[hibernate-issues] [Hibernate-JIRA] Updated: (HSEARCH-216) Delete by Term when possible

[Sanne Grinovero (JIRA)]

     [ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-216?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sanne Grinovero updated HSEARCH-216:
------------------------------------

    Description: 
discussed on mail list, in short:
when we know only one entity type is written to a specific index we could avoid the delete on the dual identifier(entityname, databaseID) which needs an IndexReader, and a query, and iterating on the results;
a deletion by Term using only database identifier would be enough and should be more efficient.

Additionally if we detect it's possible to do all the work with an IndexWriter we could skip the IndexReader opening.


  was:
discussed on mail list, in short:
when we know only one entity type is written to a specific index we could avoid the delete on the dual identifier(entityname, databaseID) which needs an IndexReader, and a query, and iterating on the results;
a deletion by term using only database identifier would be enough and should be more efficient.

Additionally if we detect it's possible to do all the work with an IndexWriter we could skip the IndexReader opening.


        Summary: Delete by Term when possible  (was: Delete by TermQuery when possible)

> Delete by Term when possible
> ----------------------------
>
>                 Key: HSEARCH-216
>                 URL: http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-216
>             Project: Hibernate Search
>          Issue Type: Improvement
>          Components: engine
>            Reporter: Sanne Grinovero
>
> discussed on mail list, in short:
> when we know only one entity type is written to a specific index we could avoid the delete on the dual identifier(entityname, databaseID) which needs an IndexReader, and a query, and iterating on the results;
> a deletion by Term using only database identifier would be enough and should be more efficient.
> Additionally if we detect it's possible to do all the work with an IndexWriter we could skip the IndexReader opening.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira