Re: [websocket-dev] Can we start work?

[Mark Thomas <markt@xxxxxxxxxx>]

On 03/04/18 13:13, Roman Grigoriadi wrote:
> We can start work now.

Excellent.

> I suggest PRs to be created and eventually
> accepted to master branch.

Is the expectation that every change comes in via a PR? Or can
committers commit directly to master?

My preference is to keep the process as light as possible and allow
committers to commit directly to master but expect them to use judgement
as to when going via a PR to allow discussion is appropriate.

> First release under Eclipse umbrella will be
> done from EE4J branch, which should be pretty much the same as code in
> master during donation. Changes from master will be picked up by next
> releases following EE4J branch release.

What is the process for agreeing changes to back-port from master to the
EE4J branch? There are, for example, a number of Javadoc typos that have
been reported that it makes sense to fix in the first release.

> Please see my comments inline.
> 
> On 16.3.2018 08:44, Mark Thomas wrote:
>> On 15/03/18 07:04, Roman Grigoriadi wrote:
>>> Hi,
>>>
>>> although repositories are now visible under eclipse-ee4j project
>>> transfer is not complete.  All projects are required to have their
>>> intellectual property contribution approved and build / integration
>>> pipelines migrated to Eclipse infrastructure.
>>> Until approved by PMC
>>> (https://projects.eclipse.org/projects/ee4j/pmc) we are not accepting
>>> any contributions besides critical ones or those required to make JavaEE
>>> donation possible. However this does not mean PRs  cannot be created.
>> Who is the "we" in that sentence?
> We - websocket committers.
>> Looking at [1] the IP contribution has been approved.
> You're right, IP contribution is complete for websocket.
>> [2] suggests the CI pipeline is still to be migrated. Is that in hand
>> somewhere or are we (the websocket committers) meant to be doing
>> something to make that happen?
> We (contributors working on donation) will migrate initial pipelines for
> both Websocket and Tyrus. You are welcome to participate on Jenkins
> setup if you like, but you are not meant / don't have to do so.

I'd help if I could but that isn't my area of expertise. If it is in
hand then I'm happy to leave it to folks that know more than me.

Mark


> 
> Thank you,
> Roman
>> I also assume we can have discussions on the issues as we all as create
>> PRs for the ones that need fixing and closing the ones that are invalid.
>>
>> Mark
>>
>>
>>
>> [1] https://dev.eclipse.org/ipzilla/show_bug.cgi?id=15766
>> [2] https://ci.eclipse.org/
>>
>>> Looking forward to integrate fixes for #208 and #240 when possible.
>>>
>>>
>>> Thank you,
>>> Roman
>>>
>>>> On 14 Mar 2018, at 23:46, arjan tijms <arjan.tijms@xxxxxxxxx
>>>> <mailto:arjan.tijms@xxxxxxxxx>> wrote:
>>>>
>>>> Hi,
>>>>
>>>> I wonder about this too, and also about the moment when it's actual
>>>> possible to do make API changes (even if only in an experimental branch).
>>>>
>>>> The RI just arrived at GitHub as well,
>>>> see https://github.com/eclipse-ee4j/tyrus
>>>>
>>>> From the JSF perspective we're still quite eager to fix this
>>>> one: https://github.com/eclipse-ee4j/websocket-api/issues/240
>>>>
>>>> Currently I have implemented a workaround in Mojarra that's reflective
>>>> based, Tyrus specific, and nothing short of an ugly hack. I'd love to
>>>> replace that by a proper solution.
>>>>
>>>> Kind regards,
>>>> Arjan Tijms
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Wed, Mar 14, 2018 at 11:20 PM, Mark Thomas <markt@xxxxxxxxxx
>>>> <mailto:markt@xxxxxxxxxx>> wrote:
>>>>
>>>>     Hi all,
>>>>
>>>>     I see that the API code has made it to GitHub:
>>>>     https://github.com/eclipse-ee4j/websocket-api
>>>>     <https://github.com/eclipse-ee4j/websocket-api>
>>>>
>>>>     Woot!
>>>>
>>>>     Are we able to start committing?
>>>>
>>>>     I understand that we don't want to make any API changes at this point.
>>>>     However, there are some bug reports like this one:
>>>>     https://github.com/eclipse-ee4j/websocket-api/issues/208
>>>>     <https://github.com/eclipse-ee4j/websocket-api/issues/208>
>>>>
>>>>     that relate to copy/paste errors in the Javadoc which should be
>>>>     safe to fix.
>>>>
>>>>     Is there any reason we shouldn't start triaging the open bugs and
>>>>     fixing
>>>>     issues like 208 that don't have an API impact?
>>>>
>>>>     Mark
>>>>     _______________________________________________
>>>>     websocket-dev mailing list
>>>>     websocket-dev@xxxxxxxxxxx <mailto:websocket-dev@xxxxxxxxxxx>
>>>>     To change your delivery options, retrieve your password, or
>>>>     unsubscribe from this list, visit
>>>>     https://dev.eclipse.org/mailman/listinfo/websocket-dev
>>>>     <https://dev.eclipse.org/mailman/listinfo/websocket-dev>
>>>>
>>>>
>>>> _______________________________________________
>>>> websocket-dev mailing list
>>>> websocket-dev@xxxxxxxxxxx <mailto:websocket-dev@xxxxxxxxxxx>
>>>> To change your delivery options, retrieve your password, or
>>>> unsubscribe from this list, visit
>>>> https://dev.eclipse.org/mailman/listinfo/websocket-dev
>>>
>>> _______________________________________________
>>> websocket-dev mailing list
>>> websocket-dev@xxxxxxxxxxx
>>> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
>>> https://dev.eclipse.org/mailman/listinfo/websocket-dev
>>>
>> _______________________________________________
>> websocket-dev mailing list
>> websocket-dev@xxxxxxxxxxx
>> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/websocket-dev
> 
> 
> 
> _______________________________________________
> websocket-dev mailing list
> websocket-dev@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
> https://dev.eclipse.org/mailman/listinfo/websocket-dev
>