Re: [topbraid-users] exporting spin functions to fuseki

2015-02-26 Thread Holger Knublauch
Hi Jack, short answer: You'll need the SPIN API [1] as a plugin to Fuseki. Fuseki by itself does not know what to do with the RDF declarations of SPIN functions - they are just triples for Fuseki. BTW afn:localname is not just a TopBraid function, but is originally from Jena. So it works in

[topbraid-users] exporting spin functions to fuseki

2015-02-26 Thread Jack Hodges
I have exported a bunch of graphs to a .ttl file and then imported it into fuseki. If I run a query on fuseki to see if the functions are there - they are. If I run a query on fuseki to show the spin:body or spin:constraints they show them as anonymous objects (is that the correct term?). If I r

Re: [topbraid-users] SDB vs TDB

2015-02-26 Thread Irene Polikoff
Charlie, Transactions against SDB are atomic. There will never be a partial commit. There is an in-memory buffering graph in the write-through cache making sure that operations are indeed atomic. Data doesn¹t corrupt. And multiple concurrent transactions should execute the same as if they were do

Re: [topbraid-users] SPIN API usage

2015-02-26 Thread Holger Knublauch
Sorry for this kind of question we need a minimal reproducible example/test case. It is really difficult to guess what's wrong. If you suspect sp:text, then you could confirm this yourself if you just switch your file back to RDF triples. Holger On 2/26/2015 20:55, Claudia Grieco wrote: Ther

Re: [topbraid-users] SDB vs TDB

2015-02-26 Thread Charles Mead
Hi Scott and Irene --) As always, thank you for taking the time to address the questions, all of which came up in the context of discussing EVN and its SDB (or Virtuoso or other) persistence bindings. The core question(s) revolve around the notion of how the definition of "transaction" -- which i

Re: [topbraid-users] SDB vs TDB

2015-02-26 Thread Scott Henninger
Charlie; In terms of transaction management, one way to understand it is that since a relational back-end is used to store all data, the transactional guarantees from the relational system hold. I.e., all changed are executed as fully or roll-back and retry; however the relational system performs

Re: [topbraid-users] SDB vs TDB

2015-02-26 Thread Irene Polikoff
Charlie, I am not sure I fully understand the question, but I will try to answer in somewhat high level terms. If you need more details, maybe Scott could answer or we could pass your question on to the development team. As multiple users make edits, these edits are written into a database. Mor

Re: [topbraid-users] SDB vs TDB

2015-02-26 Thread Charles Mead
Thanks Irene. A follow-up question is how the transaction management that lives in SDB (and that no one here doubts) is (or is not) manifest in EVN per se. In other words, what is the notion of "transaction management" at the graph level independent of its commitment to persistence through SDB?

Re: [topbraid-users] SDB vs TDB

2015-02-26 Thread Irene Polikoff
We recommend SDB in the multi user read/write scenarios because it has better support for transactions. If you have a load and read-only scenario, TDB is a good choice. TopBraid includes a streaming data loader for TDB. Note, however, that there are some known issues with TDB on Windows. They ha

[topbraid-users] SDB vs TDB

2015-02-26 Thread Charles Mead
Are there performance and size metrics available for why one would choose to use one or the other of these approaches as an export/transfer format? Thanks -- charlie -- You received this message because you are subscribed to the Google Group "TopBraid Suite Users", the topics of which include

Re: [topbraid-users] SPIN API usage

2015-02-26 Thread Claudia Grieco
There are some rules that are applied to super classes of the instances, but some simple rules (ex. insert height and width to an instance of 'rectangle') shouldn't have these problems. My rules are in sp:text, is this a problem? I'm using Spin API 1.4.0 bye Claudia Il giorno giovedì 26 febbraio