I changed the tokenizer class from KeywordTokenizerFactory to
WhitespaceTokenizerFactory for the query analyzer using the Schema API, it
seems to have solved the problem.
Sent from Mail for Windows 10
From: ufuk yılmaz
Sent: 02 March 2021 20:47
To: solr-user@lucene.apache.org
Subject: Default conjunction behaving differently after field type change
Hello all,
>From the Solr 8.4 (my version) documentation:
“The OR operator is the default conjunction operator. This means that if there
is no Boolean operator between two terms, the OR operator is used. To search
for documents that contain either "jakarta apache" or just "jakarta," use the
query:
"jakarta apache" jakarta
or
"jakarta apache" OR jakarta”
I had a field type=”string” in my old schema:
<fieldType name="string" class="solr.StrField" sortMissingLast="true"
docValues="true"/>
<field name="username" type="string"/>
I could use queries like:
username: (user1 user2 user3)
So it would find the documents of all 3 users (conjunction is OR)
-
Recently I changed the field definition in a new schema:
<fieldType name="string_ci" class="solr.TextField" sortMissingLast="true"
omitNorms="true" stored="true" docValues="false">
<analyzer type="query">
<tokenizer class="solr.KeywordTokenizerFactory"/>
<filter class="solr.LowerCaseFilterFactory"/>
</analyzer>
</fieldType>
<field name="username" type="string_ci" indexed="true"/>
When I search with the same query:
username: (user1 user2 user3)
I get no results unless I change it to either:
username: (user1 OR user2 OR user3) //
username: (“user1” “user2” “user3”)
First I was thinking the default conjunction operator changed to AND, but it
seems now standart query parser thinks user1 user2 user3 is a single string
containin spaces I guess?
I couldn’t find how the default “string” field queries are analyzed, what is
the difference that may cause this behavior?
--ufuk yilmaz
Sent from Mail for Windows 10
