[infinispan-dev] migrating the JPA cache store to the new API

2013-09-06 Thread Mircea Markus
Hi Ray, Part of Infinispan 6.0 we revamped the CacheStore API. I've migrated the LevelDB store to the new API, do you feel like giving it a go with the Hibernate one? Cheers, -- Mircea Markus Infinispan lead (www.infinispan.org) ___ infinispan-

Re: [infinispan-dev] migrating the JPA cache store to the new API

2013-09-06 Thread Ray Tsang
I'd be happy to take a look. I also wrote an offheap memory store on my GitHub if you missed it. It can be easily refitted for general purpose mapdb store. On Sep 6, 2013, at 9:33, Mircea Markus wrote: > Hi Ray, > > Part of Infinispan 6.0 we revamped the CacheStore API. I've migrated the > Lev

Re: [infinispan-dev] migrating the JPA cache store to the new API

2013-09-11 Thread Ray Tsang
Mircea, I saw I could use CacheLoaderException for load()... But what should be used for write()? Should errors occurred during write() continue to throw CacheLoaderException? or, can it throw any RuntimeException? Thanks, On Fri, Sep 6, 2013 at 9:38 AM, Ray Tsang wrote: > I'd be happy to ta

Re: [infinispan-dev] migrating the JPA cache store to the new API

2013-09-12 Thread Ray Tsang
What's the best way to deal w/ storing MarshalledValue in a cache store such as JPA CacheStore? It seems like if MarshalledValue is stored, the same type is expected when loaded? Thanks, On Wed, Sep 11, 2013 at 5:24 PM, Ray Tsang wrote: > Mircea, > > I saw I could use CacheLoaderException for

Re: [infinispan-dev] migrating the JPA cache store to the new API

2013-09-18 Thread Galder ZamarreƱo
On Sep 13, 2013, at 12:05 AM, Ray Tsang wrote: > What's the best way to deal w/ storing MarshalledValue in a cache store such > as JPA CacheStore? > It seems like if MarshalledValue is stored, the same type is expected when > loaded? You get a MarshalledEntry, and calling getValueBytes() will