You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cayenne.apache.org by Aristedes Maniatis <ar...@maniatis.org> on 2016/02/04 11:16:33 UTC
Re: work on ROP
On 21/01/2016 11:39pm, Dzmitry Kazimirchyk wrote:
> One of the questions here is still whether we want to go all the way and introduce changes which can potentially brake things for upgrading users or we should preserve the old HessianROPServlet and the logic around it and have a new pluggable approach as an alternative to what we currently have.
What is the reason that half the implementation is in org.apache.cayenne.rop and the other half in org.apache.cayenne.configuration.rop? Why are some http parts in org.apache.cayenne.rop.http but not all of them?
Given that we are breaking backward compatibility for people who have overridden specific bits of this, should we tidy up?
Ari
--
-------------------------->
Aristedes Maniatis
GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A
Re: work on ROP
Posted by Aristedes Maniatis <ar...@maniatis.org>.
On 4/02/2016 9:16pm, Aristedes Maniatis wrote:
> Why are some http parts in org.apache.cayenne.rop.http but not all of them?
OK, should we move just this bit then?
Ari
--
-------------------------->
Aristedes Maniatis
GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A
Re: work on ROP
Posted by Andrus Adamchik <an...@objectstyle.org>.
ServerRuntime is in "org.apache.cayenne.configuration.server", so "configuration.rop.client" is kind of symmetrical.
A.
> On Feb 4, 2016, at 1:36 PM, Dzmitry Kazimirchyk <dk...@gmail.com> wrote:
>
> There are mainly DI module configurations in "configuration.rop" packages, which I think makes sense. However in "configuration.rop.client" there are also ClientRuntime classes, which I'm guessing can be moved "org.apache.cayenne.rop" package for consistency.
>
> Dima
>
> On 2/4/16 1:16 PM, Aristedes Maniatis wrote:
>> On 21/01/2016 11:39pm, Dzmitry Kazimirchyk wrote:
>>> One of the questions here is still whether we want to go all the way and introduce changes which can potentially brake things for upgrading users or we should preserve the old HessianROPServlet and the logic around it and have a new pluggable approach as an alternative to what we currently have.
>>
>>
>> What is the reason that half the implementation is in org.apache.cayenne.rop and the other half in org.apache.cayenne.configuration.rop? Why are some http parts in org.apache.cayenne.rop.http but not all of them?
>>
>> Given that we are breaking backward compatibility for people who have overridden specific bits of this, should we tidy up?
>>
>> Ari
>>
>>
>
Re: work on ROP
Posted by Dzmitry Kazimirchyk <dk...@gmail.com>.
There are mainly DI module configurations in "configuration.rop"
packages, which I think makes sense. However in
"configuration.rop.client" there are also ClientRuntime classes, which
I'm guessing can be moved "org.apache.cayenne.rop" package for consistency.
Dima
On 2/4/16 1:16 PM, Aristedes Maniatis wrote:
> On 21/01/2016 11:39pm, Dzmitry Kazimirchyk wrote:
>> One of the questions here is still whether we want to go all the way and introduce changes which can potentially brake things for upgrading users or we should preserve the old HessianROPServlet and the logic around it and have a new pluggable approach as an alternative to what we currently have.
>
>
> What is the reason that half the implementation is in org.apache.cayenne.rop and the other half in org.apache.cayenne.configuration.rop? Why are some http parts in org.apache.cayenne.rop.http but not all of them?
>
> Given that we are breaking backward compatibility for people who have overridden specific bits of this, should we tidy up?
>
> Ari
>
>