I put in an issue on JIRA, UIMA-519
(https://issues.apache.org/jira/browse/UIMA-519).
-Matt
Thilo Goetz wrote:
Hi Matt,
the endless loop is certainly a bug. Feel free to open a Jira issue,
else I'll do it later today. Thanks for reporting this.
--Thilo
Matthew Campbell wrote:
Actually...I may have put my foot in my mouth on Friday concerning
DocumentAnnotation containing only one child annotation. The bug in the
AnnotationIndexImpl addChildredn() method may still be an issue though.
Sorry for the confusion.
-Matt
Matthew Campbell wrote:
Hey folks:
I was just wondering if anyone else was having any luck with the
tree() method in AnnotationIndexImpl. I've been walking through it
with the debugger and believe that the child subiterator in
addChildren() never gets pushed to the next annotation, so we get
stuck in the while loop. I changed the while loop to the code below
and it gets through fine:
while (it.hasNext()) {
annot = (AnnotationFS) it.next();
dtr = new AnnotationTreeNodeImpl();
dtr.set(annot);
node.addChild(dtr);
addChildren(dtr, subiterator(annot));
}
On a somewhat unrelated note, I'm basically trying to make a tree
of all the annotations in the document, so I'm passing in the
DocumentAnnotation into the tree function. Is there a reason the
DocumentAnnotation only has one child via a subiterator (which is a
SourceDocumentInformation annotation)? I assumed that since it's
range covers the entire document's length and is the first annotation
in the index, a subiterator on this annotation would iterate over all
of the strict unambiguous annotations in the document. Thanks in
advance for the help - I've only been working with UIMA for a few
weeks now so I'm still getting my head around everything, but it's a
great tool.
-Matt
