Re: [Spellcheck] NullPointerException on QueryComponent.mergeIds

2013-12-07 Thread Jean-Marc Desprez
James,
Sorry for the late response.
The shard.qt parameter actually solved my problem !

Thanks
Jean-Marc


2013/11/12 Dyer, James james.d...@ingramcontent.com

 Jean-Marc,

 This might not solve the particular problem you're having, but to get
 spellcheck to work properly in a distributed enviornment, be sure to set
 the shards.qt parameter to the name of your request handler.  See
 http://wiki.apache.org/solr/SpellCheckComponent#Distributed_Search_Support.

 James Dyer
 Ingram Content Group
 (615) 213-4311


 -Original Message-
 From: Jean-Marc Desprez [mailto:jm.desp...@gmail.com]
 Sent: Tuesday, November 12, 2013 8:57 AM
 To: solr-user@lucene.apache.org
 Subject: [Spellcheck] NullPointerException on QueryComponent.mergeIds

 Hello,

 I'm following this tutorial : http://wiki.apache.org/solr/SolrCloud with a
 SolR 4.5.0

 I'm at the very first step, only two replica and two shard and I have only
 *one* document in the index.

 When I try to get a spellcheck, I have this error :
 java.lang.NullPointerException
 at

 org.apache.solr.handler.component.QueryComponent.mergeIds(QueryComponent.java:843)

 I do not understand what I'm doing wrong and how I can get an error on
 mergeIds with only one document in the index (merge this doc with ... ??)

 Some technical details :
 URL :

 http://127.0.0.1:8983/solr/bench/select?shards.qt=ri_spell_fr_FRq=sistemdistrib=true
 If I set distrib to false, no error.

 My uniqueKey is indexed and stored :

 field name=ref type=string indexed=true stored=true
 multiValued=false /
 uniqueKeyref/uniqueKey


 My conf :
 requestHandler name=ri_spell_fr_FR class=solr.SearchHandler
 lazy=true
   lst name=defaults
 bool name=spellchecktrue/bool
 str  name=spellcheck.onlyMorePopulartrue/str
 str  name=spellcheck.extendedResultstrue/str
 str  name=spellcheck.collateExtendedResultstrue/str
 str  name=spellcheck.maxCollationTries3/str
 str  name=spellcheck.collatetrue/str
 str  name=spellcheck.count5/str
 str  name=spellcheck.dictionaryri_spell_fr_FR/str
 str  name=spellcheck.buildfalse/str
   /lst

   arr name=components
 strspellcheck_fr_FR/str
   /arr
 /requestHandler

 searchComponent name=spellcheck_fr_FR class=solr.SpellCheckComponent
   str name=queryAnalyzerFieldTypesuggest_fr_FR/str

   lst name=spellchecker
 str name=nameri_spell_fr_FR/str
 str name=fieldspell_fr_FR/str
 str name=spellcheckIndexDir./spellchecker_fr_FR/str
 str

 name=distanceMeasureorg.apache.lucene.search.spell.JaroWinklerDistance/str
   /lst

   ...

 /searchComponent

 With this URL :
 http://127.0.0.1:8983/solr/bench/select?qt=ri_spell_fr_FRq=sistem

 I have no error but the response is empty :
 responselst name=responseHeaderint name=status0/intint
 name=QTime1/int/lst/response


 Thanks
 Jean-Marc



RE: [Spellcheck] NullPointerException on QueryComponent.mergeIds

2013-11-12 Thread Dyer, James
Jean-Marc,

This might not solve the particular problem you're having, but to get 
spellcheck to work properly in a distributed enviornment, be sure to set the 
shards.qt parameter to the name of your request handler.  See 
http://wiki.apache.org/solr/SpellCheckComponent#Distributed_Search_Support .

James Dyer
Ingram Content Group
(615) 213-4311


-Original Message-
From: Jean-Marc Desprez [mailto:jm.desp...@gmail.com] 
Sent: Tuesday, November 12, 2013 8:57 AM
To: solr-user@lucene.apache.org
Subject: [Spellcheck] NullPointerException on QueryComponent.mergeIds

Hello,

I'm following this tutorial : http://wiki.apache.org/solr/SolrCloud with a
SolR 4.5.0

I'm at the very first step, only two replica and two shard and I have only
*one* document in the index.

When I try to get a spellcheck, I have this error :
java.lang.NullPointerException
at
org.apache.solr.handler.component.QueryComponent.mergeIds(QueryComponent.java:843)

I do not understand what I'm doing wrong and how I can get an error on
mergeIds with only one document in the index (merge this doc with ... ??)

Some technical details :
URL :
http://127.0.0.1:8983/solr/bench/select?shards.qt=ri_spell_fr_FRq=sistemdistrib=true
If I set distrib to false, no error.

My uniqueKey is indexed and stored :

field name=ref type=string indexed=true stored=true
multiValued=false /
uniqueKeyref/uniqueKey


My conf :
requestHandler name=ri_spell_fr_FR class=solr.SearchHandler
lazy=true
  lst name=defaults
bool name=spellchecktrue/bool
str  name=spellcheck.onlyMorePopulartrue/str
str  name=spellcheck.extendedResultstrue/str
str  name=spellcheck.collateExtendedResultstrue/str
str  name=spellcheck.maxCollationTries3/str
str  name=spellcheck.collatetrue/str
str  name=spellcheck.count5/str
str  name=spellcheck.dictionaryri_spell_fr_FR/str
str  name=spellcheck.buildfalse/str
  /lst

  arr name=components
strspellcheck_fr_FR/str
  /arr
/requestHandler

searchComponent name=spellcheck_fr_FR class=solr.SpellCheckComponent
  str name=queryAnalyzerFieldTypesuggest_fr_FR/str

  lst name=spellchecker
str name=nameri_spell_fr_FR/str
str name=fieldspell_fr_FR/str
str name=spellcheckIndexDir./spellchecker_fr_FR/str
str
name=distanceMeasureorg.apache.lucene.search.spell.JaroWinklerDistance/str
  /lst

  ...

/searchComponent

With this URL :
http://127.0.0.1:8983/solr/bench/select?qt=ri_spell_fr_FRq=sistem

I have no error but the response is empty :
responselst name=responseHeaderint name=status0/intint
name=QTime1/int/lst/response


Thanks
Jean-Marc