Error tdb-loading a third-party rdf file

[Eric Scott <er...@att.net>]

I'm loading an xml/rdf file provided by a third party using tdbloader2, 
and it throws this error:
--
Base URI is null, but there are relative URIs to resolve.: <>
--
The header of the rdf file does not have in fact have an xml:base 
clause, so I guess I'd need to edit the original RDF source to provide one?

In the hope of providing a meaningful value for this, I'd like to be 
able to see just what statement contained the relative URI that 
triggered this error, but tdbloader doesn't seem to be providing line 
numbers.  Searching for <> turns up nothing.  Is there a utility I can 
use that will give me this information?

Failing that, does it even matter what value I use to provide the base 
URI? If not, I guess I can just stick in any old thing.

Thanks,

Re: Error tdb-loading a third-party rdf file

[Dave Reynolds <da...@gmail.com>]

On Mon, 2011-10-24 at 09:30 -0700, Eric Scott wrote: 
> I'm loading an xml/rdf file provided by a third party using tdbloader2, 
> and it throws this error:
> --
> Base URI is null, but there are relative URIs to resolve.: <>
> --
> The header of the rdf file does not have in fact have an xml:base 
> clause, so I guess I'd need to edit the original RDF source to provide one?

Yes (in code you can set a base URI in the read call).

> In the hope of providing a meaningful value for this, I'd like to be 
> able to see just what statement contained the relative URI that 
> triggered this error, but tdbloader doesn't seem to be providing line 
> numbers.  Searching for <> turns up nothing.  Is there a utility I can 
> use that will give me this information?


Can't think of one. The chances are it is a line containing

    rdf:about=""   or rdf:about=''

so a judicious regexp search might turn it up.

> Failing that, does it even matter what value I use to provide the base 
> URI? If not, I guess I can just stick in any old thing.

Depends on the data and what you are going to do with it.

The most common use of "" as a URI is to deliberately refer to the
document, eg in an owl:Ontology. In that case the URI should really be
the URL you got the data from. However, if you regard your copy as a
different document then give it any base you like.

Dave