--
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/bc736582-4c38-4553-a668-d4e6f20cfe03n%40googlegroups.com.
Dear Lam,
VocBench’s backend platform, Semantic Turkey, has API for accessing and modifying its content.
http://semanticturkey.uniroma2.it/doc/user/first_access.jsf
http://semanticturkey.uniroma2.it/doc/user/web_api.jsf
You can write your own client in Python for using them.
Kind Regards,
Armando
.
To view this discussion on the web visit https://groups.google.com/d/msgid/vocbench-user/CAP9uDus6nX35fzXn_6a7vhY6qFd1E1uGOQQ%2B8-GVmrJfvtVcmw%40mail.gmail.com.
Dear all,
this is a more direct approach. The advantage in directly connecting to the triple store (GraphDB or RDF4J, whichever you are using) is that you can reuse some existing client for accessing specific tripe stores or at least compliant with the SPARQL protocol.
However, by using the ST API you would benefit from the full range of high-level API that can take care of all specific triples to be modified following a certain action. Also, passing by ST would be better in case history or validation are activated on the project.
In principle, both can be used. Let’s say that through-ST you benefit from all the existing logic of the platform and is generally safer. The other one is like tinkering under the hood, but you benefit from existing clients :-)
Kind Regards,
Armando
From: vocben...@googlegroups.com <vocben...@googlegroups.com>
On Behalf Of Roland Wingerter
Sent: Tuesday, May 3, 2022 11:32 AM
To: vocbench-user <vocben...@googlegroups.com>
Subject: Re: [vocbench-user] ERROR: Query evaluation exception caught
Hi Lam,
.
To view this discussion on the web visit https://groups.google.com/d/msgid/vocbench-user/0c33ed27-c074-460c-ba3a-fe3679074d23n%40googlegroups.com.
Dear Lam,
A few notes first about the needs you mentioned in the paragraph before the questions. While not being a VB vs WebProtege thing, I just picked your list in order to check if other aspects you found in WebProtege can be easily covered in VB. While we can’t change the platform in a breeze, it’s important for us to get feedback or to show to you things you were not aware of.
Lightweight editor, flexible editing: I’m interested in this as, since you considered WebProtege, you are probably dealing with ontology and/or data (thesauri and lexicons are not supported by WebProtege). Is there anything that seems easier to be done?
Speaking about OWL, I find the Manchester editor quite easy and fast to be used, with syntax completion (not limited to the language, but extended to the same elements you are modeling), highlighting and toggle for syntactic/semantic checks. The rest: creating resources, linking them, should be pretty easy. Anyhow, if you find margin for improvement, we are all ears :-)
Comment function. You can go with two approaches:
this is even better than just comments as you can keep discussions going on a separated platform, create issues of different types, mark them as in progress, resolved etc… You can do everything from within VocBench only using the connected tool as a backend or you can open links to the specific connected tool in order to work directly from there and use some tool-specific functionality
Now, to the explicit question you made:
The scenario you depicted is ok, but a precise answer depends on your specific case. If the resources are separated resources, the best thing is to keep them as separated projects and then align them.
For alignment, there are various solutions, from completely manual, passing through some automation to full automation and validation, see these links:
http://vocbench.uniroma2.it/doc/user/data_view.jsf#alignment_to_an_external_resource
http://vocbench.uniroma2.it/doc/user/alignment_validation.jsf
http://vocbench.uniroma2.it/doc/user/alignment_systems.jsf
However, if there’s a specific owl:import relation (e.g. one resource is an ontology and one another is a dataset importing that same ontology), then it is better that you manage the ontology in one project, periodically (I mean, when you freeze a new release) you export it, and then you owl:import it (http://vocbench.uniroma2.it/doc/user/namespaces_imports.jsf ) into your dataset. That is not linking indeed, it’s modeling (that is, your ontology resources are a vocabulary for describing the instance data in your dataset).
Another approach, if you want to keep things together, is that you put everything in one project. Let’s take for example the previous case of ontology O + dataset D and say that you have already a starting dataset and a starting ontology.
You create a project for dataset D (with baseuri/namespace for dataset D), then you owl:import the ontology O , so it will be imported in another graph.
Now, normally you will be able to edit only the dataset and the ontology will be read-only. However, you can switch the working graph on VB and change it to the one of the ontology, and then edit the ontology.
I personally prefer the first one, as I believe that, if the ontology has a life of its own and is possibly used by other datasets, it should have its own project and, in any case, you should change it periodically and not continuously. However, you might decide to follow this latter approach in a case of a more tight connection between ontology and data.
Kind Regards,
Armando
To view this discussion on the web visit https://groups.google.com/d/msgid/vocbench-user/2365272e-8dca-4550-bb67-e7121c559d89n%40googlegroups.com.
curl -X 'POST' \
'http://localhost:1979/semanticturkey/it.uniroma2.art.semanticturkey/st-core-services/Auth/login?email=mye...@gmail.com&password=mypassword&_spring_security_remember_me=true' \
-H 'accept: application/json' \
-H 'Content-Type: application/json' \
-d '{}'
curl -X 'POST' \
'http://127.0.0.1:1979/semanticturkey/it.uniroma2.art.semanticturkey/st-core-services/Projects/listProjects' \
-H 'accept: application/json' \
-H 'Content-Type: application/json' \
-d '{}'
Dear Lam,
Eventually we will develop some tool to automatically produce client API docs for all the services in the platform. However, since currently there are hundreds of services, we described various ways to learn the API from available material (javadocs and source code, which are both available).
However, short recipe that can help you going directly to your needs and easily reproducible: run your browser and activate the devtools (e.g. CTRL+SHIFT+J on Chrome/Edge, CTRL+SHIFT+I on Firefox) , select “network”, then check all calls being done. You just need then to reproduce them when invoking your methods. Just remember each time you run a batch with various calls to preempt the login first (as you have already learned).
Kind Regards,
Armando
From: vocben...@googlegroups.com <vocben...@googlegroups.com>
On Behalf Of Lam Nguyen
Sent: Friday, May 6, 2022 11:10 AM
To: Armando Stellato <stel...@uniroma2.it>
Cc: Tiziano Lorenzetti <tiziano.l...@gmail.com>; vocbench-user <vocben...@googlegroups.com>
Subject: {Disarmed} Re: [vocbench-user] ERROR: Query evaluation exception caught
Dear Armando,
I followed the instruction from your links.
I could log in using the command below:
curl -X 'POST' \
'http://localhost:1979/semanticturkey/it.uniroma2.art.semanticturkey/st-core-services/Auth/login?email=mye...@gmail.com&password=mypassword&_spring_security_remember_me=true' \
-H 'accept: application/json' \
-H 'Content-Type: application/json' \
-d '{}'
However, I could not list the projects:
To view this discussion on the web visit https://groups.google.com/d/msgid/vocbench-user/CAP9uDutJSsiqRO3DrCTFj0gES6%2BRknZ6zZtGXBBOXv4V%3DahSaw%40mail.gmail.com.