[
https://issues.apache.org/jira/browse/JENA-488?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13707693#comment-13707693
]
Claude Warren commented on JENA-488:
------------------------------------
I suppose the other question is do we need to validate the reader class? It
seems like the tests to ensure that the reader returned by
model.getReader("N-TRIPLE") actually reads N-TRIPLEs should be sufficient.
After looking at the code. It would be simple to add the expected class to the
TestingModelFactory (it is available and varies based on who/what is building
the model), but the N-TRIPLE reader is the only one we verify the class for.
The "TestReaders.java" tests also verify that the N-TRIPLE and RDF readers can
read local and remote files. Using the reader to actually read N-TRIPLES is,
in my opinion, sufficient and the testGetNTripleReader() can be safely removed.
> TestReaders.testGetNTripleReader() is dependent upon ARQ not being loaded.
> --------------------------------------------------------------------------
>
> Key: JENA-488
> URL: https://issues.apache.org/jira/browse/JENA-488
> Project: Apache Jena
> Issue Type: Bug
> Components: Jena, RIOT
> Affects Versions: Jena 2.10.1
> Reporter: Claude Warren
> Assignee: Claude Warren
> Labels: easyfix
>
> The testGetNTripleReader test assumes that the NTriple reader will be an
> instance of NTripleReader, which is correct if ARQ has not been loaded. When
> ARQ has been loaded the result is an instance of RDFReaderRIOT.
> This bug is to fix the test to accept RDFReaderRIOT or NTripleReader as valid
> results.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
