On 4/18/12 11:16 , Per-Erik Svensson wrote:
Hi and thank you,
What I take from this is that there will never be a fully standardized way
of doing what Felix OBR does (or at least not in a foreseeble future) so I
don't have to "feel bad" in picking a specific implementation. Part of my
concern was picking an implementation only to find out that there is a
standard way of doing the same thing in the pipeline of spec-writers.
Yes, there currently isn't a standard way in the pipeline, but never say
never.
-> richard
Thanks for the responses!
Best regards,
Per-Erik Svensson
On Wed, Apr 18, 2012 at 5:04 PM, Richard S. Hall<[email protected]>wrote:
On 4/18/12 08:12 , Neil Bartlett wrote:
I assume you're talking about the new Resolver and Repository
specifications that are in the OSGi Enterprise R5 release? Because "OBR" is
properly the name for the Felix implementation, which has been available
and released for several years.
An early draft release of the Enterprise R5 spec was made publicly
available in March and can be downloaded now from the OSGi site. The
official final release of these specs will be some time later, depending on
votes and lawyers etc, however the technical content is very unlikely to
change.
I believe that the Felix org.apache.felix.**bundlerepository
implementation will NOT be updated to comply with the new R5 specs. Richard
is working on the RI for the R5 spec within Felix but it is new code, based
on the resolver in the core Felix framework.
I would like to see the Felix OBR implementation evolve to use the new R5
resolver and repository specs, for sure.
Per-Erik, your confusion is likely due to the fact that the scope of
RFC-112 has changed over the course of its lifetime. Originally, it was an
API that was the precise basis for the Felix OBR implementation. Over time,
some holes were discovered in the API and Felix OBR evolved into having its
own API and well as supporting the original RFC-112.
During the spec process for RFC-112, it became clear that the scope was
too big (and there were too many bike shed issues) to complete it in a
reasonable amount of time, so the scope was pared back (even then it still
took a long time to complete). So, originally, RFC-112 dealt with
repositories, resolving, and deployment/provisioning, but now it only deals
with repositories and resolving. Further, there is no direct relationship
between the repositories and resolving.
In short, it became more low-level and only provides some of the building
blocks necessary to implement something like Felix OBR. So, yes, I'd like
to see Felix OBR built on top of these new building blocks, but in the end
it will still require that you tie yourself to a specific implementation to
get the full features of the current Felix OBR.
-> richard
Regards
Neil
On Wednesday, 18 April 2012 at 13:01, Per-Erik Svensson wrote:
Hi,
Is there any timeframe on the release of OBR (the api, not the felix
implementation). I know this might not be the right place to ask but... I
figured that people like Richard, Peter or Neil would be able to answer
and
I know they subscribe. :)
The reason I ask is because I'm thinking of using the Felix
implementation
of the API (org.apache.felix.**bundlerepository) but before commiting
to this
I would like to know if that implementation is supposed to be fully
compliant with the spec and if the spec will ever get released.
Best regards,
Per-Erik Svensson
------------------------------**------------------------------**---------
To unsubscribe, e-mail:
users-unsubscribe@felix.**apache.org<[email protected]>
For additional commands, e-mail: [email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
