You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@jena.apache.org by Charles Li <ch...@gmail.com> on 2013/09/04 20:03:50 UTC

How to switch off checking for illegal reuse of rdf:ID's?

Hi, All:

When I run a TDBLoader (came with Jena 2.10.1), I got a warning

     WARN  riot                 :: {W137} Input is large. Switching off
checking for illegal reuse of rdf:ID's.

- How do I switch off rdfLID checking? I looking into "TDBLoader --help"
but couldn't find any clue.
- I suppose the whole loading would be faster if I switch it off, correct?

Thanks a lot for all your help!
- Charles

Re: How to switch off checking for illegal reuse of rdf:ID's?

Posted by Andy Seaborne <an...@apache.org>.
On 04/09/13 19:03, Charles Li wrote:
> Hi, All:
>
> When I run a TDBLoader (came with Jena 2.10.1), I got a warning
>
>       WARN  riot                 :: {W137} Input is large. Switching off
> checking for illegal reuse of rdf:ID's.
>
> - How do I switch off rdfLID checking? I looking into "TDBLoader --help"
> but couldn't find any clue.

There is some kind of RDF/XML option. tdbloader does not expose it; 
you'd need code to set it.

> - I suppose the whole loading would be faster if I switch it off, correct?

Maybe but unlikely.  The cost of loading is usually raw XML parsing + 
loading (which don't play well together).

You must be at some degree of scale to the the warning.

My recommendation is parse the RDF/XML to N-triples so that it is 
completely checked, then load the NT.

Loading N-triples is faster than RDF/XML (the parser slows the loader 
down!).

With a file of any size, checking it first, so the load does not break 
half-way through, is usually a good idea to incorporate in a workflow.

	Andy

>
> Thanks a lot for all your help!
> - Charles
>