I don't think that the number of bundles is an issue: if the user uses
bundle:list (without -t 0), he doesn't see that ;)
For instance, in some projects, like ACE or Aries, we have a lot of
bundles, and it's not a big deal.
Regards
JB
On 03/13/2013 04:26 PM, Guillaume Nodet wrote:
On Wed, Mar 13, 2013 at 4:01 PM, Jean-Baptiste Onofré <j...@nanthrax.net>wrote:
Thanks Guillaume for this remember (or introduction for some of us I think
;)).
I think that on trunk we made some progress in the way that you describe.
For instance, unlike that we have in Karaf 2.x, modules on trunk are
structured like this:
- core provide OSGi services
- commands use the core services
- MBeans use the core services
- an end-user can use core services if he wants
Yes, and from a purely technical side, it's really nice. As Ioannis said,
it can be troublesome for users that Karaf comes with 80 bundles ... It has
the nasty drawback of not looking lightweight anymore ...
Where I'm fully agree is to avoid to go too "deep" in granularity: we
already discussed of Karaf utils: it's a jar embedded in other bundles,
it's not a bundle exposing a service or API used by other bundles.
I'm not fully agree with Christian. OSGi doesn't mean that we have to
expose all as OSGi, for instance, it doesn't make sense for Karaf utils (we
are not in a developer bullshit approach where we turn all in OSGi just for
"fun" or "elegance", we have to keep things simple, maintainable, and
coherent).
My 0.02€
Regards
JB
On 03/13/2013 11:21 AM, Guillaume Nodet wrote:
Starting a new thread for discussing those points.
The idea for OSGi is modularity, but it should be done at the right level.
And modularity is different from code sharing.
In OSGi, the main idea is to have bundles exposing API and services.
That's the way we leverage the most of OSGi.
Unlike projects like CXF or Camel, we develop for OSGi so we should try to
use it in the best way.
Let's consider the 3 different things we can have in OSGi : apis,
implementations and libraries.
We have basic rules:
* service api packages should be exported
* service implementation packages should be kept private
When a service bundle ships the api in the same bundle as the
implementation, the rule is to export and import the api package to let
the
framework use package substitution if needed. For libraries, the rule is
to export the packages and not import those, as subsitution should never
occur.
One first case is service api and implementations. A service bundle can
either provide its own API or the API be provided by a separate package.
With the osgi compendium jar, it's recommended to not use it because it
breaks modularity: this bundle provides lots of different service apis, so
you can't change them one by one. That's why osgi service implementation
usually ship their own API. For apis we control, things are slightly
different, as we don't have those big bundles. For those cases, the best
thing is actually to ship the API and the service implementation in
different bundles. This allows updating the service implementation
without
requiring a refresh of the other bundles.
Let's now discuss the bundle lifecycle pov. If a bundle providing a
service depends for the implementation on libraries packages provided by
external bundles, it breaks the abstraction and modularity to some degree
by exposing internal constraints. Those constraints are better captured
by
service dependencies.
Let's take an example: we have bundle A and B depending on a library
provided by bundle C. If you want to update C to provide a fix needed for
A, this will also impact B, which could cause a regression to B
functionality. There are transient ways around that: i.e. deploy a new
version of bundle C or update C and only refresh A. But those are only
transient as in felix, the wiring isn't retained across restarts. And
this
can't really be controlled well with the tools we have at hand.
On the other hand, if both bundle A and B embeds the C library needed you
can update A and B independently, so a better modularity at the cost of
less code sharing.
So in an ideal OSGi world, service APIs would be shipped in individual
bundles, and service implementation would have no other constraints than
on
other services.
We're not in an ideal world though. For big projects such as Camel, CXF,
ActiveMQ, there's not always a real API and other constraints may come in,
mostly that they are not architected purely for OSGi.
Anyway, for small library bundles that can easily be embedded, I think we
should do it: there's no drawbacks I can see, and it improves modularity.
On Wed, Mar 13, 2013 at 10:05 AM, Christian Schneider <
ch...@die-schneider.net> wrote:
I do not agree with embedding bundles except for some rare cases. They
make it much more difficult to work with those projects. In maven you
always
get the list of dependencies including the embedded ones unless you
exclude them.
I agree though that ops4j contains too many fine grained bundles. Instead
of embedding I propose to check if we could just merge some of these
libs.
Christian
On 13.03.2013 07:41, Guillaume Nodet wrote:
+1
A few comments though
When I started the first time, karaf failed to install the additional
features (ssh, management, etc...)
I then removed my ~/.m2/settings.xml which were pointing to a nexus and
restarted from clean. That worked, but the bundles took a long time to
install. While it works, I think going to the internet when starting
is a
really bad idea.
karaf@root()> instance:connect test
undefined option -p
Try <command> --help' for more information.
On a side note, the completion usability is really lessened by the
subshells. We really need to fix that somehow. Maybe by having
subshells
ending with ':'so that completion would behave better.
Another problem with subshells: when going out of a subshell, the scope
isn't modified. So it's effectively as if you were still inside the
subshell and you loose the default scope behavior, which is especially
troubling with the list command. I.e.:
> feature
> exit
> list
the list command gives you the feature list and not the bundle list.
I think some bundles are too fine grained to be deployed as is. I'm
mostly referring to org.ops4j.base/* and org.ops4j.pax.swissbox/*
bundles
which are pure helper libraries and should imho be embedded whenever
possible. I think we already had those discussions, but ideally,
library
bundles should be avoided.
On Tue, Mar 12, 2013 at 4:26 AM, Jamie G. <jamie.goody...@gmail.com>
wrote:
Hi,
We resolved 964 issues in this release (web page will be published
post RC promotion):
https://svn.apache.org/repos/****asf/karaf/site/trunk/src/**main/**<https://svn.apache.org/repos/**asf/karaf/site/trunk/src/main/**>
webapp/index/community/****download/karaf-3.0.0.RC1-****release.page<
https://svn.**apache.org/repos/asf/karaf/**site/trunk/src/main/webapp/
**index/community/download/**karaf-3.0.0.RC1-release.page<https://svn.apache.org/repos/asf/karaf/site/trunk/src/main/webapp/index/community/download/karaf-3.0.0.RC1-release.page>
NOTE: This is a technology preview release candidate.
Staging repository:
https://repository.apache.org/****content/repositories/**<https://repository.apache.org/**content/repositories/**>
orgapachekaraf-019/<https://**repository.apache.org/content/**
repositories/orgapachekaraf-**019/<https://repository.apache.org/content/repositories/orgapachekaraf-019/>
Release tags:
https://svn.apache.org/repos/****asf/karaf/tags/karaf-3.0.0.****RC1/<https://svn.apache.org/repos/**asf/karaf/tags/karaf-3.0.0.**RC1/>
<https://svn.apache.org/**repos/asf/karaf/tags/karaf-3.**0.0.RC1/<https://svn.apache.org/repos/asf/karaf/tags/karaf-3.0.0.RC1/>
3.0.x Dependencies table:
https://svn.apache.org/repos/****asf/karaf/site/trunk/src/**main/**<https://svn.apache.org/repos/**asf/karaf/site/trunk/src/main/**>
webapp/index/documentation/****karaf-dependencies/karaf-deps-**
**3.0.x.page<https://svn.**apache.org/repos/asf/karaf/**
site/trunk/src/main/webapp/**index/documentation/karaf-**
dependencies/karaf-deps-3.0.x.**page<https://svn.apache.org/repos/asf/karaf/site/trunk/src/main/webapp/index/documentation/karaf-dependencies/karaf-deps-3.0.x.page>
Please vote to approve this release:
[ ] +1 Approve the release
[ ] -1 Veto the release (please provide specific comments)
This vote will be open for 72 hours.
--
Christian Schneider
http://www.liquid-reality.de
Open Source Architect
http://www.talend.com
--
Jean-Baptiste Onofré
jbono...@apache.org
http://blog.nanthrax.net
Talend - http://www.talend.com
--
Jean-Baptiste Onofré
jbono...@apache.org
http://blog.nanthrax.net
Talend - http://www.talend.com
