Word does not get analysed properly using StemmerOverrideFilterFactory and SnowballPorterFilterFactory for Dutch language

Question

Solr: 3.5

Hi,

I created a dutch field type according to the following fieldType definition:

    <fieldType name="text_nl" class="solr.TextField" positionIncrementGap="100">
        <analyzer>
            <tokenizer class="solr.StandardTokenizerFactory"/>
            <filter class="solr.WordDelimiterFilterFactory" generateWordParts="1" generateNumberParts="1"
                    catenateWords="1" catenateNumbers="1" catenateAll="0" preserveOriginal="1"/>
            <filter class="solr.LowerCaseFilterFactory"/>
            <filter class="solr.StemmerOverrideFilterFactory" words="lang/stemdict_nl.txt"  ignoreCase="true"/>
            <filter class="solr.SnowballPorterFilterFactory" language="Kp" words="lang/stemdict_nl.txt"/>
        </analyzer>
    </fieldType>

stemdict_nl.txt is using 45710 word rules according to the http://snowball.tartarus.org/algorithms/kraaij_pohlmann/stemmer.html algorithm.

Most of the search queries seem to be working fine and I am getting mostly correct suggestions.

However there is an issue when I search on 'etiketje'. According to my rules:

etiket                        etiket
etiketten                     etiket
etiketteren                   etiketteer
etikettering                  etiketteer
etiketje                      etiket

It should fallback on 'etiket'. Except however it fallsback on 'etik'. When I analyse my field, SOLR returns:

etiketje
etiketje
etiketje
etiketje
etik

I would love for SOLR to analyse 'Etiketje' as:

etiketje
etiket

Hopefully someone here can point me in the right direction.

Answer 1

Try changing your definition to the exact syntax as shown on the wiki i.e. change

<filter class="solr.StemmerOverrideFilterFactory" 
        words="lang/stemdict_nl.txt"  ignoreCase="true"/>
<filter class="solr.SnowballPorterFilterFactory" 
        language="Kp" words="lang/stemdict_nl.txt"/>

to

<filter class="solr.StemmerOverrideFilterFactory" 
        dictionary="lang/stemdict_nl.txt"/>
<filter class="solr.SnowballPorterFilterFactory" 
        language="Kp"/>

You do not need ignoreCase=true on the StemmerOverrideFilter since you are using LowerCaseFilter before that filter anyway.