[
https://issues.apache.org/jira/browse/JENA-71?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13049249#comment-13049249
]
Andy Seaborne commented on JENA-71:
-----------------------------------
The patch looks reasonable but I wonder if it's the best way of approaching
this.
The output side (OutoputLangUtils) is fine - I've applied that part. no tests
yet because the current test structure relies on reading things in again to
check.
The tokenizer (text or bytes) recognizes various beyond-RDF things - keywords
and various symbols (e.g "*").
A keyword is a word without a ":" in it. ":" means prefix name. They provide a
way to add language on top of the character stream using keywords for
structure. I can envisage wanting to write other "meta Nodes" and using
keywords is one way to do that.
In jena3, the Node hierarchy might be more flexible for sub-system specific
additions. The TDB optimizer uses meta Nodes like to indicate when a triple
pattern will a fixed value in a slot , but does not what it is. It has to use
something like Triple because Node isn't extensible.
If this works for you, conversion to Node.ANY would be done in Token.asNode by
seeing a keyword with text ANY.
> OutputLangUtils cannot encode Node.ANY
> --------------------------------------
>
> Key: JENA-71
> URL: https://issues.apache.org/jira/browse/JENA-71
> Project: Jena
> Issue Type: Bug
> Components: ARQ
> Reporter: Laurent Pellegrino
> Priority: Minor
> Labels: arq, node.any
> Attachments: JENA-71-r1131151.patch
>
>
> As stated in the mailing list
> (http://mail-archives.apache.org/mod_mbox/incubator-jena-dev/201106.mbox/%[email protected]%3E)
> and with JENA-70, Node.ANY cannot be encoded with OutputLangUtils.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
