synonym.txt文件中使用的特殊字符(例如*)在solr中不起作用

Question

对于版本为4.10.4的solr，我已经创建了文件synonyms.txt并应用了synonymFilterFactory，如下所示：

<fieldType name="text_general" class="solr.TextField">  
     <analyzer>     
        <tokenizer class="solr.StandardTokenizerFactory"/>
        <filter class="solr.LowerCaseFilterFactory"/>
        <filter class="solr.WordDelimiterFilterFactory" generateWordParts="1" generateNumberParts="0" splitOnCaseChange="0" splitOnNumerics="0" catenateWords="1" catenateNumbers="0" catenateAll="0" preserveOriginal="1" stemEnglishPossessive="0"/>
     </analyzer>
     <analyzer type="query">
        <tokenizer class="solr.StandardTokenizerFactory"/>
        <filter class="solr.SynonymFilterFactory" synonyms="synonyms.txt" ignoreCase="true" expand="true"/>
        <filter class="solr.LowerCaseFilterFactory"/>
     </analyzer>
</fieldType>

synonyms.txt的内容如下：

holland* => holland

holland, netherland, netherlands, niederlande

我在应用程序中有某些条件可以生成术语，例如：holland*

在这种情况下，我希望显示的结果与我输入术语holland、netherland、netherlands、niederlande时得到的结果相同。

但目前，对于术语holland*，它不能给出匹配的结果。荷兰的结果*包含与“荷兰”或“荷兰”相同的结果，但这些结果处于底部，所以我们可以提高这些结果吗？

有没有人知道，我怎么才能做到呢？

以下是更多细节：

对于holland，我得到了一些结果，当我调试查询时，它显示为

"debug": {
    "rawquerystring": "holland",
    "querystring": "holland",
    "parsedquery": "(name:holland name:netherland name:netherlands name:niederlande)/no_coord",
    "parsedquery_toString": "name:holland name:netherland name:netherlands name:niederlande",
    "explain": {
      "country-NLD-de": "\n7.42217 = (MATCH) sum of:\n  7.42217 = (MATCH) weight(name:niederlande in 1775593) [DefaultSimilarity], result of:\n    7.42217 = score(doc=1775593,freq=1.0), product of:\n      0.5213204 = queryWeight, product of:\n        14.237252 = idf(docFreq=14, maxDocs=8413113)\n        0.036616646 = queryNorm\n      14.237252 = fieldWeight in 1775593, product of:\n        1.0 = tf(freq=1.0), with freq of:\n          1.0 = termFreq=1.0\n        14.237252 = idf(docFreq=14, maxDocs=8413113)\n        1.0 = fieldNorm(doc=1775593)\n",
      "country-NLD-en": "\n7.3550315 = (MATCH) sum of:\n  7.3550315 = (MATCH) weight(name:netherlands in 230095) [DefaultSimilarity], result of:\n    7.3550315 = score(doc=230095,freq=1.0), product of:\n      0.5189572 = queryWeight, product of:\n        14.172713 = idf(docFreq=15, maxDocs=8413113)\n        0.036616646 = queryNorm\n      14.172713 = fieldWeight in 230095, product of:\n        1.0 = tf(freq=1.0), with freq of:\n          1.0 = termFreq=1.0\n        14.172713 = idf(docFreq=15, maxDocs=8413113)\n        1.0 = fieldNorm(doc=230095)\n",
      "place-49218-de": "\n5.0385056 = (MATCH) sum of:\n  5.0385056 = (MATCH) weight(name:holland in 385574) [DefaultSimilarity], result of:\n    5.0385056 = score(doc=385574,freq=1.0), product of:\n      0.4295267 = queryWeight, product of:\n        11.730367 = idf(docFreq=183, maxDocs=8413113)\n        0.036616646 = queryNorm\n      11.730367 = fieldWeight in 385574, product of:\n        1.0 = tf(freq=1.0), with freq of:\n          1.0 = termFreq=1.0\n        11.730367 = idf(docFreq=183, maxDocs=8413113)\n        1.0 = fieldNorm(doc=385574)\n",

在荷兰*的情况下，结果包含一些来自荷兰的记录，但调试部分如下：

"debug": {
    "rawquerystring": "holland*",
    "querystring": "holland*",
    "parsedquery": "name:holland*",
    "parsedquery_toString": "name:holland*",
    "explain": {
      "place-51432-de": "\n1.0 = (MATCH) ConstantScore(name:holland name:hollandarod name:hollande name:hollander name:hollanderei name:hollandia name:hollandischer name:hollands name:hollandsbjerg name:hollandsch name:hollandsche name:hollandscheveld name:hollandsdiep name:hollandskamp name:hollandske), product of:\n  1.0 = boost\n  1.0 = queryNorm\n",
      "place-49196-de": "\n1.0 = (MATCH) ConstantScore(name:holland name:hollandarod name:hollande name:hollander name:hollanderei name:hollandia name:hollandischer name:hollands name:hollandsbjerg name:hollandsch name:hollandsche name:hollandscheveld name:hollandsdiep name:hollandskamp name:hollandske), product of:\n  1.0 = boost\n  1.0 = queryNorm\n",
      "place-49207-de": "\n1.0 = (MATCH) ConstantScore(name:holland name:hollandarod name:hollande name:hollander name:hollanderei name:hollandia name:hollandischer name:hollands name:hollandsbjerg name:hollandsch name:hollandsche name:hollandscheveld name:hollandsdiep name:hollandskamp name:hollandske), product of:\n  1.0 = boost\n  1.0 = queryNorm\n",

在上面的dubug部分，如果我们检查"parsedquery“部分，它在holland和holland*的情况下是不同的。所以我认为，特殊字符*对SynonymFilterFactory不起作用。

Answer 1

据我所知，同义词文件不支持通配符。

在您的例子中，您还可能遇到另一个问题，因为查询中的通配符通常用于搜索不完全匹配的结果。它取决于查询中使用的查询解析器。

换句话说，查询"holland*“搜索所有以"holland”开头的文档。

如果您希望Solr将通配符视为简单的字符，则应该对其进行转义。

我看到的另一个错误是，在您的字段定义中，您应该为案例(索引和查询)定义analyzer类型。

如果您为字段类型提供了一个简单的定义，如上面的示例所示，那么它将同时用于索引和查询。

Answer 2

请尝试：

<fieldType name="text_general" class="solr.TextField" positionIncrementGap="100" multiValued="true">
  <analyzer type="index">
    <tokenizer class="solr.StandardTokenizerFactory"/>
    <filter class="solr.LowerCaseFilterFactory"/>
  </analyzer>
  <analyzer type="query">
    <tokenizer class="solr.StandardTokenizerFactory"/>
    <filter class="solr.SynonymGraphFilterFactory" synonyms="synonyms.txt" ignoreCase="true" expand="true" tokenizerFactory="solr.StandardTokenizerFactory"/>
    <filter class="solr.LowerCaseFilterFactory"/>
  </analyzer>
</fieldType>