I'd prefer us to do this in a minor version, i.e. 5.6.
As we expose Query to users, the removal of these methods may cause
compilation errors on their side. Also the deprecation of Filter may
cause many deprecations in client code.
Both seem not appropriate for a micro release to me.
I don't think we need to rush it. Compared to previous times, we are
keeping up with new Lucene versions closely these days, so doing it
during the envisioned 5.6 time frame seems fast enough to me. Or, if
you really want to get it out sooner, we can do a quick 5.6 off
master, I suppose prior to those first ES commits and release the
latter as 5.7.
2015-12-16 14:22 GMT+01:00 Sanne Grinovero <sanne(a)hibernate.org>:
Apache Lucene 5.4.0 is released as stable, and while it includes
several benefits - not least a performance fix from myself which I'd
be keen to take advantage of - there are no significant changes
visible to end users, expect I think this one is worth a warning:
LUCENE-6590: Query.setBoost(), Query.getBoost() and Query.clone() are
gone. In order to apply boosts, you now need to wrap queries in a
Although, it doesn't affect any of our code not examples so a release
note might be acceptable.
There are several other interesting performance improvements.
So I'm tempted to upgrade to this version in the next maintenance
(micro) release of Hibernate Search 5.5.
What do you all think about that?
hibernate-dev mailing list