Equinox will not export the annotation package (via Export-Package)
because it will lead to a bad practice of using Import-Package for these
non-runtime packages. Whatever the solution ends up being to get the
annotations on the classpath of the project, PDE must do so without
requiring the
You probably should open a bug and give the details on how to reproduce. A
stacktrace may also help and the error log if it exists, At this point I
don't understand what is going wrong and I am not sure what to suggest.
With the Luna release (framework org.eclipse.osgi version 3.10) there was
> From: Stephan Herrmann
> Hi Tom,
>
> Thanks for your comprehensive answer.
> Yes, we can easily agree that SynchronousLogListener is dangerous :)
>
> More questions inline ...
>
>
> Sounds promising, but ...
>
> > But be aware that nobody is going to see your logs until there is
> a
Hi
The Eclipse Log API (org.eclipse.core.runtime.ILog) has the unfortunate
contract that log listeners registered with it will be called
synchronously. This can cause the issues you are seeing with class
loading if we get into a circularity. In Equinox we actually have one and
only one log
ext = null;
}
}
Tom
From: Thomas Watson/Austin/IBM@IBMUS
To: Equinox development mailing list <equinox-dev@eclipse.org>
Date: 05/16/2016 09:15 AM
Subject:Re: [equinox-dev] Safe logging from a WeavingHook?
Sent by:equinox-dev-boun...@eclipse.org
Hi
The
You may be interested in reading an article that I wrote about
experimenting with JPMS layers and representing an OSGi framework as a
JMPS Layer [1].
Tom
[1] http://blog.osgi.org/2016/08/osgi-with-java-modules-all-way-down.html
___
equinox-dev
It depends on the launcher you are using. Here I am going to assume you
are using the Equinox launcher as-is. The thing about
FrameworkEvent.STARTED is that it is only fired when the framework has
reached its "beginning framework start-level" during the call to
Framework.start(). By default
Some bundle in your environment looks like it has no Bundle-SymbolicName.
Can you open a bug against p2. It needs a null check at
ConfigApplier.refreshPackages(ConfigApplier.java:401)
But I am curious how you got p2 to install a bundle with no
Bundle-SymbolicName. I thought that was
: Thomas Watson/Austin/IBM@IBMUS
To: Equinox development mailing list <equinox-dev@eclipse.org>
Date: 06/30/2016 07:51 AM
Subject:Re: [equinox-dev] Fwd: Starting a runtime Eclipse fails
with Exception in
org.eclipse.equinox.internal.simpleconfigurator.Activator.start() of
I may be wrong, but I think the limitations are on the Eclipse console
view implementation. You may want to try posting a message to
https://dev.eclipse.org/mailman/listinfo/jdt-debug-dev to get details on
enabling that for the eclipse console.
Tom
From: Christian Schneider
A 'reference' install will not copy the content of the bundle file into the framework's internal storage. Instead it will load the bundle directly out of the file URL the 'reference' URL is pointing to. Note that 'reference' URLs only support embedded 'file' URLs (not http etc.) This has
2017 6:38 AM, Thomas Watson wrote:
EventAdmin is also the RI for OSGi R6. There are no updates to EventAdmin for R7.Thanks. One follow up on EventAdmin. I've read that the Equinox EventAdmin uses a single thread for dispatch...i.e. for all topics...and it seems like that's correct given a
tatype
Sent by:equinox-dev-boun...@eclipse.org
On 1/19/2017 12:36 PM, Thomas Watson wrote:
Equinox metatype implementation continues to be the reference
implementation for OSGi and implements the very latest released
specification.
I did update the CM implementation to be CM spec versi
Equinox metatype implementation continues to be the reference
implementation for OSGi and implements the very latest released
specification.
I did update the CM implementation to be CM spec version 1.5 a long time
ago (3 years?). OSGi R6 did not update the CM spec, if I recall
correctly. The
There is a bad endless loop bug in the last I-Build that the Equinox team
is requesting a respin for. Markus has already provided a fix for bug
https://bugs.eclipse.org/bugs/show_bug.cgi?id=500225 and we are ready for
a respin.
To workaround the issue in the Install dialog ensure that the
For Oxygen I plan to replace the Equinox DS implementation with the Felix
DS (SCR) implementation. See
https://bugs.eclipse.org/bugs/show_bug.cgi?id=501950
Tom
From: Todor Boev
To: equinox-dev@eclipse.org
Date: 09/28/2016 07:31 AM
Subject:[equinox-dev]
Equinox is used in many environments where there is very long-term support
for older Java levels. So far I have needed to keep the Equinox Framework
at a Java 6 level to be able to use the latest versions in these
environments.
Unfortunately Java 6 is getting harder and harder to support
in properly diagnosing these issues?
Regards,
Michal
From: equinox-dev-boun...@eclipse.org <equinox-dev-boun...@eclipse.org> on
behalf of Thomas Watson <tjwat...@us.ibm.com>
Sent: Monday, October 3, 2016 2:58:25 PM
To: Equinox development mailing list
Subject: Re: [equinox-dev]
It is hard to tell if the exceptions are issues to be concerned about
without having one as an example. It is possible that enabling debug is
just printing out some benign exception that is otherwise ignored by the
system and not logged.
Yes Equinox should easily handle 1000s of bundles being
Disabling uses is usually not a good idea. I would be interested to know
why you want to do this.
To answer your question, I would not disable this in the felix
ResolverImpl directly. Instead I would disable it at a higher level in
the equinox container by hiding the uses directives on the
I will be interested to see if you can successfully map the OSGi uses
concept into the SAT solver p2 uses. I briefly looked at that a long time
ago when we were refactoring the Equinox framework (Luna) and were
replacing the old Equinox resolver. It was far from obvious how you would
achieve
I'll also state that in Equinox (and AFAIK in Felix) the PackageAdmin
implementation is not going anywhere for the foreseeable future.
Tom
From: Pascal Rapicault
To: Equinox development mailing list
Date: 12/13/2016 08:46 PM
Subject:
I think Pascal is asking how to find an installed bundle given a specific
BSN and version (range?). I suggest you take a look at the
PackageAdminImpl for the getBundles method:
Yes, that is the plan. I had hoped to get it in for M4. But have not had
time to get it in. I will look to put it in for M5. If it slips past
that though it will get more difficult to get it in.
Tom
From: Scott Lewis
To: Equinox development mailing list
And yes, it will support the latest Declarative Services version 1.3 from
the R6 specification
Tom
From: Thomas Watson/Austin/IBM@IBMUS
To: Equinox development mailing list <equinox-dev@eclipse.org>
Date: 11/29/2016 07:55 AM
Subject:Re: [equinox-dev] Felix SC
I'm not sure I understand the scenario.
What I do know is that in the latest Equinox (since the Luna release) the
framework will eagerly flush updated or uninstalled class loaders if they
are not 'reachable' by any other in-use bundle in the framework. It
sounds to me like this may be what is
This is all getting a bit fuzzy for me because PDE is still using the old
Equinox resolver API which the framework no longer actually uses at
runtime. But if I recall correctly Require-Capability reqs are modelled
with the org.eclipse.osgi.service.resolver.GenericSpecification interface
which
For awareness,
I am looking at it, but will be a bit slow while at EclipseConverge and Devoxx. Looking at the problem it seems that the resolver should find a solution pretty easily but the presence of one of the versions of org.apache.httpcomponents.httpcore_4.4.4.v20161115-1643 (there are
Over the years Equinox has become a stable project. The last major code effort was in the Luna release when the Equinox framework was refactored to use the OSGi generic capability and requirements model along with the OSGi Resolver service. Over the past few years the list of active committers
Singleton should have nothing to do with activation. Do you have a scenario to reproduce?
Tom
- Original message -From: Lars Vogel Sent by: equinox-dev-boun...@eclipse.orgTo: Equinox development mailing list Cc:Subject: [equinox-dev]
I have merged the branch we had for OSGi R7 into master for rt.equinox.framework.
I expect there may be a few compilation errors in the next build because of some places where we implement interfaces from the framework which are not intended to be implemented by client code (probably mostly
Over the years Equinox has become a stable project. The last major code
effort was in the Luna release when the Equinox framework was refactored
to use the OSGi generic capability and requirements model along with the
OSGi Resolver service. Over the past few years the list of active
ecomean OSGi bundle and at it to the target platform).But from your answer, I understand that OSGi / Equinox does currentlynot plan to simplify the consumption of the Java 9 modules. So we haveto stay with the old conversion process.Thanks again for the answer, LarsOn Wed, Oct 4, 2017 at 8:20 PM, Thomas W
hat matter).best,StephanOn 05.10.2017 15:12, Thomas Watson wrote:> Not sure if you are aware of the work I have done for JPMS inter-op [1]> During my prototyping of inter-op I also did try implementing what you suggest, by adapting Java 9 modules to OSGi bundles on the> fly (See [2]). I kn
Hi Scott,
You are correct. That package got added to Oxygen when we started using the Felix SCR implementation for our OSGI R6 Declarative Services.
Tom
- Original message -From: Scott Lewis Sent by: equinox-dev-boun...@eclipse.orgTo: Equinox development
:Subject: Re: [equinox-dev] logging jobsDate: Wed, Mar 28, 2018 9:51 AM
On Wed, Mar 28, 2018 at 10:37 AM, Thomas Watson <tjwat...@us.ibm.com> wrote:
What do you mean by runtime jobs? When I hear that in the Eclipse context I think something from org.eclipse.core.jobs.
Maybe?
But I suspec
What do you mean by runtime jobs? When I hear that in the Eclipse context I think something from org.eclipse.core.jobs. But I suspect that is not what you are asking for. Are you asking for some tracing for all background operations the framework performs? I assume you are trying to diagnose
Dictionary my be documented as obsolete, but it is not deprecated. I would not want to deprecate a spec'ed API method that takes non-deprecated types. Anyway, the discussion is happening now in the expert group.
Tom
- Original message -From: Lars Vogel Sent by:
Neil and Gunner are correct. I also question the motivation to running activators in parallel. In implies some kind of design flaw because you must have some long running BundleActivators if you want to run them in parallel. You are better off investigating why you must have long running
I've started the 4.11 Equinox planning page at https://wiki.eclipse.org/Equinox/Plan/4.11
Please add any major work items you plan to work on for 4.11.
Tom
- Original message -From: "Daniel Megert" Sent by: eclipse-dev-boun...@eclipse.orgTo: "General development mailing list of the
I added a comment to the bug report, we can continue the discussion there.
Tom
- Original message -From: Guillaume Dufour Sent by: equinox-dev-boun...@eclipse.orgTo: Equinox development mailing list Cc: tycho-...@eclipse.orgSubject: Re: [equinox-dev] Tycho Surefire Test fails to load
The equinox jobs have been migrated over to the new infrastructure, but we are experiencing some issues with the migration. See https://bugs.eclipse.org/bugs/show_bug.cgi?id=545352
Tom
- Original message -From: Mykola Nikishov Sent by: equinox-dev-boun...@eclipse.orgTo:
Thanks Mykola, I've done the same for the Equinkox framework and bundles jobs now.
Tom
- Original message -From: Mykola Nikishov Sent by: equinox-dev-boun...@eclipse.orgTo: equinox-dev@eclipse.orgCc:Subject: [EXTERNAL] [equinox-dev] Log Maven version in p2-master and p2-gerrit
That version is over 6 years old and was released before Java 8 was even available so it may have issues with running on the oldest supported Java available and I doubt it will work well on anything Java 9+. While I suspect it should just work for the most part on Java 8 there will be a few
All the source can be found in the eclipse git repositories info found at https://git.eclipse.org/c/equinox/rt.equinox.framework.git/ and https://git.eclipse.org/c/equinox/rt.equinox.bundles.git/
But for most we do publish the source JARs into maven central:
Thanks again Johah,
The blame history has been properly restored to the master branch and to the osgiR8 branch using this approach. My mistake is now properly swept under the rug.
Tom
- Original message -From: "Thomas Watson" Sent by: equinox-dev-boun...@eclipse.orgTo: e
from:
$ git blame bundles/org.eclipse.osgi.tests/META-INF/MANIFEST.MF--snip--713ce018b1 (Thomas Watson 2020-03-10 16:03:39 -0500 5) Bundle-Version: 3.15.300.qualifier
to$ git blame bundles/org.eclipse.osgi.tests/META-INF/MANIFEST.MF--snip--27288b2b5b (Alexander Kurtakov 2020-03-08 14:28
I incorrectly merged into master a long running branch I have for OSGi R8 development in the rt.equinox.framework. I have a bug opened against the foundation asking to get a force push to master to correct things: https://bugs.eclipse.org/bugs/show_bug.cgi?id=560997
As much as I dislike a
to do the force push to master for me.
Sorry for the bone-headed move on my part.
Tom
- Original message -From: Thomas Watson/Austin/IBMTo: equinox-dev@eclipse.org, platform-releng-...@eclipse.orgCc:Subject: Incorrect merge of osgiR8 branch into master for rt.equinox.frameworkDate
I would try the osgi-dev mailing list: https://mail.osgi.org/mailman/listinfo/osgi-dev
Tom
- Original message -From: Lars Vogel Sent by: equinox-dev-boun...@eclipse.orgTo: Equinox development mailing list Cc:Subject: [EXTERNAL] [equinox-dev] JAXB with Java 11 in OSGiDate: Tue, Jul 28,
I think there is simply a miscommunication here and nothing to take offence at. Regardless we have given Ranjana more than enough information.
Tom
- Original message -From: "Sravan K Lakkimsetti" Sent by: equinox-dev-boun...@eclipse.orgTo: "Equinox development mailing list"
See https://bugs.eclipse.org/bugs/show_bug.cgi?id=525368
Under the covers p2 uses SAT4J to implement a resolver for selecting what to install. I remember discussing with the original p2 developers the complications of the uses directive such that they could look into encoding the constraint
I would be in favor of deprecating the API package org.eclipse.osgi.service.resolver which is contained in the Equinox Framework (org.eclipse.osgi). The framework fragment called org.eclipse.osgi.compatibility.state holds the actual implementation, which is internal. But anything needing to get
I've certainly made fixes to this code long after Equinox has stopped using it and I don't mind maintaining fixes there when required. I do care enough to not break existing projects that are using this API. After all, I do use PDE nearly every day and we are very dependant on tycho to build.
like a rat hole to me. :-(
On 08.01.2021 16:02, Thomas Watson wrote:
You are correct, this is a rather a tricky situation. Especially because of things like org.eclipse.pde.core.plugin.IPluginModelBase.setBundleDescription(BundleDescription)
I'm not entirely sure why that is API there,
stion here to deprecate these things too?!
On 08.01.2021 15:40, Thomas Watson wrote:
I would be in favor of deprecating the API package org.eclipse.osgi.service.resolver which is contained in the Equinox Framework (org.eclipse.osgi). The framework fragment called org.eclipse.osgi.compatibility.s
To be clear, the Eclipse IDE doesn't support x86 32-bit. Nothing in the Equinox Framework prevents you from running on x86 32-bit.
Can you run the gogo console in your environment and run the following command?inspect capability osgi.native 0
I expect output like this (note this is from my
The dev class path entries are prepended to the value of the Bundle-ClassPath header value. Normal delegation rules apply for the dev class path as if they are part of the bundle's Bundle-ClassPath. That roughly is the following:1) Import-Package wires, if package is imported the search
80420-1519.jarshould this work or do I need to explode the jar files into a folder?Am 12.04.21 um 16:17 schrieb Thomas Watson:> The dev class path entries are prepended to the value of the> Bundle-ClassPath header value. Normal delegation rules apply for the dev> class path as if they ar
That is a bit of a dramatic statement that can open a can of worms. The extension registry and the Eclipse platform model with its many extension points does what it does very well. If you are developing bundles that plugin to the Eclipse platform do not "feel bad" about using the right
If all you need is a path to some resource in a bundle then there is no need for a custom header or a custom service interface. I recommend you define your own capability namespace and define a set of attributes that describe the capability for such a namespace.
For example, define a new
I assume the migration will include binaries also
(https://git.eclipse.org/c/equinox/rt.equinox.binaries.git/) or is that not
necessary?
Tom
From: equinox-dev on behalf of Aleksandar
Kurtakov
Sent: Wednesday, March 9, 2022 10:06 AM
To: Equinox development
The package org.osgi.service.component.annotations is not meant to be imported
by bundles. There is no need for a bundle to be wired to these packages at
runtime. The JARs the OSGi specification (working group) provides in maven
central are only meant to be used for compile time dependencies.
Thanks, I saw that go by.
Equinox was one of the early adopters of parallel class loaders during its
design and eventual inclusion in Java 7. Long ago we stopped needing any such
flags to prevent deadlock with our non-hierarchical class loaders. This should
have no impact for Eclipse
eclipse.incubator.equinox Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Stoyan Boshev. Stoyan Boshev is a
new full Committer on the equinox component of the eclipse.incubator
project.
Welcome!
rt.equinox.bundles Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for John Arthorne. John Arthorne is a
new full Committer on the Bundles component of the rt.equinox project.
Welcome!
+1
+1
Voting summary: http://portal.eclipse.org/
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org/mailman/listinfo/equinox-dev
rt.equinox.incubator Committers,
This automatically generated message signals that Thomas Watson has
nominated Ruibin Zhou as a Committer on the rt.equinox.incubator project.
The reason given is as follows:
As part of the move of the SODA Device Kit move to rt.equinox.incubator
(http
rt.equinox.incubator Committers,
This automatically generated message signals that Thomas Watson has
nominated Aldo Eisma as a Committer on the rt.equinox.incubator project.
The reason given is as follows:
As part of the move of the SODA Device Kit move to rt.equinox.incubator
(http
rt.equinox.incubator Committers,
This automatically generated message signals that Thomas Watson has
nominated David Lavin as a Committer on the rt.equinox.incubator project.
The reason given is as follows:
As part of the move of the SODA Device Kit move to rt.equinox.incubator
(http
rt.equinox.incubator Committers,
This automatically generated message signals that Thomas Watson has
nominated Simon Archer as a Committer on the rt.equinox.incubator project.
The reason given is as follows:
As part of the move of the SODA Service Activator Toolkit (SAT) to
rt.equinox.incubator
? Dave Stevenson
+1 Thomas Watson
? Matthew Webster
If you have any questions, please do not hesitate to contact your project
lead, PMC member, or the EMO e...@eclipse.org
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org
? Dave Stevenson
+1 Thomas Watson
? Matthew Webster
If you have any questions, please do not hesitate to contact your project
lead, PMC member, or the EMO e...@eclipse.org
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org
rt.equinox.incubator Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Aldo Eisma. Aldo Eisma is a new
full Committer on the rt.equinox.incubator project.
Welcome!
___
rt.equinox.incubator Committers,
This automatically generated message signals that Thomas Watson has
nominated Lazar Kirchev as a Committer on the rt.equinox.incubator project.
The reason given is as follows:
I would like to nominate Lazar Kirchev to the equinox incubator. Lazar has
shown
Moir
+1 Andrew Niefer
? Matthew Piggott
? Steffen Pingel
? Pascal Rapicault
? Heiko Seeberger
? Dave Stevenson
+1 Thomas Watson
? Matthew Webster
? Ruibin Zhou
If you have any questions, please do not hesitate to contact your project
lead, PMC member, or the EMO e
rt.equinox.incubator Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Lazar Kirchev. Lazar Kirchev is a
new full Committer on the rt.equinox.incubator project.
Welcome!
___
Thomas Watson voted:
+1
+1
Voting summary: http://portal.eclipse.org/
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org/mailman/listinfo/equinox-dev
rt.equinox.bundles Committers,
This automatically generated message signals that Thomas Watson has
nominated Glyn Normington as a Committer on the rt.equinox.bundles project.
The reason given is as follows:
As discussed in several Equinox meetings, I have been collaborating with
Glyn Normington
rt.equinox.bundles Committers,
This automatically generated message signals that Thomas Watson has
nominated Borislav Kapukaranov as a Committer on the rt.equinox.bundles
project. The reason given is as follows:
As discussed in several Equinox meetings, I have been collaborating with
the Virgo
Rapicault
? Heiko Seeberger
+1 Thomas Watson
If you have any questions, please do not hesitate to contact your project
lead, PMC member, or the EMO e...@eclipse.org
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org/mailman
rt.equinox.bundles Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Glyn Normington. Glyn Normington
is a new full Committer on the rt.equinox.bundles project.
Welcome!
___
rt.equinox.bundles Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Borislav Kapukaranov. Borislav
Kapukaranov is a new full Committer on the rt.equinox.bundles project.
Welcome!
rt.equinox.bundles Committers,
This automatically generated message signals that Thomas Watson has
nominated Lazar Kirchev as a Committer on the rt.equinox.bundles project.
The reason given is as follows:
Lazar has been working on a solution to extract the equinox console out of
the framework
+1 Glyn Normington
? Pascal Rapicault
? John Ross
? Heiko Seeberger
+1 Thomas Watson
If you have any questions, please do not hesitate to contact your project
lead, PMC member, or the EMO e...@eclipse.org
___
equinox-dev mailing list
rt.equinox.bundles Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Lazar Kirchev. Lazar Kirchev is a
new full Committer on the rt.equinox.bundles project.
Welcome!
___
Thomas Watson voted:
+1
+1
Voting summary: http://portal.eclipse.org/
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org/mailman/listinfo/equinox-dev
Thomas Watson voted:
+1
+1
Voting summary: http://portal.eclipse.org/
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org/mailman/listinfo/equinox-dev
Thomas Watson voted:
+1
+1
Voting summary: http://portal.eclipse.org/
___
equinox-dev mailing list
equinox-dev@eclipse.org
https://dev.eclipse.org/mailman/listinfo/equinox-dev
rt.equinox.bundles Committers,
This automatically generated message signals that Thomas Watson has
nominated Markus Keller as a Committer on the rt.equinox.bundles project.
The reason given is as follows:
I am nominating Markus Keller as a committer to rt.equinox.bundles to take
over
? Martin Lippert
? Kim Moir
+1 Glyn Normington
+1 Pascal Rapicault
+1 John Ross
? Heiko Seeberger
+1 Thomas Watson
If you have any questions, please do not hesitate to contact your project
lead, PMC member, or the EMO e...@eclipse.org
rt.equinox.bundles Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Markus Keller. Markus Keller is a
new full Committer on the rt.equinox.bundles project.
Welcome!
___
rt.equinox.framework Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for John Ross. John Ross is a new full
Committer on the rt.equinox.framework project.
Welcome!
___
rt.equinox.bundles Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Raymond Auge. Raymond Auge is a
new full Committer on the rt.equinox.bundles project.
Welcome!
___
rt.equinox.bundles Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Stefan Xenos. Stefan Xenos is a
new full Committer on the rt.equinox.bundles project.
Welcome!
___
rt.equinox Committers,
This automatically generated message marks the completion of all the legal
paperwork and webmaster provisioning for Alexander Kurtakov. Alexander
Kurtakov is a new full Committer on the rt.equinox project.
Welcome!
___
equinox-dev
701 - 796 of 796 matches
Mail list logo