|
David Sebastiao
|
Monday 07 September 2009 10:17:10 am
Hi,
I have a very strange issue with ezfind 2.0. Maybe it's holliday's effect but my instance makes difference between uppecase and lowercase. So i don't have same result with 'foo' and 'FoO'
I'm very disapointed. I've checked if spellcheck was on true. My request
{def $search=fetch( ezfind, search,
hash( query, $search_text,
'class_id', array( '37' ),
'offset',$offset,
'limit',$page_limit,
'sort_by', hash( 'name', 'asc' ),
'spell_check', array(true(),'default')
) )}
My ezfind.ini seems ok too
[SpellCheck]
SpellCheck=enabled
DefaultDictionary=default
As described here : <i>http://www.mail-archive.com/[email protected]/msg00206.html</i>
i've changed solrconfig.xml
- <str name="buildOnCommit">true</str>
+ <str name="buildOnOptimize">true</str>
Restart Solr Service, and run updatesearchindex but still no effect. Thanks a lot.
|
|
David Sebastiao
|
Wednesday 09 September 2009 3:39:43 am
Ok i found a solution.
I had sorting bug with ezstring attribute soaccording to http://issues.ez.no/IssueView.php?Id=15164 I added DatatypeMap[ezstring]=text in [SolrFieldMapSettings] section in my ezfind.ini.
# Datatype to field type map.
#
# Example:
DatatypeMap[ezstring]=string
DatatypeMap[eztext]=text
DatatypeMap[ezboolean]=boolean
DatatypeMap[ezdate]=date
DatatypeMap[ezdatetime]=date
DatatypeMap[ezfloat]=sfloat
DatatypeMap[ezinteger]=sint
DatatypeMap[ezprice]=sfloat
DatatypeMap[eztime]=date
Thanks to Carlo's and Max's post (http://ez.no/fr/developer/forum/extensions/ez_find/wildcards_and_phrases_and_spanish_order_of_the_results/re_wildcards_and_phrases_and_spanish_order_o__1)
So I resolved my issue changing DatatypeMap[ezstring]=text and creating a custom datatype.
|