There is managed schema, which means it is editable via API, and there is 'schemaless' mode that uses that to auto-define the field based on the first occurance.
'schemaless' mode does not know if the field will be multi-valued the first time it sees content for that field. So, all the fields created automatically are multivalued. You can change the definition or you can define the field explicitly using the API or Admin UI. 'schemaless' is only there really for a quick prototyping with unknown content. Regards, Alex. P.s. That's my SO answer :-) Glad you found it useful. ---- http://www.solr-start.com/ - Resources for Solr users, new and experienced On 13 March 2017 at 11:15, Furkan KAMACI <furkankam...@gmail.com> wrote: > Hi, > > I generate dummy documents to test Solr 6.4.2. I create a field like that > at my test code: > > int customCount = r.nextInt(500); > document.addField("custom_count", customCount); > > This field is indexed as: > > org.apache.solr.schema.TrieLongField > > and > > Multivalued. > > I want to use FieldCache on multivalued field and don't want it to be > multivalued. When I check managed-schema I see that: > > <fieldType name="long" class="solr.TrieLongField" > positionIncrementGap="0" docValues="true" precisionStep="0"/> > <fieldType name="longs" class="solr.TrieLongField" > positionIncrementGap="0" docValues="true" multiValued="true" > precisionStep="0"/> > > So, it seems that it's predicted as longs instead of long. > > What is the reason behind that? > > Kind Regards, > Furkan KAMACI
