Importing external change history

[Boris Pelakh]

I am in the process of migrating some taxonomies from PoolParty into TopBraid EDG. The simple path is an RDF export of the taxonomy (which is SKOS and dcterms-compliant), and then import the RDF after creating the taxonomy in EDG. I have a couple of questions:

1. Is there an API endpoint to create a new taxonomy from RDF content, i.e. without manual interaction, so I can automate the process?
2. I have a way of exporting revision history from PoolParty either as JSON or RDF. Is there a way for me to import this history into EDG (I can transform it into any schema required)

Thanks!

[Holger Knublauch]

Hi Boris,

On 2021-05-13 10:33 am, Boris Pelakh wrote:

I am in the process of migrating some taxonomies from PoolParty into TopBraid EDG. The simple path is an RDF export of the taxonomy (which is SKOS and dcterms-compliant), and then import the RDF after creating the taxonomy in EDG. I have a couple of questions:

1. Is there an API endpoint to create a new taxonomy from RDF content, i.e. without manual interaction, so I can automate the process?

Yes there are various options, depending on what input file(s) you have and whether the taxonomies already exist and you just want to replace the content. Basically if you click through the UI with a network monitor you can see which web services are called. For example you could use teamwork:ImportTrigFileService which is the service behind the + > Create Multiple Asset Collections from TriG File  button. You can also write SWP or ADS scripts that actively pull content from external web services. Or you could call a combination of the teamwork:createProject service and tbl/importFileUpload.

What is your starting point: individual Turtle files but no existing Taxonomy yet?

1. I have a way of exporting revision history from PoolParty either as JSON or RDF. Is there a way for me to import this history into EDG (I can transform it into any schema required)

The change history of EDG graphs is stored in so-called .tch graphs using RDF. For example you could create instances of teamwork:Change using SPARQL. If the PP history is already in RDF, I guess you could do an INSERT { ... } WHERE { ... } update using SPARQL.

Holger

[Boris Pelakh]

Thank you, Holger, that's great. I was able to find the ontology at https://www.topbraid.org/teamwork, but are you aware of any sample Change data I can look at as examples?

[David Price]

On 13 May 2021, at 17:37, Boris Pelakh <pel...@gmail.com> wrote:

Thank you, Holger, that's great. I was able to find the ontology at https://www.topbraid.org/teamwork, but are you aware of any sample Change data I can look at as examples?

I test running EDG from Composer, doing some changes and then exit EDG and look at the tch graph of the collection you edited. Use that pattern and you’re sure whatever you generate matches what looks nice when displayed in EDG.

Cheers,

David

--
You received this message because you are subscribed to the Google Groups "TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to topbraid-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/topbraid-users/5b3ef746-63c3-4979-9430-cb9072778382n%40googlegroups.com.

UK +44 (0) 7788 561308

US +1 (336) 283-0808‬

[Irene Polikoff]

Example of how changes look like is provided at https://doc.topquadrant.com/7.0/teamwork/

Regarding the TriG import option, there are a couple of things to know about it:

1. This import will create either asset collections in EDG (loading data into them) or RDF files in EDG’s workspace

2. Asset collections are created for graphs in the TriG file that are described as such. In other words, they have they follow the EDG naming convention (urn:x-evn-master:XY) and they specify the collection type e.g.,

<urn:x-evn-master:geo>

  a taxonomies:Taxonomy ;

  a teamwork:Vocabulary ;

  a owl:Ontology ;

This triple indicates that it should be created as a taxonomy:

<urn:x-evn-master:geo>

  a taxonomies:Taxonomy ;

If you want a graph to be created as an ontology asset collection (e.g., to contain classes or properties other than SKOS that you want to be used in your taxonomies), then skip " a taxonomies:Taxonomy”.

See: https://doc.topquadrant.com/7.0/importing/#Import_TriG_file.

Also, relevant is: https://doc.topquadrant.com/7.0/preparing-rdf-data/ 

To view this discussion on the web visit https://groups.google.com/d/msgid/topbraid-users/BD136B92-8ED4-4975-ABBE-0A83DD130BCC%40topquadrant.com.