[jira] [Created] (SOLR-4025) queryAnalyzerFieldType for spellcheck is never validated, makes example configs broken

["Hoss Man (JIRA)" <ji...@apache.org>]

Hoss Man created SOLR-4025:
------------------------------

             Summary: queryAnalyzerFieldType for spellcheck is never validated, makes example configs broken
                 Key: SOLR-4025
                 URL: https://issues.apache.org/jira/browse/SOLR-4025
             Project: Solr
          Issue Type: Bug
            Reporter: Hoss Man


Discovered this while trying to answer a quick question for "nepenthe" on the IRC channel...

* spellcheck component looks for an init param named "queryAnalyzerFieldType" butsilently ignores it and uses WhitespaceAnalyzer if it refers to a fieldtype that doesn't exist.
* as a result of silently ignoring this w/o failure, no one seems to have ever noticed that the example configs refer to a field type that was removed...
{noformat}
    <str name="queryAnalyzerFieldType">textSpell</str>
{noformat}
* as a result, the "default" example spellchecker dictionary (using DirectSolrSpellChecker) doesn't give any recomendations for words with upercase letters (because it defaults to WhitespaceAnalyzer but the "name" field used for hte dictionary uses LowercaseFilter)

For example, this gives no suggestions...

http://localhost:8983/solr/spell?q=Delll+Ultrashar&spellcheck=true&spellcheck.extendedResults=true&spellcheck.dictionary=default&spellcheck.build=true

But this does...

http://localhost:8983/solr/spell?q=delll+ultrashar&spellcheck=true&spellcheck.extendedResults=true&spellcheck.dictionary=default&spellcheck.build=true

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

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