Hi Marek
On 28/08/12 14:35, Marek Potociar wrote:
> Hi Sergey,
>
> This issue was discussed already (a long time ago though), but we didn't seem to agree on anything particular. So I thought I'd try this approach that may work better.
>
Indeed, I recall it now
> Also, I'd like to make it clear that the proposals I sent for review are certainly not carved into stone (see all the changes we're still making even in the API's that were approved earlier). If the EG decides we should ditch the API and rather defer the issue resolution to another release, we will remove the API. If EG decides to change the API, we will change it.
>
Sure
> What I could do instead is to always create new branches and first commit the proposals to these branches. Once approved I could merge those to master. However, again, this approach was earlier consider by this EG as too difficult to follow and committing directly to master seemed to be the preferred way. The changes are more readily viewable on the web site, in the snapshot API docs or maven artifacts etc. But, if we want to revisit this and perhaps change the process, we can.
Definitely no need to do it, it will be just an extra hassle for you
>
> Now to the issue - it's true that this issue was initially reviewed by EG and it was considered as CRITICAL - IOW something that MUST be addressed in JAX-RS 2.0. But that still does not imply that we have to accept my proposal, if there's a better solution.
>
OK. All know that a lot of attention went into creating this API, as far
as I'm concerned I'll just need to revisit again what it actually helps
with, hope I'll provide a better feedback
Cheers, Sergey
> Marek
>
> On Aug 28, 2012, at 3:02 PM, Sergey Beryozkin<sberyozkin_at_talend.com> wrote:
>
>> By the way, as far as the process is concerned, should we have what are seen by some people as critical api/spec issues discussed at this list first ?
>> I can see/understand why going ahead with the actual fix and then asking for the review works, especially when the time is limited and such, but I'm not sure this is the way it should work; to be honest I do not see much scope for an api like this be removed now that it's been added - we are now working with the implicit presumption this is an absolutely essential fix.
>>
>> Sergey
>>
>>
>>
>> On 24/08/12 17:34, Marek Potociar wrote:
>>> Experts,
>>>
>>> as a resolution of a critical issue
>>> http://java.net/jira/browse/JAX_RS_SPEC-72 a new ResourceContext API has
>>> been introduced. Please see the change here:
>>> http://java.net/projects/jax-rs-spec/sources/git/revision/d6a07ec69f2f0bae5c93f641247a0a493434f52b
>>>
>>> Your review feedback& comments are welcome.
>>>
>>> Many thanks,
>>> Marek
>>
>>
>