Can someone shed a ray of light on when exactly a bundle's classloader is
(supposedly) made eligible for GC? I'm working with native code and while
it is not required to re- or unload my bundle, it would be nice to have a
better grasp of potential lifecycle issues.
Right now felix (on Windows)
Todor, Richard -
thank you for your responses. I understand how various GC schemes work - I
guess I should have asked a more concise question like does the framework
explicitly keep a reference to the classloader, and if so, when is that
reference released. :) However in the meantime I found out
Nima Kaviani wrote:
I have developed a cross platform bundle that installs/uninstalls either
comm-win32 or comm-linux jars to enable communication with the serial port.
Everything works fine on Linux, however on windows, Felix seems to cache
win32com.dll along with other files and folders at
Richard S. Hall wrote:
On 10/22/09 14:01, Sahoo wrote:
I would not call that a simple class. It will certainly be better to
have a property to customize the message.
+1
We have always expected people would create their own launchers for
their own purposes. Now that R4.2 has standardized
Richard S. Hall wrote:
We have always expected people would create their own launchers for
their own purposes. Now that R4.2 has standardized on the framework
launching API, it makes more sense than ever to roll your own launcher
(or build on ours).
Here we go:
Donald Whytock wrote:
Then it occurred to me that there are multiple types of services I'd
like to do this with, which seemed to call for an abstract class,
similar to ServiceBinder.GenericActivator, that could be subclassed
with the ServiceImpl and the ServiceFactoryImpl. Thus serving as a
Justin Edelson wrote:
the Require-Bundle header, OBR should handle it. That said, you should
also raise an issue with whomever created these bundles and ask them to
fix the manifests.
Good luck with that. The Orbit repo at eclipse is its own little locked-in
universe and makes heavy use of
On Wed, Oct 13, 2010 at 3:49 AM, LongkerDandy longkerda...@gmail.com wrote:
I have the same situation.
Always using mvn install.
Seems to work just fine for me with maven3. :-)
-h
-
To unsubscribe, e-mail:
On 01.12.2010 09:04, Nima Kaviani wrote:
I have a question now. Is there a way to figure out these wirings
programmatically.
This will be standardized in OSGi 4.3. You can take a peek at the EA
spec at:
http://www.osgi.org/Download/File?url=/download/osgi-4.3-early-draft2.pdf
-h
On 12.01.2011 15:13, Clay McCoy wrote:
The whole reason that I am trying to use Embed dependencies and transitive =
true is so that I don't have to deal with dependencies. They are all
supposed to be embedded into a self contained jar.
But now I am having to deal with sun.misc. This seems
Hi,
I'm looking at the HTTP Service and was wondering if anyone can tell if or
how I can let the implementation pick a free port for a service, instead
of having to choose one myself? The reachable address would be
communicated to the outside world by different means, but I don't want to
specify
On 20.01.2011 23:50, Carl Hall wrote:
This sounds pretty interesting. I'll take a stab at this and be sure to
update the JIRA [1].
[1] https://issues.apache.org/jira/browse/FELIX-2786
Great! I would have gotten to it eventually but have other things on my
list for now. Will gladly provide
On 31.01.2011 15:25, tim.hopk...@accenture.com wrote:
Hi
I'm having a problem trying to get the xml-apis bundle to build. I'm
new to Felix and to Maven, so this might all be my fault. When I
The good news: not your fault. The bad news: you're doomed.
execute Maven against the xml-apis
On 18.04.2011 18:26, Yuri de Wit wrote:
Is it possible to only generate the OSGi Manifest with maven-bundle-plugin?
Adding OSGi metadata to existing projects without changing the
packaging type
at:
http://felix.apache.org/site/apache-felix-maven-bundle-plugin-bnd.html
-h
Sorry for writing so much but you asked for it.. :-)
On 28.04.2011 20:08, duncan wrote:
We are accessing native libraries from Java. This works outside of
OSGi and in OSGi on Windows but not on Linux probably because the dll
and so linking behave differently.
Yes and no. They work
On 29.04.2011 16:31, Daniel Faber wrote:
wow, awesome post with great detailed explanations which I am very
interested in because we had a similar problem. I have just one
Thanks!
I've read chapter 3.9 Loading Native Code Libraries in the OSGi spec
again and it seems the framework is not
On 29.04.2011 17:30, Richard S. Hall wrote:
In that case the only way out is to first try $ORIGIN and, failing that,
a fixed location like /tmp or any other directory to which your bundle
can unpack the libraries and try again. It's much more difficult to get
right since you really want to
On 10.06.2011 23:39, mvangeert...@comcast.net wrote:
Do you feel that embedding all of a file's dependencies into a bundle
is the right approach? I think it contravenes some of the core
principles of OSGi to embed dependencies except in some rare
corner-cases.
I'm not Justin, but..
Yes and
On 21.06.2011 12:22, Elvy wrote:
I'm on Windows Vista (which is 64 bits) and after switching from JVM 32bits
AFAIK there are also 32bit variants of Vista and Win7..
to JVM 64bits I had to use the following on my maven-bundle-plugin for my
app to properly load on Felix 3.0.8
On 19.08.2011 15:28, john.dun...@exceter.com wrote:
Yes, that's exactly what I need to do. Do you have a recommended approach?
Don't know what your service does, but creating a bundle on-the-fly is
quite easy (I was surprised how easy, actually). I have some code but
can't post the full version,
On 30.08.2011 12:51, Ivanhoe Abrahams wrote:
I noticed that it really depends on the start order of the bundles whether
my simple class gets woven or not.
This is because you haven't really declared any kind of logical dependency
between the bundles; if your client bundle can work without
On 30.08.2011 15:36, Richard S. Hall wrote:
While I agree with you in spirit, generic caps/reqs won't really help
when it comes to preventing one bundle from starting before another,
since they address resolve-time dependencies, not activation-time
dependencies. The weaver must be active
..and now for something complicated :)
I'm embedding a third-party jar and need to re-export all its packages. In
theory I should be able to specify inline=true since I get the classes of
the exported packages anyway, so there is no need to still keep the
complete jar embedded. Turns out this
On 27.10.2011 16:59, Stuart McCulloch wrote:
This is because embedding uses bnd's Include-Resource instruction,
which is independent of the exported/private package processing.
Includes are processed after the packages and currently always
overwrite any existing resource.
Thanks for the
On 03.01.2012 09:42, david koch wrote:
Hello,
I try to use the Bundle.adapt method with Java 1.7.0_02 in the following
way:
Framework framework;
FrameworkStartLevel startLevelObject =
framework.adapt(FrameworkStartLevel.class);
This is ok for the Eclipse compiler (1.7) but when I
On 18.01.2012 03:35, DBinSD wrote:
Does the felix implementation of the EventAdmin service support the delivery
of events, asynchronously and in parallel to multiple listeners at once? It
This was addressed as part of RFC 157 (Updates to EventAdmin) sometime
in 2010 and accepted, though I
On 03.02.2012 15:49, Richard S. Hall wrote:
I also know that the ECF guys ran into this same issue (with Equinox)
and that they solved it by introducing a separate proxy bundle under
which they'd register their service factories.
This is what I did in my (more precisely Paremus')
On 30.03.2012 15:03, Jochen Wiedmann wrote:
On Fri, Mar 30, 2012 at 1:02 PM, Felix Meschberger fmesc...@adobe.com
wrote:
If sou, you would just open an InputStream to those JAR files and had
it over to BundleContext.installBundle(String, inputStream) where
string would be an URL (or an URL
On 15.05.2012 21:34, Neil Bartlett wrote:
In other words, Hibernate is stupid, and breaks modularity! It's well
known to OSGi programmers as a difficult library to use.
What Neil said. But not all's unwell that ends unwell.
As usual there is more to it and as you can see on
Hi,
I'm trying to fix disruptor's b0rken bundle manifest and it hard-requires
sun.misc, hooray. This got me thinking: considering that sun.misc often
needs to be added to system packages/bootdelegation anyway, is it a good
idea to still explicitly import it? I *think* it is good style in order to
On 08/13/12 09:46, Felix Meschberger wrote:
Am 10.08.2012 um 20:19 schrieb Martin Lichtin:
Hi
Very rarely I'm getting the following IOException during startup of a
Pax-Exam(-Karaf) test:
[FelixStartLevel] ERROR org.apache.felix.fileinstall - Failed to install
artifact:
On 02/12/13 13:21, Imóveis Nacionais wrote:
How can I access server bundle interface type during compile time?
By adding it into a separate -api jar for (both client and server)
compile time. For runtime you can either turn that jar into a bundle as
well (so that compile-time structure
On 23.02.2013 09:12, Dhiego Abrantes de Oliveira Martins wrote:
But, I want to find a bundle that contains an specific implementation of
HiService, in this case is HiServiceImpl.
This is what service properties are for (it's all in the spec). You
don't really want HiServiceImpl, you want the
On 06.03.2013 14:30, Francis ANDRE wrote:
I would like to mavenize a huge OSGi application. What is the best way
to automaticaly produce the pom.xml from an existing MANIFEST.MF OSGi
bundle?
Unless you have a database/reverse-lookup mechanism that maps packages
to jars, the answer is you
On 02.07.2013 10:36, Christopher BROWN wrote:
What is the best strategy for integrating JDBC drivers, such as MySQL,
jTDS, and others that can start background threads or timers, so that these
can be correctly stopped when the bundle is unloaded?
Implementing the JDBC Service bits? See
On Thu, Jul 25, 2013 at 3:34 PM, Neil Bartlett njbartl...@gmail.com wrote:
It's possible that there is a dependency from LeapJava.dll to Leap.dll
within the native code. This is not something that OSGi can directly
Not just possible but in the case of JNI stubs rather mandatory. :)
36 matches
Mail list logo
