I've done a deeper analysis, here is what I am thinking about
query.SearchTimeoutException why is it not a subclass of SearchException?
query.SearchTimeoutException => query.engine.SearchTimeoutException (API) ? (I
don't see why the persistence layer could not let this exception go as is).
quer
ok, you don't have conflicts with Hardy either:
https://github.com/hibernate/hibernate-search/pull/50/files
2011/3/2 Emmanuel Bernard :
> I am thinking about moving:
> EntityInfo
> AbstractLoader
> DocumentExtractor
> Loader
> LoaderHelper
> MultiClassesQueryLoader
> ObjectLoaderHelper
> Projectio
I have pending changes for DocumentExtractor, DocumentBuilderHelper,
DocumentBuilderIndexedEntity
(if I consider only the engine package).
I can merge them later if this is blocking you, won't be much work.
Cheers,
Sanne
2011/3/2 Emmanuel Bernard :
> Which part specifically?
> I can't really fini
I am thinking about moving:
EntityInfo
AbstractLoader
DocumentExtractor
Loader
LoaderHelper
MultiClassesQueryLoader
ObjectLoaderHelper
ProjectionLoader
QueryLoader
On 2 mars 2011, at 23:19, Emmanuel Bernard wrote:
> It seems that the engine package content is entirely related to query and
> sh
Which part specifically?
I can't really finish HSEARCH-687 without it.
On 2 mars 2011, at 23:24, Sanne Grinovero wrote:
> fine for me, but we could postpone such a change?
> I believe both Hardy and me have many pending changes to integrate in
> the same area.
>
> 2011/3/2 Emmanuel Bernard :
>
fine for me, but we could postpone such a change?
I believe both Hardy and me have many pending changes to integrate in
the same area.
2011/3/2 Emmanuel Bernard :
> It seems that the engine package content is entirely related to query and
> should move somewhere under o.h.search.query.
> Any opin
It seems that the engine package content is entirely related to query and
should move somewhere under o.h.search.query.
Any opinion or objection? Most are non API non SPI classes anyways.
___
hibernate-dev mailing list
hibernate-dev@lists.jboss.org
https
On Wed, 02 Mar 2011 19:48:33 +0100, Sanne Grinovero
wrote
> so it's quite possible that this approach takes hostage of a lot of
> memory.
I guess that was our fear from the beginning.
> So to cope with this, I've now built a custom fieldCache and a
> specialized Collector, which extract the d
Hello,
I was playing with a "standard" FieldCache, having as goal to
accelerate the extraction of what we need mostly: classtype and
primary Id.
Focusing on classtype only, it seems that we have to be careful with
FieldCaches:
1 - Strings are extracted as different instances, copying over and
over
Hey, guys,
it is possible that foreign key references non-primary key, but some
candidate key. For example unique column.
But seems it was not taken into account in ForeignKey class.
And that is why we have a NPE in the case.
Could somebody review the patch which fixes the problem?
(see https:/
10 matches
Mail list logo