Thanks for the updates. Some comments inline:

On Thu, Jun 19, 2008 at 2:13 PM, Haibo Zhao <[EMAIL PROTECTED]> wrote:
> Hi Luciano,
>
>    No, not at all, just forgot to cc to the list
>
>    I am Working on GdataBindingInvoker.GetInvoker() and having some
> issues, will submit the updated patch via JIRA when I clear this part.
>
>    Thanks,
>    Haibo
>
> On Wed, Jun 18, 2008 at 5:21 PM, Luciano Resende <[EMAIL PROTECTED]> wrote:
>> Would you mind sending this to the list.
>>
>> Also, you should provide the patch using the JIRA application, and
>> discuss the aproach/design you want to take on the list as well, see
>> how others from GSoC are doing.
>>
>> On Tue, Jun 17, 2008 at 11:12 PM, Haibo Zhao <[EMAIL PROTECTED]> wrote:
>>> Hi Luciano,
>>>
>>>  Please check the project blog for the 5th weekly progress report:
>>>
>>>  Link:    http://gdatabindingtuscany.blogspot.com
>>>
>>>  Please see the attached patch file, which is still a very very
>>> preliminary thing, more like a "hello world" version. I basically
>>> modify the atom-binding code and change the binding type name, the
>>> corresponding configuration files, package names and class names.
>>>

Looks like a good start. I'd recommend having the module called
something like binding-atom-gdata, if this is kind of extension to the
atom binding. Also, looks like you added to much into your patch
(check how to configure svn:ingnore in our dev guide) and you also
concatenate the diff multiple times in the same patch, causing same
file to be showing multiple times on the patch... but it's a good
start, let me know when you have a new patch available, and let's
start using the jira to provide the patch.

>>>  After reading and modifying the Atom-binding code, I have some
>>> notes(comments in atom-binding classes) in atom-binding classes about
>>> the changes I am planning to make. These nodes are included in the
>>> patch.
>>>
>>>  Luciano, I am having a couple of questions:
>>>
>>> (1) Is this the right direction to implement gdata-binding? I mean,
>>> modifying the Atom-binding code. I feel like this should be OK because
>>> the major different is the model used in Abdera and GData, the way the
>>> entries can be parsed from the message and the data conversion from/to
>>> collection items. So could you confirm me that I am heading a right
>>> direction?
>>>

I guess there are two ways of getting this GData integration done, one
by creating a totally new binding (Douglas approach) and another way
would be extending the current Atom support (your aproach)... so I
guess you are in the right direction.

>>> (2) It seems that the authentication part is not completed in Atom-binding.

Yes, but I also noticed that it looks like Abdera already have some
extensions supporting GData authentication, it would be great to take
a look and see how it could be used to our advantage.


Also note that you and Douglas are going to have the same requirements
regarding availability of GData dependencies in a maven repo. Let's
collaborate in finding a solution for this issue.


>>>
>>>
>>>
>>>
>>>  Thanks a lot,
>>>
>>>  Haibo
>>>
>>
>>
>>
>> --
>> Luciano Resende
>> Apache Tuscany Committer
>> http://people.apache.org/~lresende
>> http://lresende.blogspot.com/
>>
>



-- 
Luciano Resende
Apache Tuscany Committer
http://people.apache.org/~lresende
http://lresende.blogspot.com/

Reply via email to