Publish Ontology

44 views
Skip to first unread message

Imran Asif

unread,
Jan 10, 2023, 6:23:04 AM1/10/23
to vocbench-user
Hi,

I have a question that can we publish ontology from vocbench to GitHub/w3id.org for making the namespace consistent?

Thanks and would appreciate your answers.

Kind Regards,
Imran

Armando Stellato

unread,
Jan 10, 2023, 11:22:45 AM1/10/23
to Imran Asif, vocbench-user

Dear Imran,

 

Let’s divide it into two points:

 

There are two aspects here:

so then this is a plain storage on a server.
VocBench has a functionality for deploying dumps to locations; see here: http://vocbench.uniroma2.it/doc/user/global_data_management.jsf#export_data and here: http://semanticturkey.uniroma2.it/doc/sys/deployer.jsf . As you can see you can use the SFTP deployer if you have SFTP access to the web server where you deploy your ontology, or you can configure the generic HTTP deployer if you have other HTTP-based services. Furthermore, you could use the GraphStore deployer if you are deploying a whole dataset (thus using a triple store and not just a file, as it is common for ontologies)

 

  • since you mentioned GitHub, are you considering using GitHub/pages or the GIT service of GitHUB? Concerning the second one, currently there is no deployer for GIT. However, the deployer functionality is an extension point, and plugins for that can be developed. If interested in it, pls let us know, we can provide indications for that

 

Kind Regards,

 

Armando

 

 

--
You received this message because you are subscribed to the Google Groups "vocbench-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vocbench-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/vocbench-user/00678ec5-a42f-4553-934d-e206025ab17en%40googlegroups.com.

Imran Asif

unread,
Jan 11, 2023, 9:23:53 AM1/11/23
to vocbench-user
Thanks, Armando,

You explained very well, and I got your explanation. Yes, I am interested in storing my ontology on GitHub and configuring this ontology with w3id.org. Because w3id.org (https://w3id.org/#new) provide quite well and easy configuration for GitHub.

One more question regarding the Ontology mapping/alignment/matching, Is there any feature that I can use VocBecnh to achieve ontology matching?

Thanks again for your valuable response.

Kind Regards,
Imran

Armando Stellato

unread,
Jan 13, 2023, 6:41:27 AM1/13/23
to Imran Asif, vocbench-user

Dear Imran,

 

about the alignment, pls give a look at:

 

http://vocbench.uniroma2.it/doc/user/alignment_validation.jsf

and the subpage:

http://vocbench.uniroma2.it/doc/user/alignment_systems.jsf

 

NAISC is currently not updated to the latest version of the alignment interface though. Other linkable systems are coming though thanks to collaborations with their respective development groups

 

About GiT, my question was if you are interested in developing the related plugin, so that, in case, we can provide you indications. Otherwise, we might develop that in the future, but it’s not on our short-term roadmap.

 

Kind Regards,

 

Armando

 

 

Imran Asif

unread,
Jan 16, 2023, 4:03:39 AM1/16/23
to vocbench-user
Hi Armando,

Thanks for your reply. I will take a look at the alignment links you gave me. About the GIt, can you please provide the documentation where I can find the plugin's development?

Very kind regards,
Imran

Manuel Fiorelli

unread,
Jan 17, 2023, 5:44:39 PM1/17/23
to Imran Asif, vocbench-user
Dear Imran,

you can find a general overview of extension point development on  the Semantic Turkey site (http://semanticturkey.uniroma2.it/documentation/dev/extension_point_devguide.jsf); however, it has not been updated to recent developments of the framework.

So, the best thing to do in my opinion is to look on existing implementations:
Conceptually, these two deployers differ in the nature of their input: the former is fed with a data stream (being completely indifferent to its content), while the latter is fed with "abstract" RDF data (expressed using the RDF4J API), so that any conversion/marshalling must be implemented internally.

Let us look at the latter in details.

Firstly, each core extension is implemented in a more or less self-contained Maven module, which can be used as a basis to implement a new one.

  •  ShowVocDeployerFactory.java instantiate an object from this extension class. Being a ConfigurableExtensionFactory, this factory requires a configuration to produce the instance. The interface
    PUScopedConfigurableComponent enables storing the configuration at the PU scope and more general ones. This interfaces enables the "save configuration" button near the selector of a configuration in the UI.
  • ShowVocDeployer.java is the actual implementation. Your main job is to implement the method void deploy(RepositorySource source)
  • XXXConfiguration are the configuration classes. The idea is that a configuration is modeled as a class with instance attributes annotated with STProperty representing configuration properties
  • the file st-showvoc-deployer-context.xml is a spring context that instantiates the factory and makes it available on the service context.
My suggestion on how to proceed. Try to identify which kind of deployer you need, and the using one of the two I indicated to you, create a new Maven project outside of Semantic Turkey. Your main duty will be to consistently rename the classes and their references (e.g. in the Spring context definition). As a start, clear the deploy method and simply write a message to stdout. When you can successibile deploy this empty deployer, you can start with the implementation.

Regards,
Manuel


To view this discussion on the web visit https://groups.google.com/d/msgid/vocbench-user/9d2507f6-b51a-4321-9be3-e6a7d1cf538en%40googlegroups.com.


--
Manuel Fiorelli
Reply all
Reply to author
Forward
0 new messages