On 09/11/10 16:59, Jeremy Carroll wrote:
I could be accused of having gone overboard ...
yes ... :-)
Having changed job and looking at this from a different perspective I am
less convinced by the pickiness.
At least leave the picky mode there, please. It's been very helpful in
working with data generated by one organisation and used by another.
Script generated IRIs have been
Once dubious data gets into a database, it's troublesome to get it out.
Having the data storage only cover the required spec common ground is
better.
TDB uses the IRI profile (not RDF URI references) as strict as possible
except that these two warnings are switched off:
ViolationCodes.UNWISE_CHARACTER
that is "{", "}", "|", "\", "^", "`"
It's the "|" that I found used in generated IRIs.
Whitespace is handled separately.
"<" ">" are syntax errors.
ViolationCodes.UNREGISTERED_IANA_SCHEME
because people invent schemes for data.
I did find the most strict mode a bit too strict :-)
Andy
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
For additional commands, e-mail: general-h...@incubator.apache.org
