[
https://issues.apache.org/jira/browse/CTAKES-341?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15029027#comment-15029027
]
britt fitch commented on CTAKES-341:
------------------------------------
output before update:
{code}
26 Nov 2015 10:52:38 INFO ContextDependentTokenizerAnnotator - process(JCas)
4.5 3.5
4.5
4.7
{code}
output after update:
{code}
26 Nov 2015 10:53:42 INFO ContextDependentTokenizerAnnotator - process(JCas)
4.5
3.5
4.7
{code}
Checked in:
* FractionFSM - update as mentioned above
* AggregateAE - update for test case
* TestContextDependentTokenizerAnnotator - new test case
> FractionFSM annotates incorrect span
> ------------------------------------
>
> Key: CTAKES-341
> URL: https://issues.apache.org/jira/browse/CTAKES-341
> Project: cTAKES
> Issue Type: Bug
> Components: ctakes-context-tokenizer
> Affects Versions: 3.2.0
> Reporter: britt fitch
> Assignee: britt fitch
> Fix For: 3.2.3
>
>
> It appears that when a decimal is followed by a range that the FractionFSM
> incorrectly annotates the FractionToken
> given:
> {code}
> FOO 4.5 3.5-4.7
> {code}
> produces the following FractionTokens:
> * "4.5"
> * "4.5 3.5"
> * "4.7"
> after fsm.reset we need to also add the following in order to move the start
> position and allow sequential END states to be handled correctly:
> {code} tokenStartMap.put(fsm, tokenStartIndex); {code}
> i will create a test case and verify this solution before committing it back
> to trunk.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
