I know the only way to achieve it in old releases was through scripting.
I'm wondering if that limitation was overcome in the latest release and potentially we could you the standard command, indicating the correct format/interval (i.e):
scripting is one way to go. But you could also use the reindex API and add the day of the week in its own field and then run a terms aggregation on that - which would be waaaay faster than the scripting solution.
Is there any possibility to factor that in as well during the reindex operation (maybe via a second field in addition to the day of the week add the timezone and base your queries on that)?
Only if we index all the time zone available... not feasible in this scenario.
We've got it up running for other date formats (year, month and year etc)
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.