[
https://hibernate.onjira.com/browse/HSEARCH-471?page=com.atlassian.jira.p...
]
Sanne Grinovero commented on HSEARCH-471:
-----------------------------------------
About _IndexingActionTypeonDelete_ use cases, some of the people requesting this issue
wanted it primarily for performance reasons. One request is to be able to have the
indexing engine not touch the index during peak hours (index will be kept optimized, and
take better advantage of FieldCache and filter caches, reusing the same IndexReader);
today one can change the ReaderProvider to not refresh itself, being able to disable index
changes is better. I had this configuration myself on some systems, the reindex all over
the night to recover consistency; Having this is better than avoiding to register the
listener as you can control it per entity-type.
{quote}
Another problem is that because we exclude non indexed fields and don't trigger a Work
event in this case, we need to force the user to index all the fields that will possibly
involve the soft delete logic even if the field is not meant to be indexed int he first
place.
An alternative is to list the fields that ought to force a Work trigger. Ideas?
{quote}
This is how I think it could work:
OperationType.REMOVE -> We remove, whatever our information on dirtyness: the entity
must be "hidden".
OperationType.SKIP -> We avoid creating any work, whatever our information on
dirtyness.
OperationType.PASS_THROUGH -> We look at dirtyness (business as usual).
Basically we need to invoke the interceptor before looking at the dirtyness.
Ability to selectively index an entity based on its state
---------------------------------------------------------
Key: HSEARCH-471
URL:
https://hibernate.onjira.com/browse/HSEARCH-471
Project: Hibernate Search
Issue Type: New Feature
Components: engine, mapping
Affects Versions: 3.1.1.GA
Reporter: Dobes Vandermeer
In our system we have entities that are searched but not all of them are available for
search - some of them are flagged as "removed". It would improve the efficiency
of our search subsystem if we could implement a kind of "filter" that blocked
these entities from being added to the search index, since we wouldn't have to make
that a search term and our indexes would be somewhat smaller.
--
This message is automatically generated by JIRA.
For more information on JIRA, see:
http://www.atlassian.com/software/jira