Thanks Jack.
our schema version is 1.3
<schema name="watchdox-solr" version="1.3">
we are using the official solr 3.4 release. actually we use maven to
download solr war and artifacts
<dependency>
<groupId>org.apache.solr</groupId>
<artifactId>solr</artifactId>
<version>3.4.0</version>
<type>war</type>
</dependency>
No, I did not modify the schema at anytime, all documents where indexed with
the same schema.
Yes, we have additional copyFields into the text field. usually none of them
will contain the same text as the document name, its mostly owner
information.
to make the picture clearer:
we are indexing text documents, every document has a db row, and the content
on disk space.
we index the db with DataImportHandler. among other columns we index the
document name which is our 'doc' column, another field 'docname' which is
the document display name,usually the same as 'doc', and we also index the
document content in 'content' field (the content is indexed in the same
DataImportHandler process).
we copy 'doc' and 'content' into the 'text' field plus some other fields
usually owner information like email address etc. it may be that the content
contains the document name or parts of it.
Thank you
--
View this message in context:
http://lucene.472066.n3.nabble.com/search-on-default-field-returns-less-documents-tp3999896p4000238.html
Sent from the Solr - User mailing list archive at Nabble.com.
