Robert,

I thought I would just summarise what I've done on beanutils to date, so
that its easier for you (or others to review) - theres basically four main
pieces of work which I highlight in the "Summary" below, but I've also
detailed all the bugs I changed under "Detail" below.

SUMMARY
=========
1)   Committed fixes for Bug 28684 to the "numeric" locale converters and
added test cases.
2)   Added patch for review to Bug 29772 to fix issues in
DateLocaleConverter and addtional tests
3)   Added patch for review to Bug 23815 to fix issues in
PropertyUtilsBean - however these changes are also incorporated in the patch
attached to Bug 28813 below and its that patch that I hope to committ.
4)   Added patch for review to Bug 28813 to fix issues in
PropertyUtilsBean - working on addtional tests which I will attach to the
bug for review when I've completed them.


DETAIL
======

1) Bug 28254    getIndexedProperty problems on JDK 1.4.1
http://issues.apache.org/bugzilla/show_bug.cgi?id=28254

I closed this because it was a duplicate of Bug 28358 (same reporter and
same wording!)


2) Bug 28684    BigDecimalLocaleConverter seems not working as designed
http://issues.apache.org/bugzilla/show_bug.cgi?id=28684

BigDecimalLocaleConverter was basically returning the results of whatever
the DecimalFormat.parse() returned - i.e. either a Double or Long. The same
issue also existed in three of the other locale converter clases. This IMO
was a pretty straight forward bug and so I have actually committed fixes for
these and added unit tests for all the "numeric" locale converters and fixed
another small bug I found while testing.


3) Bug 29772    DateLocaleConverter does NOT return a default value if a
conversion error occurs
http://issues.apache.org/bugzilla/show_bug.cgi?id=29772

This bug I think might already have been resolved by previous changes.
Anyway I decided to add some more tests to DateLocaleConverterTestCase to
see if I could reproduce the problem and I found that while it worked under
JDK 1.3, I had a number of problems under JDK 1.4.  I have attached  a patch
to the bug with a workaround for these issues along with the modified Test
Case. Feedback on whether you (or anyone else) are happy for me to apply
this patch would be much appreciated.


4) Bug 23815, 26904 and 29571

Bug 23815   PropertyUtils.getNestedProperty() doesn't allow getXxxx on
Map-Instances any longer
Bug 26904  (g|s)etSimpleProperty fails when passing a Map
Bug 29571  Inconsistent behaviour of BeanUtils.setProperty() and
BeanUtils.getProperty() methods
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=23815
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=26904
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=29571

These bugs all related to the same issue. Craig McClanahan posted concerns
in one of the bugs that what was being requested would break backward
compatibility. I believe that the change in behaviour I'm proposing
satisfies both camps and resolves the inconsistencies between some of the
methods. I've documented on Bug  23815 the changes in behaviour to
PropertyUtilsBean I'm proposing.


5) Bugs 23108 and 28813

Bug 28813    Can't use . (dot) in mapped properties for setProperty or
getPropertyDescriptor
Bug 23108    Setting complex mapped properties using PropertyUtils
http://issues.apache.org/bugzilla/show_bug.cgi?id=28813
http://issues.apache.org/bugzilla/show_bug.cgi?id=23108

Both these bugs deal with the same issue so I closed Bug 23108 as a
duplicate of 28813

PropertyUtils was changed in Bug 10478 so that "mapped" propery keys could
contain dot notation - but only the "get" method was changed, not the "set"
or getPropertyDescriptor" methods. I have attached a patch to the bug with
the changes to PropertyUtilsBean - the changes in the "diff" file look more
complex than they seem - all I did was copy the "while()" loop logic from
the get method to the set and getPropertyDescriptor methods, so that they
are all working consistently the same way.

These bugs were for the same class (PropertyUtilsBean) as for the Bugs in 4)
so the patch attached to Bug 28813 also includes the changes for the bugs in
4). Feedback on whether you (or anyone else) are happy for me to apply this
patch would be much appreciated.

I'm currently working on adding additional tests for this patch and will
attach it to the Bug 28813 when I've finished it.


Niall

----- Original Message ----- 
From: "robert burrell donkin" <[EMAIL PROTECTED]>
To: "Jakarta Commons Developers List" <[EMAIL PROTECTED]>
Sent: Sunday, July 11, 2004 9:53 PM
Subject: Re: [beanutils] LazyDynaClass and LazyDynaBean Bug [29879]


> cool.
>
> i see you've added reviewed a good number of bugs reports (thanks).
> i'll try to find time to go through as many as possible of them
> tomorrow.
>
> the general plan for beanutils (and some background) is on the wiki. we
> need to cut a 1.7.0 service release that will give compatibility with
> both collections 2.x and 3.x. this will enable a similar release to be
> cut for digester and then (hopefully) one for struts as well.
>
> since beanutils has a load of outstanding bugs and enhancements against
> it, the plan is to try to aim for a relatively quick 1.7.1 bug
> fix/enhancement release.
>
> the major item of work remaining before 1.7.0 is released is a review
> of the interfaces for the new beanified classes. i'll probably post
> some more about this is a better named thread. i'd be grateful if you'd
> take a look and comment.
>
> - robert



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to