[
https://issues.apache.org/jira/browse/UIMA-1119?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12646550#action_12646550
]
Burn Lewis commented on UIMA-1119:
----------------------------------
Our documentation refers to the old XCAS format and the new XMI format for
representing CASes in XML. I think it is acceptable to use the terms XMI and
XCAS when referring to XML representations of CASes. Also we provide a class
org.apache.uima.util.XmlCasDeserializer that will read either ... its Javadocs
say:
Deserializes a CAS from a standoff-XML format. This class can read the XMI
format introduced in UIMA v1.4 as well as the XCAS format from previous
versions.
Eclipse cannot find any callers of any of the setDocumentLocator methods ...
perhaps we should be creating a Locator and setting it as it would be somewhat
helpful to know where the error is.
Can we not simply change the error messages to refer to XML parsing? It'll be
pretty obvious which format is being used.
Yes, and we'd also need a comment about XMI in XCASParsingException.java
Or change to xmlcas.properties & XmlParsingException.java
> The XmiCasDeserializer throws NoSuchElementException if an XCAS is corrupted,
> but doesn't report the offending element.
> -----------------------------------------------------------------------------------------------------------------------
>
> Key: UIMA-1119
> URL: https://issues.apache.org/jira/browse/UIMA-1119
> Project: UIMA
> Issue Type: Improvement
> Components: Core Java Framework
> Affects Versions: 2.2.2
> Reporter: Burn Lewis
> Assignee: Marshall Schor
> Priority: Minor
> Attachments: UIMA-1119-3.patch
>
>
> Obviously it's dangerous to edit XCASes but it is often the quickest way to
> produce a variety of test cases for an annotator. By changing a few lines we
> can report the invalid id.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
