Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: HA: [mdt-ocl.dev] Compatibility Support

Hi

Great, I think we are agreed. 1.4.x is an enhanced maintenance release for Helios. Since we will be maintaining 1.3.0, 1.3.1, 1.3.2 in parallel with 2.0.0, supporting 1.4.0 can't be that different. Perhaps 1.4.0 gets no further maintenance release to avoid a 1.4.1, 2.0.1 and 2.1.0; not today's
problem anyway.

One minor problem; API tooling.

Because UML2 went from 2.x to 3.x for Galileo, the MDT-OCL UML plugin had to go from 1.x to 2.x, so 1.3.x and every subsequent release has a mismatch of MDT-OCL numbers; 2.0.0 will have a 3.0.0 for UML. Do we want to skip 2.0.0 and go straight to 3.0.0 (any of my earlier comments for 3.0.0 features should then be interpreted as 4.0.0)? This has an advantage in avoiding any confusion
that MDT OCL 1.x is OCL 1.x and MDT-OCL 2.0.0 is OCL 2.x.

If EMF or UML2 or ... has a major API change in Helios, then we cannot go to 1.4.0 for a maintenance release while complying with API tooling policy. If this happens, I think we perform API testing in so far as possible ignoring only the major increment. i.e. in our workspace we temporarily name the 1.4.0 plugins as 2.0.0 check the API compliance, then restore the name to 1.4.0 and distribute with a
major version API exclusion  filter.

   Regards

      Ed Willink



Alexander Igdalov wrote:
Hi Adolfo and Ed,
Adolfo, I agree with your description of MDT OCL 1.4.0 and 2.0.0. I also hope it is possible to practically implement this idea. Different versions of several plugins peacefully co-exist in Galileo, e.g. javax.wsdl 1.5.1 and 1.6.2. There are projects which depend on the earlier version of the plugin while others depend on the newer. Best,
Aelx.
------------------------------------------------------------------------
*From:* mdt-ocl.dev-bounces@xxxxxxxxxxx [mailto:mdt-ocl.dev-bounces@xxxxxxxxxxx] *On Behalf Of *Adolfo Sanchez-Barbudo Herrera
*Sent:* Friday, July 17, 2009 5:36 PM
*To:* MDT OCL mailing list
*Subject:* Re: HA: [mdt-ocl.dev] Compatibility Support

Hi Ed,

As far as I have understood 1.4.0 is intended to be shipped in Helios as a version of the current status of MDT-OCL as a (probably incorrect, incomplete, etc) implementation of the OCL 2.0 specification. The efforts needed could be updating dependencies and solving any issue arisen from changes in the dependent projects (EMF, UML, ...). Therefore, any actual MDT-OCL client could perfectly fit in Helios release, without any backward compatibility problem.

If clients want to align with the OCL 2.2 specification and/or take advantage of any new feature or enhancement introduced in Helios, they will have to move on the MDT-OCL 2.0.0. We will be mainly focused on this release....

Again, I have some doubts about if this, which seems to be conceptually sensible, can be done in the practice.... which would need some help from a releng expert (yeah, Nick could add a valuable point of view).

Cheers,
Adolfo.

Ed Willink escribió:
When I wrote:
If the sole purpose of 1.4.0 is to provide a slightly more confidence inspiring name than 1.3.3 for Helios then I have no problem with someone doing that. I just don't have time to contribute to any problems that arise as a result. If the intention is to offer more than maintenance functionality in 3.3 please elaborate.

I think we should proceed with OCL 2.2 as far as we can understand it on MDT-OCL 2.0.0 and address any real compatibility issues as they arise.

I was trying to find out what option A means. I would still like to know what is planned for inclusion in 1.4.0, for which I provided my best guess above. (Alex's original proposal discussed the practicalities of A/B but did not discuss content.)

In the above I was sort of suggesting that perhaps we do both A and B since there may not really be very much difference.

Moving to a vote is not very helpful, because I certainly do not understand what we are actually voting on. We will only have another discussion later on as to what the result of the vote meant.

   Regards

      Ed Willink

_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev


--

	*Adolfo Sánchez-Barbudo Herrera*
adolfosbh(at)opencanarias(dot)com <mailto:adolfosbh%28at%29opencanarias%28dot%29com>
C/Elías Ramos González, 4, ofc. 304
38001 SANTA CRUZ DE TENERIFE
Tel.: +34 922 240231 / +34 617 718268

------------------------------------------------------------------------

_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev




Back to the top