The cause of CJKBigramFilter(outputUnigrams=true) Not Expanding to MultiPhraseQuery

[Mingchun Zhao]

Hi,

When performing a phrase query using CJKBigramFilter with the
outputUnigrams=true in Solr 9, it does not expand into a MultiPhraseQuery.
Could this be due to a specification change?

Below are the details:

Field Definition:
*```*
<field name="city" type="text_bigram" indexed="true" stored="true"
multiValued="true" />
<fieldType name="text_bigram" class="solr.TextField">
<analyzer>
<tokenizer class="solr.StandardTokenizerFactory"/>
<filter class="solr.CJKBigramFilterFactory" outputUnigrams="true"/>
</analyzer>
</fieldType>
```

When performing the phrase search with q=city:"東京"("東京" is the kanji
representation of "Tokyo"):

In Solr 4.10.4:
```
"debug": {
"rawquerystring": "city:\"東京\"",
"querystring": "city:\"東京\"",
"parsedquery": "MultiPhraseQuery(city:\"(東 東京) 京\")",
"parsedquery_toString": "city:\"(東 東京) 京\"",
*```*

In Solr 9.4.0:
```
"debug":{
"rawquerystring":"city:\"東京\"",
"querystring":"city:\"東京\"",
"parsedquery":"PhraseQuery(city:\"東 京\") city:東京",
"parsedquery_toString":"city:\"東 京\" city:東京",
"explain":{ },
"QParser":"LuceneQParser",
*```*


I’d appreciate any information you might have about this.

Regards,
Mingchun