Question

这是我的查询分析器定义：

<tokenizer class="solr.WhitespaceTokenizerFactory"/>
<filter class="solr.SynonymFilterFactory" synonyms="companysyns.txt" ignoreCase="true" expand="true"/>
<filter class="solr.WordDelimiterFilterFactory" generateWordParts="1" generateNumberParts="1" catenateWords="0" catenateNumbers="0" catenateAll="0" splitOnCaseChange="0"/>
<filter class="solr.LowerCaseFilterFactory"/>

在“companysyns.txt”中，我对典型的公司字词进行了一些扩展，例如：

inc, inc., incorporated

当我发送这样的查询时：

test:"some company inc"

我在solr debug中看到了这个意外的结果：

<str name="rawquerystring">test:"some company inc"</str>
<str name="querystring">test:"some company inc"</str>
<str name="parsedquery">
MultiPhraseQuery(test:"some company inc (inc incorporated)")
</str>
<str name="parsedquery_toString">test:"some company inc (inc incorporated)"</str>

与“Some Company，Inc。”不匹配。但是，如果我删除了WordDelimiterFilterFactory，那么对于同一个查询，我会看到：

<str name="rawquerystring">test:"some company inc"</str>
<str name="querystring">test:"some company inc"</str>
<str name="parsedquery">
MultiPhraseQuery(test:"some company (inc inc. incorporated)")
</str>
<str name="parsedquery_toString">test:"some company (inc inc. incorporated)"</str>

哪个匹配。

如果我保留WordDelimiterFilterFactory，请删除“inc。”来自同义词的条目（具有句点的条目），然后它也有效：

<str name="rawquerystring">test:"some company inc"</str>
<str name="querystring">test:"some company inc"</str>
<str name="parsedquery">
MultiPhraseQuery(test:"some company (inc incorporated)")
</str>
<str name="parsedquery_toString">test:"some company (inc incorporated)"</str>

知道为什么WordDelimiterFilterFactory弄乱同义词扩展查询？

谢谢！

Answer 1

WordDelimiterFilterFactory删除inc中的点。更改过滤器的顺序，它应该可以正常工作：

    <tokenizer class="solr.WhitespaceTokenizerFactory"/>
<filter class="solr.LowerCaseFilterFactory"/>
<filter class="solr.WordDelimiterFilterFactory" generateWordParts="1" generateNumberParts="1" catenateWords="0" catenateNumbers="0" catenateAll="0" splitOnCaseChange="0"/>
<filter class="solr.SynonymFilterFactory" synonyms="companysyns.txt" ignoreCase="true" expand="true"/>

在Solr中，SynonymFilterFactory扩展为分隔符+ WordDelimiterFilterFactory =＆gt;奇怪的结果查询

1 个答案: