[hibernate-issues] [Hibernate-JIRA] Issue Comment Edited: (HSEARCH-867) input stream support

adam (JIRA) noreply at atlassian.com
Wed Aug 24 10:40:02 EDT 2011


    [ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-867?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=43338#comment-43338 ] 

adam edited comment on HSEARCH-867 at 8/24/11 9:39 AM:
-------------------------------------------------------

Sanne,
  as I read the documentation originally, it uses the stringValue() method on Fieldable to get the string, so it just pushes the toString down the line so-to-speak.  I'll look into what Fieldable can do and suggest changes to the documentation if you'd like.  

BTW, the first one is to modify the documentation to either:
# more clearly link the source 
# provide the complete object definition of the LazyField



      was (Author: abrin):
    Sanne,
  as I read the documentation originally, it uses the stringValue() method on Fieldable to get the string, so it just pushes the toString down the line so-to-speak.  I'll look into what Fieldable can do and suggest changes to the documentation if you'd like.  

BTW, the first one is to modify the documentation to define LazyField more directly to explicitly implement Fieldable.

thanks,

adam
  
> input stream support
> --------------------
>
>                 Key: HSEARCH-867
>                 URL: http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-867
>             Project: Hibernate Search
>          Issue Type: Improvement
>          Components: analyzer, integration
>    Affects Versions: 3.4.0.Final
>            Reporter: adam
>            Priority: Minor
>
> The current hibernate search functionality is not optimized for dealing with large text contents.  Two use cases:
> 1. indexing an external PDF that's 100MB where an @Field is set on a getter
> 2. indexing a @Lob field
> in both cases, the method must return a string, or a base class, which might mean that you have an InputStream that's 50MB, which gets concatenated into a string, and then passed to an analyzer bundled into a Reader object.  I'm unclear what HibernateSearch is doing when the getter for the @Field annotation is called, but it would be ideal if it could use a reader instead of a string 

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        


More information about the hibernate-issues mailing list