Hoss Man commented on SOLR-11978:

attached a patch that i plan to commit soon.  Adds {{text_gen_sort}} (using 
same analyzer as {{text_general}} ) to both configsets.  in {{_default}} a 
{{<dynamicField name="*_txt_sort"}} is also added, while in 
{{sample_techproducts_configs}} this type is instead used explicitly by the 
{{manu, title, subject, author}} fields.

(FWIW: I waned to use it on techproduct's "name" field as well, but that broke 
all sorts of things with how the spellchecker is configured in that configset, 
because the ConjunctionSpellChecker requires that all of the SpellCheckers it 
wraps use query analyzers that are {{==}} ... so IIUC effectively they must 
have the exact same {{<fieldType/>}} ... identical analyzer configs isn't 
actaully good enough?)

> include SortableTextField in _default and sample_techproducts configsets
> ------------------------------------------------------------------------
>                 Key: SOLR-11978
>                 URL: https://issues.apache.org/jira/browse/SOLR-11978
>             Project: Solr
>          Issue Type: Sub-task
>      Security Level: Public(Default Security Level. Issues are Public) 
>            Reporter: Hoss Man
>            Assignee: Hoss Man
>            Priority: Major
>         Attachments: SOLR-11978.patch
> since SortableTextField defaults to docValues="true" it has additional on 
> disk overhead compared to TextField that means I don't think we should 
> completley replace all suggested uses of TextField at this point – but it 
> would still be good to include it in our configsets similar to the way we 
> include declarations for a variety of text analysis options.
> I also think several "explicit" fields in the techproducts schema would 
> benefit from using this.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org

Reply via email to