[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [mdt-ocl.dev] Question regarding submitting EMF Event Manager
|
Hi,
well, there seems to be controversy about what a committer can commit
and what not. In order not to compromise IP, let's shoot for the safe
side. I'll raise an IPzilla as soon as someone is as kind as sending me
the link for the CQ form or a link to the IP Policy which unfortunately
I was so far unable to locate in the portal (which currently I find
difficult to navigate and search...).
I'm glad we seem to have agreement that we can (given IP clarification)
commit the event manager to OCL examples.
I hope you're fine with me committing the 323223 stuff in slices, one of
which is the event manager. The other things, as I noted also in the
respective Bugzillas, are pending other Bugzillas.
Thanks and best,
-- Axel
On 1/10/2011 6:21 PM, Ed Willink wrote:
Hi Axel
I do not have a problem with the EventManager being hosted as an OCL
Examples plugins in Indigo. This provides a convenient 'Incubation'
location prior to promotion to one of:
- EMF core plugins
- an EMF sub-project
- OCL core plugins
in Indigo+1.
This is I thought what we were doing in
https://bugs.eclipse.org/bugs/show_bug.cgi?id=323223.
Unfortunately you jumped the gun on committing to CVS; now corrected.
Since this is a non-trivial body of code written by (at least one)
non-committer (at the time), it must be subjected to Eclipse IP
approval, and not committed until the checkintocvs keyword is attached
to the IPzilla. Your current status as a committer and SAP's status as
an Eclipse member should expedite IP approval, but IP approval is required.
I was surprised when you referred to
https://bugs.eclipse.org/bugs/show_bug.cgi?id=329336, which seems to
represent a duplicate submission of the code. I have WONTFIXed 329336
and introduced the missing linkage to/from 323223 which will hopefully
avoid further confusion.
Regards
Ed Willink
On 10/01/2011 13:06, Adolfo Sánchez-Barbudo Herrera wrote:
Hi all,
Coming back from my holidays break...
To be constructive... Axel what about if you try to obtain some
feedback in the bugzilla or somewhere about when (1 week, 1 month,
post indigido, never ?) the contribution could be reviewead and/or
committed ?. Depending on such an estimation we could take some
decisions about it. We all want to make the different pieces to be
placed at the proper place... From my experience, if we committ the
event manager into OCL, it will probably stay there forever.... So my
preference is waiting until the event manager is committed into EMF,
where it should really do exist...
Best Regards,
Adolfo.
El 07/01/2011 0:05, Uhl, Axel escribió:
Ed,
-----Original Message-----
From: Ed Willink [mailto:ed@xxxxxxxxxxxxx]
Sent: Thursday, January 06, 2011 11:51 PM
To: Uhl, Axel
Cc: Adolfo Sánchez-Barbudo Herrera
Subject: Re: [mdt-ocl.dev] Question regarding submitting EMF Event
Manager
Hi Axel
As you agreed in an earlier response, you must remove the event manager
and (impact analyzer) from CVS because it contains unapproved material.
It seems some is not even intended for OCL at all. If it is still there
tomorrow, I will have to remove it myself.
I "cvs rm"ed the contents, but of course they are still in the
history. My ssh account on dev.eclipse.org is restricted, so I can't
wipe it from /cvsroot/... by any means. If you can, please go ahead.
When I indicated that you could commit to event/impact without
approval,
I intended that lack of oversight to apply to minor/moderate
maintenance, not initial/major commit where Bugzillas await residual
review. I was not aware that the event manager was intended for EMF. So
long as this is the case it should not be committed to OCL.
Ed Merks hasn't done anything in a few months now. I don't see a
problem with first committing this to OCL's examples section. If Ed
Merks later thinks it should go into EMF Core, fine, then I'm happy
to support him in moving it there. Until that happens, we need this
for the Impact Analyzer. I won't hold the Impact Analyzer back for
this solvable problem.
It's very odd that Ed has made no comment at all. Reviewing the
comments, it is now very unclear that he needs to comment since you say
you're migrating to OCL. However he may wake up in his PMC role and ask
why the OCL commit has happened without IP approval...
So long as the EMF Bugzilla is open and a comment is expected from Ed,
there should be an attachment that Ed can sensibly review. Please
provide it; a patch with org.eclipse.emf... names.
Ed Merks made it clear at ESE that *he* will look into naming and
other implementation details after I explicitly offered my
assistance. That's as much as I can do.
I'm not clear why you request permission today to do what you did
yesterday in breach of committer guidelines. Please read the guidelines
Could we please stop this? I admitted my novice mistake and
apologized. I'm now trying to get it fixed. It would be helpful if
you assisted me in this rather than keeping reminding me of my
mistakes. I asked whether top open a new bug for the MDT/OCL
component in preparation of an IP review for a later commit to that
component. You now suddenly suggest you don't want the event manager
bundle in the examples section although you know the impact analyzer
depends on it. This would block the whole thing until Ed Merks moves
which he hasn't since ESE. This is unnecessary, in my opinion,
because---as I mentioned above---the one event manager bundle can
easily be migrated to EMF Core later. It's just a package rename, and
only from an *examples* package.
Therefore, I just would like to know what the best way is to get IP
approval for committing this to the MDT/OCL examples section. If you
plan to veto this, let me know now, because then this would mean that
the Impact Analyzer gets blocked by Ed Merks' inactivity on
https://bugs.eclipse.org/bugs/show_bug.cgi?id=329336 until who knows
when.
Best,
-- Axel
carefully. The integrity of Eclipse repositories is VERY important.
Breaking the guidelines leaves both of us in trouble; advertising the
breach on Bugzilla is not very helpful either.
Please wait for other committers to comment before taking unilateral
actions and please ask for advice when in doubt.
Regards
Ed Willink
On 06/01/2011 20:49, Axel Uhl wrote:
The event manager is a pre-requisite for the OCL Impact Analyzer. At
ESE Ed Merks had indicated he would want to integrate it somwhere in
EMF Core. That's why I created
https://bugs.eclipse.org/bugs/show_bug.cgi?id=329336. Meanwhile, Ed
Merks may not have found the time to look into this. I think this
shouldn't hold up the progress of the Impact Analyzer. Therefore, I'd
like to commit the event manager to the OCL examples section.
Question: For IP review, should I use the existing
https://bugs.eclipse.org/bugs/show_bug.cgi?id=329336 bug or should I
create a new bug specific to OCL/MDT?
Best,
-- Axel
_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1191 / Virus Database: 1435/3363 - Release Date:
01/06/11
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1191 / Virus Database: 1435/3367 - Release Date: 01/08/11
_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev
