharov
>         Attachments: DefailtPersistenceDelegateTest.patch, DefailtPersistenceDelegateTest2.patch
> Currently 19 tests from org.apache.harmony.beans.tests.java.beans.DefaultPersistenceDelegateTest
class fail due various reasons. The attached patch corrects some of these failures. Please
see the list of affected tests below.
> testInstantiate_NullProperty
> testInstantiate_EmptyProperty
> testInstantiate_NonExistingGetterNullEncoder
> Invalid tests indeed. Fail if NullPointerException isn't thrown. Neither RI nor our implementation
throws such exception here. Corrected. 
> testInstantiate_NonExistingGetter
> testInstantiate_NoGetter
> Are also invalid tests. They test the behavior that is specific to some third-party implementation
of beans (not RI). I corrected the sophisticated but invalid exception handling I've found
in these tests. Now it reflects the RI behavior. Please note that using of singleton class
CallVerificationStack is thread unsafe here. I have replaced CallVerificationStack with locally
defined variable.

This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message