You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@stanbol.apache.org by Fabian Christ <ch...@googlemail.com> on 2013/01/02 16:02:37 UTC
Re: OWL API in Maven Central
Hi,
is there a chance that we can solve this soon? Otherwise I am planning to
create a branch for the commons release and exclude the OWL stuff.
Best,
- Fabian
2012/12/19 Fabian Christ <ch...@googlemail.com>
> Hi,
>
> I tried to update the commons/owl with these deps
>
> <dependency>
> <groupId>net.sourceforge.owlapi</groupId>
> <artifactId>owlapi-api</artifactId>
> <version>3.4.2</version>
> </dependency>
> <dependency>
> <groupId>net.sourceforge.owlapi</groupId>
> <artifactId>owlapi-apibinding</artifactId>
> <version>3.4.2</version>
> </dependency>
>
> and this embed instructions
>
> <Embed-Dependency>owlapi-api,owlapi-apibinding</Embed-Dependency>
>
> while the commons compiles fine with test I had no luck with the
> Ontologymanager.
>
> These tests are now failing
>
>
> org.apache.stanbol.ontologymanager.multiplexer.clerezza.ontology.TestOntologyReconciliation.versionedOnlyFromURL()
>
> org.apache.stanbol.ontologymanager.multiplexer.clerezza.ontology.TestOntologyReconciliation.versionedOnlyFromStream()
>
> and I get a compile error in the registry manager because the method
> 'add(org.semanticweb.owlapi.model.IRI)' can not be found now.
>
> I gave up at this point but maybe my try helps to get it right.
>
> Best,
> - Fabian
>
> 2012/12/14 Alessandro Adamou <ad...@cs.unibo.it>
>
>> On 12/13/12 9:38 PM, Fabian Christ wrote:
>>
>>> okay no problem. Is there something I could do to remove the embedded
>>> repo
>>> that is inline what you have in mind?
>>>
>>
>> I am not lucid enough to think of one, but I do remember having tried to
>> replace the embedded artifact with at least a couple of those in Maven
>> (owlapi-api and owlapi-apibinding IIRC) in commons/owl. I think only the
>> reasoner's unit tests were failing.
>>
>> As for the reasoner I was thinking about HermiT again - to have it as an
>> externally-contributed LGPL bundle not in the release. But a new HermiT has
>> to come out first.
>>
>> Just let me get out of this thesis thing and I'll have a look.
>>
>> Alessandro
>>
>>
>> 2012/12/13 Alessandro Adamou <ad...@cs.unibo.it>
>>>
>>> There isn't a compatible OWL 2 reasoner that can be distributed yet. I
>>>> guess we can release commons-owl and make do without one for the time
>>>> being. I'm confident a reasoner will show up eventually.
>>>>
>>>> However I am absolutely unable to do this right now. The deadline for my
>>>> Ph.D. thesis is in the next 2 days. Real sorry.
>>>>
>>>> Alessandro
>>>>
>>>>
>>>>
>>>> On 12/13/12 4:29 PM, Fabian Christ wrote:
>>>>
>>>> Hi,
>>>>>
>>>>> meanwhile OWL API 3.4.2 is available at Maven central.
>>>>>
>>>>> Can we solve this old issue now? [1]
>>>>>
>>>>> Would be really great since this blocks a release of the commons module
>>>>> that I would like to make.
>>>>>
>>>>> [1] https://issues.apache.org/****jira/browse/STANBOL-626<https://issues.apache.org/**jira/browse/STANBOL-626>
>>>>> <https:**//issues.apache.org/jira/**browse/STANBOL-626<https://issues.apache.org/jira/browse/STANBOL-626>
>>>>> >
>>>>>
>>>>>
>>>>>
>>>>> 2012/9/14 Fabian Christ <christ.fabian@googlemail.com <mailto:
>>>>> christ.fabian@**googlemail.com <ch...@googlemail.com>**>>
>>>>>
>>>>>
>>>>>
>>>>> Hi,
>>>>>
>>>>> thanks for trying this. I have no problem with waiting for a
>>>>> compatible version. It just keeps to be a little annoying
>>>>> situation
>>>>> but if there is hope that there will be a compatible version
>>>>> soon, we
>>>>> can wait.
>>>>>
>>>>> It is more a question of when do we and our users would like to
>>>>> see a
>>>>> clean release of this stuff.
>>>>>
>>>>> Best,
>>>>> - Fabian
>>>>>
>>>>> 2012/9/13 Alessandro Adamou <adamou@cs.unibo.it
>>>>> <ma...@cs.unibo.it>>:
>>>>>
>>>>> > Hi,
>>>>> >
>>>>> > the OWL API finally made it to Maven Central. check
>>>>> > net.sourceforge.owlapi:owlapi-*****
>>>>>
>>>>> >
>>>>> > However it's versions 3.3 onwards, and the latest HermiT is
>>>>> compatible with
>>>>> > OWL API 3.2.2. I tried it with the version on Maven but it did
>>>>> not work
>>>>> > (reasoning unit tests failed).
>>>>> >
>>>>> > I would propose to wait until there's a compatible OWL 2
>>>>> reasoner, and then
>>>>> > replace the artifacts in org.apache.stanbol.commons.owl wdyt?
>>>>> >
>>>>> > Best,
>>>>> > Alessandro
>>>>> >
>>>>> > --
>>>>> > M.Sc. Alessandro Adamou
>>>>> >
>>>>> > Alma Mater Studiorum - Università di Bologna
>>>>> > Department of Computer Science
>>>>> > Mura Anteo Zamboni 7, 40127 Bologna - Italy
>>>>> >
>>>>> > Semantic Technology Laboratory (STLab)
>>>>> > Institute for Cognitive Science and Technology (ISTC)
>>>>> > National Research Council (CNR)
>>>>> > Via Nomentana 56, 00161 Rome - Italy
>>>>> >
>>>>> >
>>>>> > "I will give you everything, just don't demand anything."
>>>>> > (Ettore Petrolini, 1917)
>>>>> >
>>>>> > Not sent from my iSnobTechDevice
>>>>> >
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Fabian
>>>>> http://twitter.com/fctwitt
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Fabian
>>>>> http://twitter.com/fctwitt
>>>>>
>>>>>
>>>> --
>>>> M.Sc. Alessandro Adamou
>>>>
>>>> Alma Mater Studiorum - Università di Bologna
>>>> Department of Computer Science
>>>> Mura Anteo Zamboni 7, 40127 Bologna - Italy
>>>>
>>>> Semantic Technology Laboratory (STLab)
>>>> Institute for Cognitive Science and Technology (ISTC)
>>>> National Research Council (CNR)
>>>> Via Nomentana 56, 00161 Rome - Italy
>>>>
>>>>
>>>> "I will give you everything, just don't demand anything."
>>>> (Ettore Petrolini, 1917)
>>>>
>>>> Not sent from my iSnobTechDevice
>>>>
>>>>
>>>>
>>>
>>
>> --
>> M.Sc. Alessandro Adamou
>>
>> Alma Mater Studiorum - Università di Bologna
>> Department of Computer Science
>> Mura Anteo Zamboni 7, 40127 Bologna - Italy
>>
>> Semantic Technology Laboratory (STLab)
>> Institute for Cognitive Science and Technology (ISTC)
>> National Research Council (CNR)
>> Via Nomentana 56, 00161 Rome - Italy
>>
>>
>> "I will give you everything, just don't demand anything."
>> (Ettore Petrolini, 1917)
>>
>> Not sent from my iSnobTechDevice
>>
>>
>
>
> --
> Fabian
> http://twitter.com/fctwitt
>
--
Fabian
http://twitter.com/fctwitt
Re: OWL API in Maven Central
Posted by Alessandro Adamou <ad...@cs.unibo.it>.
On 01/03/2013 07:31 PM, Rupert Westenthaler wrote:
> I think commons.owl is the correct module to embed them. Just make
> sure to use "<_exportcontents>" instead of "<Export-Package>" for
> packages exported from embedded jar files. Also make sure that you add
> the correct "version" tags to exported packages.
Yep, commons.owl was already using <_exportcontents> to preserve the old
owlapi jar. I've updated it to reference the maven modules. I've also
set the version tags to a value set for new parameter "owlapi-version"
in the parent.
Now I've removed the repository and upgraded to 3.3, also closed
STANBOL-265 <https://issues.apache.org/jira/browse/STANBOL-265> and
STANBOL-626. <https://issues.apache.org/jira/browse/STANBOL-626>
Alessandro
Re: OWL API in Maven Central
Posted by Rupert Westenthaler <ru...@gmail.com>.
> I expect some integration tests to fail, because we'll need to embed many
> additional OWL API packages: those two artifacts are enough for compiling,
> but I'm afraid we're going to need a lot more for the ontologymanager to
> work in the launcher, unless there is some readily available OSGi bundle for
> the OWL API and its bindings. If not, is it OK if I use the commons.owl
> bundle for embedding them?
I think commons.owl is the correct module to embed them. Just make
sure to use "<_exportcontents>" instead of "<Export-Package>" for
packages exported from embedded jar files. Also make sure that you add
the correct "version" tags to exported packages.
best
Rupert
>
> Best,
> Alessandro
>
--
| Rupert Westenthaler rupert.westenthaler@gmail.com
| Bodenlehenstraße 11 ++43-699-11108907
| A-5500 Bischofshofen
Re: OWL API in Maven Central
Posted by Alessandro Adamou <ad...@cs.unibo.it>.
Hi,
On 01/03/2013 07:51 AM, Fabian Christ wrote:
>> btw it's eating a lot more than before: 1.5g heap and 256m permgen was no
>> longer enough when I tried it on a virtualised Linux.
> Yes, it is because of the new security components introduced by Reto. They
> are activated in the full launcher by default. Maybe you need your own
> launcher configuration if you do not want this.
Actually it is the _build_ that's eating that much, goes out of memory
long before reaching the launchers. I needed at least -Xmx2g and
-XX:MaxPermSize=384m to get it to compile.
At any rate, I've tried the OWL API 3.3 dependencies instead of 3.4.2,
like this:
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
<artifactId>owlapi-api</artifactId>
<version>3.3</version>
</dependency>
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
<artifactId>owlapi-apibinding</artifactId>
<version>3.3</version>
</dependency>
It managed to build and test fine on me under this environment:
Apache Maven 3.0.4
Maven home: /usr/share/maven
Java version: 1.6.0_24, vendor: Sun Microsystems Inc.
Java home: /usr/lib/jvm/java-6-openjdk-i386/jre
Default locale: en_US, platform encoding: UTF-8
OS name: "linux", version: "3.2.0-35-generic", arch: "i386", family: "unix"
BUT so far I've only tried commons.owl and the ontologymanager reactor.
Now I'm rebuilding the whole Stanbol. I'll keep you updated.
I expect some integration tests to fail, because we'll need to embed
many additional OWL API packages: those two artifacts are enough for
compiling, but I'm afraid we're going to need a lot more for the
ontologymanager to work in the launcher, unless there is some readily
available OSGi bundle for the OWL API and its bindings. If not, is it OK
if I use the commons.owl bundle for embedding them?
Best,
Alessandro
Re: OWL API in Maven Central
Posted by Fabian Christ <ch...@googlemail.com>.
2013/1/2 <ad...@cs.unibo.it>
> btw it's eating a lot more than before: 1.5g heap and 256m permgen was no
> longer enough when I tried it on a virtualised Linux.
>
Yes, it is because of the new security components introduced by Reto. They
are activated in the full launcher by default. Maybe you need your own
launcher configuration if you do not want this.
--
Fabian
http://twitter.com/fctwitt
Re: OWL API in Maven Central
Posted by ad...@cs.unibo.it.
I just returned able to compile Stanbol. Let me give it a try like tomorrow.
btw it's eating a lot more than before: 1.5g heap and 256m permgen was no
longer enough when I tried it on a virtualised Linux.
Alessandro
> Hi,
>
> is there a chance that we can solve this soon? Otherwise I am planning to
> create a branch for the commons release and exclude the OWL stuff.
>
> Best,
> - Fabian
>
>
> 2012/12/19 Fabian Christ <ch...@googlemail.com>
>
>> Hi,
>>
>> I tried to update the commons/owl with these deps
>>
>> <dependency>
>> <groupId>net.sourceforge.owlapi</groupId>
>> <artifactId>owlapi-api</artifactId>
>> <version>3.4.2</version>
>> </dependency>
>> <dependency>
>> <groupId>net.sourceforge.owlapi</groupId>
>> <artifactId>owlapi-apibinding</artifactId>
>> <version>3.4.2</version>
>> </dependency>
>>
>> and this embed instructions
>>
>> <Embed-Dependency>owlapi-api,owlapi-apibinding</Embed-Dependency>
>>
>> while the commons compiles fine with test I had no luck with the
>> Ontologymanager.
>>
>> These tests are now failing
>>
>>
>> org.apache.stanbol.ontologymanager.multiplexer.clerezza.ontology.TestOntologyReconciliation.versionedOnlyFromURL()
>>
>> org.apache.stanbol.ontologymanager.multiplexer.clerezza.ontology.TestOntologyReconciliation.versionedOnlyFromStream()
>>
>> and I get a compile error in the registry manager because the method
>> 'add(org.semanticweb.owlapi.model.IRI)' can not be found now.
>>
>> I gave up at this point but maybe my try helps to get it right.
>>
>> Best,
>> - Fabian
>>
>> 2012/12/14 Alessandro Adamou <ad...@cs.unibo.it>
>>
>>> On 12/13/12 9:38 PM, Fabian Christ wrote:
>>>
>>>> okay no problem. Is there something I could do to remove the embedded
>>>> repo
>>>> that is inline what you have in mind?
>>>>
>>>
>>> I am not lucid enough to think of one, but I do remember having tried
>>> to
>>> replace the embedded artifact with at least a couple of those in Maven
>>> (owlapi-api and owlapi-apibinding IIRC) in commons/owl. I think only
>>> the
>>> reasoner's unit tests were failing.
>>>
>>> As for the reasoner I was thinking about HermiT again - to have it as
>>> an
>>> externally-contributed LGPL bundle not in the release. But a new HermiT
>>> has
>>> to come out first.
>>>
>>> Just let me get out of this thesis thing and I'll have a look.
>>>
>>> Alessandro
>>>
>>>
>>> 2012/12/13 Alessandro Adamou <ad...@cs.unibo.it>
>>>>
>>>> There isn't a compatible OWL 2 reasoner that can be distributed yet.
>>>> I
>>>>> guess we can release commons-owl and make do without one for the time
>>>>> being. I'm confident a reasoner will show up eventually.
>>>>>
>>>>> However I am absolutely unable to do this right now. The deadline for
>>>>> my
>>>>> Ph.D. thesis is in the next 2 days. Real sorry.
>>>>>
>>>>> Alessandro
>>>>>
>>>>>
>>>>>
>>>>> On 12/13/12 4:29 PM, Fabian Christ wrote:
>>>>>
>>>>> Hi,
>>>>>>
>>>>>> meanwhile OWL API 3.4.2 is available at Maven central.
>>>>>>
>>>>>> Can we solve this old issue now? [1]
>>>>>>
>>>>>> Would be really great since this blocks a release of the commons
>>>>>> module
>>>>>> that I would like to make.
>>>>>>
>>>>>> [1]
>>>>>> https://issues.apache.org/****jira/browse/STANBOL-626<https://issues.apache.org/**jira/browse/STANBOL-626>
>>>>>> <https:**//issues.apache.org/jira/**browse/STANBOL-626<https://issues.apache.org/jira/browse/STANBOL-626>
>>>>>> >
>>>>>>
>>>>>>
>>>>>>
>>>>>> 2012/9/14 Fabian Christ <christ.fabian@googlemail.com <mailto:
>>>>>> christ.fabian@**googlemail.com <ch...@googlemail.com>**>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> thanks for trying this. I have no problem with waiting for a
>>>>>> compatible version. It just keeps to be a little annoying
>>>>>> situation
>>>>>> but if there is hope that there will be a compatible version
>>>>>> soon, we
>>>>>> can wait.
>>>>>>
>>>>>> It is more a question of when do we and our users would like to
>>>>>> see a
>>>>>> clean release of this stuff.
>>>>>>
>>>>>> Best,
>>>>>> - Fabian
>>>>>>
>>>>>> 2012/9/13 Alessandro Adamou <adamou@cs.unibo.it
>>>>>> <ma...@cs.unibo.it>>:
>>>>>>
>>>>>> > Hi,
>>>>>> >
>>>>>> > the OWL API finally made it to Maven Central. check
>>>>>> > net.sourceforge.owlapi:owlapi-*****
>>>>>>
>>>>>> >
>>>>>> > However it's versions 3.3 onwards, and the latest HermiT is
>>>>>> compatible with
>>>>>> > OWL API 3.2.2. I tried it with the version on Maven but it
>>>>>> did
>>>>>> not work
>>>>>> > (reasoning unit tests failed).
>>>>>> >
>>>>>> > I would propose to wait until there's a compatible OWL 2
>>>>>> reasoner, and then
>>>>>> > replace the artifacts in org.apache.stanbol.commons.owl wdyt?
>>>>>> >
>>>>>> > Best,
>>>>>> > Alessandro
>>>>>> >
>>>>>> > --
>>>>>> > M.Sc. Alessandro Adamou
>>>>>> >
>>>>>> > Alma Mater Studiorum - Università di Bologna
>>>>>> > Department of Computer Science
>>>>>> > Mura Anteo Zamboni 7, 40127 Bologna - Italy
>>>>>> >
>>>>>> > Semantic Technology Laboratory (STLab)
>>>>>> > Institute for Cognitive Science and Technology (ISTC)
>>>>>> > National Research Council (CNR)
>>>>>> > Via Nomentana 56, 00161 Rome - Italy
>>>>>> >
>>>>>> >
>>>>>> > "I will give you everything, just don't demand anything."
>>>>>> > (Ettore Petrolini, 1917)
>>>>>> >
>>>>>> > Not sent from my iSnobTechDevice
>>>>>> >
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Fabian
>>>>>> http://twitter.com/fctwitt
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Fabian
>>>>>> http://twitter.com/fctwitt
>>>>>>
>>>>>>
>>>>> --
>>>>> M.Sc. Alessandro Adamou
>>>>>
>>>>> Alma Mater Studiorum - Università di Bologna
>>>>> Department of Computer Science
>>>>> Mura Anteo Zamboni 7, 40127 Bologna - Italy
>>>>>
>>>>> Semantic Technology Laboratory (STLab)
>>>>> Institute for Cognitive Science and Technology (ISTC)
>>>>> National Research Council (CNR)
>>>>> Via Nomentana 56, 00161 Rome - Italy
>>>>>
>>>>>
>>>>> "I will give you everything, just don't demand anything."
>>>>> (Ettore Petrolini, 1917)
>>>>>
>>>>> Not sent from my iSnobTechDevice
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>> --
>>> M.Sc. Alessandro Adamou
>>>
>>> Alma Mater Studiorum - Università di Bologna
>>> Department of Computer Science
>>> Mura Anteo Zamboni 7, 40127 Bologna - Italy
>>>
>>> Semantic Technology Laboratory (STLab)
>>> Institute for Cognitive Science and Technology (ISTC)
>>> National Research Council (CNR)
>>> Via Nomentana 56, 00161 Rome - Italy
>>>
>>>
>>> "I will give you everything, just don't demand anything."
>>> (Ettore Petrolini, 1917)
>>>
>>> Not sent from my iSnobTechDevice
>>>
>>>
>>
>>
>> --
>> Fabian
>> http://twitter.com/fctwitt
>>
>
>
>
> --
> Fabian
> http://twitter.com/fctwitt
>