lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Koji Sekiguchi <k...@r.email.ne.jp>
Subject Re: (Solr-UIMA) Doubt regarding integrating UIMA in to solr - Configuration.
Date Tue, 12 Jul 2011 00:39:33 GMT
I don't think you have wrong setting in UIMA, but you may have the request handler
named "/update/javabin" in solrconfig.xml is not correct?

koji
-- 
http://www.rondhuit.com/en/

(11/07/12 0:52), Sowmya V.B. wrote:
> Hi
>
> I just added the fields which are added to the index by one of the
> annotators to the index, in the fieldmappings section. I am not getting any
> compilation errors and still see the admin interface. However, when I index,
> i just get a SolrException,
>
> org.apache.solr.common.SolrException: Bad Request.
>
> On the server log, I don't see anything except for this:
> 127.0.0.1 - - [11/Jul/2011:17:44:04 +0200]  "POST
> /apache-solr-3.3.0/update/javabin?wt=javabin&version=2 HTTP/1.1" 400 1328
>
> Here is my UpdateRequestProcessorChain in sorlconfig.xml (Just changed the
> original path names for privacy's sake!)
>
> ***********
>      <!-- An update request processor chain for UIMA   -->
> <updateRequestProcessorChain name="uima">
>      <processor
> class="org.apache.solr.uima.processor.UIMAUpdateRequestProcessorFactory">
>        <lst name="uimaConfig">
>          <lst name="runtimeParameters">
>          <!-- Add here, the runtime parameters like Model File paths, list
> file paths etc. -->
>            <str
> name="tokenizerModelFileLocation">/Users/svajjala/Documents/EnglishTok.bin.gz</str>
>            <str
> name="taggerModelFileLocation">/Users/svajjala/Documents/tag.bin.gz</str>
>           <str name="AcceptedLanguages">english</str>
>              <str name="SetLanguageOnEmptyAcceptList">false</str>
>               <str
> name="GreenlineLists">/Users/svajjala/Documents/NewGreenline</str>
>         </lst>
>
>          <str
> name="analysisEngine">/Users/svajjala/Documents/ir4icallPipeline.xml</str>
>          <!-- Set to true if you want to continue indexing even if text
> processing fails.
>               Default is false. That is, Solr throws RuntimeException and
>               never indexed documents entirely in your session. -->
>             <bool name="ignoreErrors">false</bool>
>          <!-- This is optional. It is used for logging when text processing
> fails.
>               If logField is not specified, uniqueKey will be used as
> logField.
>          <str name="logField">id</str>
>          -->
>          <lst name="analyzeFields">
>            <bool name="merge">false</bool>
>            <arr name="fields">
>              <str>text</str>
>            </arr>
>          </lst>
>
>           <fieldMapping>
>
>              <type name = "org.apache.uima.annotators.vocablistClassfier">
>                  <map feature="Generic_Level5" field="Generic_Level5" />
>                  <map feature="Generic_Level6" field="Generic_Level6" />
>                  <map feature="Generic_Level7" field="Generic_Level7" />
>                  <map feature="Generic_Level8" field="Generic_Level8" />
>                  <map feature="Generic_Level9" field="Generic_Level9" />
>                  <map feature="Generic_Level10" field="Generic_Level10" />
>                  <map feature="Generic_Level11" field="Generic_Level11" />
>                  <map feature="Generic_Ratio" field="Generic_Ratio" />
>                  <map feature="Generic_TotalWordCount"
> field="Generic_TotalWordCount" />
>              </type>
>
>          </fieldMapping>
>        </lst>
>      </processor>
>      <processor class="solr.LogUpdateProcessorFactory" />
>      <processor class="solr.RunUpdateProcessorFactory" />
>    </updateRequestProcessorChain>
>
>      ************
> I don't understand where exactly can I see a more detailed log of why its
> not getting indexed.
>
> Sowmya.
>
> On Mon, Jul 11, 2011 at 5:26 PM, Koji Sekiguchi<koji@r.email.ne.jp>  wrote:
>
>> disclaimer: I'm not an expert of UIMA. I've just started using it when Solr
>> 3.1
>> integrated UIMA!
>>
>>
>>   Thanks for the clarification. Now, I get it.
>>> Should<fieldMapping>   section mention all the annotators, even if the
>>> annotators do not add any new fields?
>>>
>>
>>   For example, if I have a pipeline, starting from "parser", "tokenizer" and
>>> "tagger", all of them operate on a field called "text"..which is the<html>
>>> of the document. but all these annotators do not add any new fields to the
>>> index. Should I still write fieldmappings for these annotators inside
>>> SolrConfig.?
>>>
>>
>> I don't think so. Try it and if you got errors, come back to this list!
>>
>>
>> koji
>> --
>> http://www.rondhuit.com/en/
>>
>
>
>


Mime
View raw message