The way the append processor works is that it appends the values in value to the field specified in field, and it operates independently on each document. So if I use the _simulate API:
Edit to expand a bit: Since you're not providing an array field in the input documents, the processor is creating a new, empty array and appending value to that empty array.
Argh. That makes sense. Upon reading the description in the docs, I had hoped that each time you index the doc it appended the value to an array rather than overwriting it.
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.