[hibernate-dev] R: [HSEARCH] @DateBridge and @Field(index=Index.UN_TOKENIZED)

Sanne Grinovero sanne.grinovero at gmail.com
Wed May 26 08:03:37 EDT 2010


true, I don't support the idea to enforce this, but it would be nice
to provide the correct default, or maybe log a warning.
each bridge could have a

boolean recommendAnalyzersOn();

?

2010/5/26 Emmanuel Bernard <emmanuel at hibernate.org>:
> I have been thinking some more and it looks like its a more general problem and not as easy a solution as it seems.
> If you think about it many bridges should not be analyzed (date, URL, class, numbers, enum etc: most of the built-in ones actually).
>
> That being said a smart person might want to convert a class into its fqcn and then analyze that with a specific analyzer. Same for all the others. So making decisions is probably not a good idea after all :)
>
>
> On 20 mai 2010, at 20:04, Sanne Grinovero wrote:
>
>> Makes sense, maybe we could default to untokenized? Same idea should apply to the soon-to-be-added NumericFields.
>>
>>
>>> Il giorno 20/mag/2010 14:58, "Emmanuel Bernard" <emmanuel at hibernate.org> ha scritto:
>>>
>>> Does it ever make sense to use @DateBridge without @Field(index=Index.UN_TOKENIZED) or NO_NORM
>>>
>>> I got caught up with doing
>>>
>>> @DateBridge(resolution=DAY) @Field
>>> public String getDate() { return date; }
>>>
>>> and having unexpected issues.
>>>
>>> If it never makes sense we could raise an exception when that happens, or simply force the index strategy to UN_TOKENIZE.
>>>
>>> WDYT?
>>>
>>>
>>> _______________________________________________
>>> hibernate-dev mailing list
>>> hibernate-dev at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/hibernate-dev
>>
>
> _______________________________________________
> hibernate-dev mailing list
> hibernate-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/hibernate-dev
>



More information about the hibernate-dev mailing list