[
https://issues.apache.org/jira/browse/UIMA-190?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Thilo Goetz closed UIMA-190.
----------------------------
Resolution: Fixed
First of all it wasn't a type priority test, but a regular CAS index iterator
test. I didn't stop to figure out exactly why the test case failed, but just
replaced the custom setup by a regular, though empty, annotator. This made the
test case fail for all JVMs, thankfully all in the same place. After
meditating on this for a while, I finally figured out that the failing part of
the test case was invalid and had only accidentally worked before. When doing
a reverse iteration over a sorted index, the sequence of FSs is not guaranteed
to be the reverse of a forward iteration. The only thing we guarantee is that
for reverse iteration, the FSs come out in reverse order with respect to the
sort order, not with respect to forward iteration. This is because there may
be FSs in the sorted index that are equal to each other with respect to the
sort order. An iterator will return those in arbitrary order.
I have commented out the test case, but left the code in.
> Type priority test case failing with IBM JDK 1.5
> ------------------------------------------------
>
> Key: UIMA-190
> URL: https://issues.apache.org/jira/browse/UIMA-190
> Project: UIMA
> Issue Type: Bug
> Components: Build, Packaging and Test
> Affects Versions: 2.1
> Reporter: Thilo Goetz
> Assigned To: Thilo Goetz
> Priority: Minor
> Fix For: 2.1
>
>
> The reason the test case fails seems to be due to the incorrect setup of the
> test case. The way the test is set up, the type priority sort order is
> initialized more than once, resulting in erratic behavior.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
