1

I'm trying to make symbols in searches not relevant.

If we see the following field value has a "perr" Token at indexation, as well as the value in query.

Index content

Why isn't a query like this returning anything? (extra fields remove for brevety)

<lst name="params">
<str name="q">per-ro</str>
<str name="qt">ri_dismax_mixed_es</str>
<str name="indent">true</str>
<str name="wt">xml</str>
<str name="debugQuery">true</str>
<str name="_">1651060917537</str>
</lst>
...
<result name="response" numFound="0" start="0">
</result>
<lst name="debug">
    <lst name="queryBoosting">
       <str name="q">per#ro</str>
       <null name="match"/>
    </lst>
<str name="rawquerystring">per-ro</str>
<str name="querystring">per-ro</str>
<str name="parsedquery">(+DisjunctionMaxQuery((... | exactLoose_es_PRODUCTO_DESCRIPCION_DETALLADA_es_str:"per (ro perr)"^8.0 | ... </str>
<str name="parsedquery_toString">+(... | exactLoose_es_PRODUCTO_DESCRIPCION_DETALLADA_es_str:"per (ro perr)"^8.0 | ...</str>
 <lst name="explain"/>
 <str name="QParser">DisMaxQParser</str>

QT ri_dismax_mixed_es (extra fields remove for brevety)

   <requestHandler class="solr.SearchHandler" name="ri_dismax_mixed_es" >
    <lst name="defaults">

      
      <str name="facet.method">enum</str>

      <str name="defType">dismax</str>
      <str name="echoParams">explicit</str>
      <int name="rows">20</int>

      
      <float name="tie">0.1</float>

      
      <str name="qf">... exactLoose_es_PRODUCTO_DESCRIPCION_DETALLADA_es_str^8 ...</str>

      
      <str name="pf">... exactLoose_es_PRODUCTO_DESCRIPCION_DETALLADA_es_str^8 ...</str>

      <int name="ps">100</int>

      
      <str name="fl">...</str>

      
      <str name="mm">1</str>
      <str name="facet">true</str>
      <int name="facet.limit">-1</int>
      <int name="facet.mincount">1</int>
      <str name="sort">score desc</str>
      <str name="q.alt">*:*</str>
    </lst>
    <lst name="appends">
        <str name="facet.field">hierarchy_es_front</str>
        ...
    </lst>
    <arr name="last-components">
      <str>elevator_es</str>
    </arr>
  </requestHandler>
Alejandro Armenteros
  • 61
  • 3
  • I've look around and it seems the problem comes with the positions of the resulting tokens. I've remade the WordDelimiterFilter to make all tokens from one word have the same position of the original token. so the parts are not dependant of each other so Per-ro would end up being "Per OR ro OR Perro" I think this was solved with the FlattenGraphFilter but i'm using solr 4.2 so i don't have access to that – Alejandro Armenteros Apr 28 '22 at 15:04

0 Answers0