How do you filter for text in DNG that contains an underscore "_"?
Accepted answer
In fact starting version 6.0.4, searches containing an underscore should work again.
But you may need to run a reindex first as outlined below.
See section 'Searches that contain an underscore character might not return the correct results' in article 'Workarounds and Limitations: Known issues in DNG 6.0.4':
https://jazz.net/library/article/89437#38
There is a workaround described in this article that resolves the problem:
==============
Workaround
Starting in version 6.0.4, searches that contain the underscore character should work properly. However, if you upgrade or migrate data from a release earlier than 6.0.4, an administrator must reindex the text indexes. This operation can take a significant amount of time depending on the repository size.
For details about reindexing only the text store indexes, see Repository tools command to regenerate indexes. For a quick summary of reindexing only the text store indexes, see the following examples.
Windows
Open a command prompt and enter the following command:
cd C:\Program Files\IBM\JazzTeamServer\server\
repotools-rm.bat -reindex teamserver.properties=conf\rm\teamserver.properties scope=search
Note: Remember to include scope=search; otherwise, the Jena triplestore will be indexed and the process will take much longer.
UNIX
Open a command prompt and enter the following command:
cd opt/IBM/JazzTeamServer/server/
./repotools-jts.sh -reindex teamserver.properties=conf/jts/teamserver.properties scope=search
Note: Remember to include scope=search; otherwise, the Jena triplestore will be indexed and the process will take much longer.
==============
One other answer
The full text search feature in CLM utilizes Apache Lucene, which does not store underscore (_) as a "term" in its database in most configurations. See the "Tokenizing" section of the below article for more details.
https://jazz.net/library/article/824
Text is also broken up at non-letter non-numerical characters (e.g. ‘.’, ‘?’, ‘!’ and /’), capital characters (e.g. CamelCase is broken into Camel Case)