[ 
https://issues.apache.org/jira/browse/JENA-811?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Andy Seaborne closed JENA-811.
------------------------------
       Resolution: Fixed
    Fix Version/s: Jena 2.12.2

This fixes the presenting issue, making behaviour (more) consistent across all 
Jena parsers.  There is not change in global IRI policy; the fix is to be more 
relaxed when resolving theRDF/XML base IRI in RIOT.

> org.apache.jena.riot.system.IRIResolver doesn't allow for any control over 
> violation error/warning settings
> -----------------------------------------------------------------------------------------------------------
>
>                 Key: JENA-811
>                 URL: https://issues.apache.org/jira/browse/JENA-811
>             Project: Apache Jena
>          Issue Type: Bug
>          Components: IRI
>    Affects Versions: Jena 2.11.2
>            Reporter: Matthew Jarvis
>            Assignee: Andy Seaborne
>            Priority: Minor
>             Fix For: Jena 2.12.2
>
>
> The IRIResolver class has a public static iriFactory that is used for 
> creating IRIs, but consumers never have a chance to call 
> IRIResolver.iriFactory#setIsError or IRIResolver.iriFactory#setIsWarning 
> because the iriFactory is used in a static block when the class is loaded.  
> Any attempts to call either the setIsError or setIsWarning method throw the 
> following exception:
> {noformat}
> java.lang.IllegalStateException: Cannot reinitialize IRIFactory after first 
> use.
>       at 
> org.apache.jena.iri.impl.IRIFactoryImpl.initializing(IRIFactoryImpl.java:255)
>       at org.apache.jena.iri.IRIFactory.setIsError(IRIFactory.java:579)
> ...
> {noformat}
> A simple test case that illustrates the problem is:
> {noformat}
> @Test
> public void testIgnoreLowercaseViolation() {
>   IRIResolver.iriFactory.setIsError(ViolationCodes.LOWERCASE_PREFERRED, 
> false);
>   IRIResolver.resolveIRI("http://EXAMPLE.com";);
> }
> {noformat}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to