You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@clerezza.apache.org by "Enrico Daga (JIRA)" <ji...@apache.org> on 2014/04/11 19:04:23 UTC
[jira] [Commented] (CLEREZZA-908) Cannot save literals of datatype
rdf:XMLLiteral with the virtuoso storage
[ https://issues.apache.org/jira/browse/CLEREZZA-908?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13966789#comment-13966789 ]
Enrico Daga commented on CLEREZZA-908:
--------------------------------------
I propose to do the following, while we wait for Virtuoso people to fix these issues.
1) Remove the TTL function trick and use the classic and more efficient method
2) Shadow the rdf:XMLLiteral datatype so that it is saved in the storage as a generic custom datatype, say <urn:x-clerezza-virtuoso:rdf#XMLLiteral>
in this way we will have both things. Of course if the applications uses the Virtuoso storage as shared database, this will not be good. But we can live with this for the moment.
> Cannot save literals of datatype rdf:XMLLiteral with the virtuoso storage
> -------------------------------------------------------------------------
>
> Key: CLEREZZA-908
> URL: https://issues.apache.org/jira/browse/CLEREZZA-908
> Project: Clerezza
> Issue Type: Bug
> Reporter: Enrico Daga
> Assignee: Enrico Daga
>
> This problem raised some time ago for the first time but it needs some time to be clarified. Discussions has been done in the virtuoso mailing list [1].
> Specifically, inserting a triple with a value of type rdf:XMLLiteral caused a string "0" to be inserted. At that time a partial solution was to serialize the whole triple as TTL and use a stored procedure db.dba.ttl(...) to save the triple. This worked, but caused the triple to loose the datatype (see [2]).
> This is also what is currently implemented.
> Anyway, this solution is bad for other reasons, especially:
> - performance is too bad (serializes any triple...)
> - you still don't have the correct datatype (there are two tests with "Ignore" annotation in VirtuosoMGraphTest that witness this)
> [1] http://sourceforge.net/p/virtuoso/mailman/message/31987368/
> [2] http://sourceforge.net/p/virtuoso/mailman/message/31992357/
--
This message was sent by Atlassian JIRA
(v6.2#6252)