2014-11-06 10:49 GMT+01:00 Chetan Mehrotra <[email protected]>:
> On Thu, Nov 6, 2014 at 3:08 PM, Tommaso Teofili > <[email protected]> wrote: > > yes, that's right, sorry I should've expanded this a bit: I personally > > think we should expose all/most of Lucene configuration points, including > > Codecs, so I'd like to take an approach that works for all the > > configuration bits. > > Agreed. But do you think people would be providing new codec > implementation or composing new codec based on ones already provided > by Lucene. Looking up core Lucene codes and posting format should > work. Problem would be if you need to make your custom codec impl > visible to Lucene > what I see as more probable is people demanding the possibility of composing existing Lucene codecs, which is what we have done with OakCodec (using default but with compression disabled for stored fields). > > So probably we can started with #C for now for exposing analyzers > customisation which seems to cover majority of requirements. Later we > can tackle the Codec case > ok for me, I'd also be happy to help as that would probably imply exposing Lucene classes in OSGi too, right? Regards, Tommaso > > Chetan Mehrotra >
