I agree to utilize existing implementation. But there are still others missing. For example, java.util.Collection and its subclasses[1]. [1]http://issues.apache.org/jira/browse/HARMONY-4327
Yes, this is true. We don't have persistence delegates for collection types yet. If you like to implement some of them - good news, go ahead. :) However, I think it make sense to keep all persistence delegates in one package - currently it is "org.apache.harmony.beans".
I agree to regard these test cases as API tests. And then these persistenceDelegate implementations better be placed at src/main/java/java/beans corresponding to test case layout.
Please see my previous comment to Paulex's letter. Regards, 2007/7/5, Spark Shen <[EMAIL PROTECTED]>:
2007/7/5, Yang Paulex <[EMAIL PROTECTED]>: > > 2007/7/4, Alexei Zakharov <[EMAIL PROTECTED]>: > > > > Spark Chen wrote: > > > > If no objection, I will go to implement those missing persistence > > > > functionality. > > > > FYI we already have number of persistence delegates classes located in > > org.apache.harmony.beans package, and FieldPersistenceDelegate is one > > of them. I agree to utilize existing implementation. But there are still others missing. For example, java.util.Collection and its subclasses[1]. [1]http://issues.apache.org/jira/browse/HARMONY-4327 It contains the algorithm that is very close to one you've > > implemented in the patch for HARMONY-4321. So I don't think we should > > add new FieldPersistenceDelegate, let's fix the existing one instead > > (if it needs fixing). BTW, we also have a special folder > > (src/test/java-internal/org/apache/harmony/beans) where tests for > > persistence delegates are located. Shouldn't we create something like > > FieldPersistenceDelegateTest there? > > > Seems the tests in src/test/java-internal/org/apache/harmony/beans are > implementation tests, all of which fail on RI with message like " > java.lang.NoClassDefFoundError: > org/apache/harmony/beans/ArrayPersistenceDelegate ". But I think the > persistent delegate mechanism are very similar with serialization, on > which > we need to be compatible with other Java SE implementation, so it makes > sense to make them the API tests which pass on RI. Did I miss something? I agree to regard these test cases as API tests. And then these persistenceDelegate implementations better be placed at src/main/java/java/beans corresponding to test case layout. Thanks, > > > > 2007/7/3, Tony Wu <[EMAIL PROTECTED]>: > > > On 7/3/07, Spark Shen <[EMAIL PROTECTED]> wrote: > > > > I find beans.XMLEncoder does not persist > java.lang.reflect.Fieldproperly. > > > > And I suspect there are more classes not properly handled. > > > > > > > > > > Yes, I think so. > > > > I have reported a JIRA: > > > > https://issues.apache.org/jira/browse/HARMONY-4321 > > > > > > > > If no objection, I will go to implement those missing persistence > > > > functionality. > > > > > > > > > > please go head:) > > > > > > > > -- > > Alexei Zakharov, > > Intel ESSD > > > > > > -- > Paulex Yang > China Software Development laboratory > IBM > -- Spark Shen China Software Development Lab, IBM
-- Alexei Zakharov, Intel ESSD
