Solr has a nice new feature for the KeywordMarkerFilterFactory in field
types in schema.xml – the ability to specify tokens to mark (e.g., to
suppress stemming) using a regular expression pattern specified in the new
“pattern” attribute. But, you wouldn’t know it from reading the Solr
CHANGES.txt.
The issue here is that this was a Lucene change that bubbles up through
Solr. But how many Solr users are going to read the Lucene CHANGES.txt
closely enough to realize that UCENE-4822 ("Add PatternKeywordTokenFilter to
marks keywords based on regular expressions") affects the "Solr"
KeywordMarkerFilterFactory (which is technically in "Lucene")?
It seems to me that changes to Lucene that have a Solr schema impact should
be noted in the Solr CHANGES.txt, although, admittedly it is a gray area.
Still, there should be a way for Solr users to find out about new features
they can use without diving deep into Lucene.
-- Jack Krupansky
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
