We wanted follow-up on our previous question on incorporating hyphen in our searches. We've implemented a custom analyzer as described in your example and added it to classpath as a jar file. In our implementation we simply return the WhitespaceAnalyzer.
We expect it to treat hyphen not as whitespace and interpret hyphenated words as a whole to be an indexed term. However we don't see any difference in the results we get.
Is there a way to check on Stardog which Lucene analyzer is actively being used?
Also, we'd appreciate any resources on writing a custom analyzer.
Hi,
Are you able to share your AnalyzerFactory code as well as the file you are including in META-INF/services? Where in the classpath are you adding this jar?
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.