[
https://issues.apache.org/jira/browse/JENA-1251?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15962149#comment-15962149
]
ASF GitHub Bot commented on JENA-1251:
--------------------------------------
Github user afs commented on the issue:
https://github.com/apache/jena/pull/211
Could you close it from you end? These mirrored GH-repos don't behave
exactly as if they were the master and we, the PMC, don't get a "close" button.
We can close unmerged ones but it means committing to the code base.
I have found one issue - it should not be using `NodeToLabel labels` at
all. In the clear-up of RIOT, the label mapping is passed in. I have fixed this
in PR#234.
Side-effect - all nodes get created via `ParserProfile`/`FactoryRDF` which
includes checking of URIs: this addresses JENA-1251.
However, it seems that java-jsonld is emitting bnodes without their
original label - I get "_:b0" out despite the input having "_:BBB", when the
blank node is created in `JsonLDReader.createNode(Map)`
> Parsing JSON-LD needs to do more checking of terms.
> ---------------------------------------------------
>
> Key: JENA-1251
> URL: https://issues.apache.org/jira/browse/JENA-1251
> Project: Apache Jena
> Issue Type: Bug
> Affects Versions: Jena 3.1.0, Fuseki 2.4.0
> Reporter: Andy Seaborne
>
> https://lists.apache.org/thread.html/d38c9d81b88448f57d68adfc6ed336e90c6a5314fdc96b8786ca0832@%3Cusers.jena.apache.org%3E
> JsonLDReader calls {{ParserProfileChecker.createURI}} but term checking of
> IRIs happens in CheckerIRI which is a tokenizer level feature and JSON-LD
> does not use RIOT tokenizing.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
