I have all of the mappings for my fields nested under "properties" is this necessary? I'm not sure when I would use properties and when I would not.
Also, the "fields" setting under "myfield" is making "myfield" a "keyword" type in addition to a "text" type right? Is writing it this way just less verbose then defining it twice with each type?
Just so I understand how this works, when I look at my events I see a "myfield" and a "myfield.keyword".
If I want to do a full text search on the field I use "myfield" which is analyzed (tokenized).
If there is a specific value I am looking for ( or to filter on ) I can use myfield.keyword.
I realized I have a bunch of my fields defined as "type": "text" AND "type": "keyword", but many of them are just ID fields that I would never need to do a full text search on. I could save space and indexing time by changing them to just keyword right?
If I did change it I would just change it to this right:
"myfield": {
"type": "keyword",
"norms": false,
}
Also, I can still do wildcard searches on keyword fields right?
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.