Yes. The sorting is done internally with objects a that are NOT documents.
Unindexed fields are only available when building Document. The
<>Collector in Lucene fills the aforementioned objects from index fields,
when it finds a qualifying dcoument it looks for position of the sort
field in the field pos index
and its not there. Probably doesn't even look ... if you think about it.
On Sunday, October 7, 2012 3:05:55 AM UTC-4, Jae wrote:
Hi
I thought that I could sort with the date field which was not indexed but
sorting didn't work.
I checked the sort value in the response and all were '0', which means, if
I want to sort with the field, I should index that field? Am I correct?
This is not what I expected. I thought sorting was not related with the
index.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.