Julian Feinauer created CALCITE-3345: ----------------------------------------
Summary: Implement time_bucket function Key: CALCITE-3345 URL: https://issues.apache.org/jira/browse/CALCITE-3345 Project: Calcite Issue Type: New Feature Reporter: Julian Feinauer See here for information on the `time_bucket` function: https://docs.timescale.com/latest/api#time_bucket This is a more powerful version of the standard PostgreSQL date_trunc function. It allows for arbitrary time intervals instead of the second, minute, hour, etc. provided by date_trunc. The return value is the bucket's start time. This would especially help with time averaging but keeps everything SQL compliant. E.g. queries like Example query from (https://www.timescale.com/): {code:sql} SELECT time_bucket('10 seconds', time) AS ten_second, machine_id, avg(temperature) AS "avgT", min(temperature) AS "minT", max(temperature) AS "maxT", last(temperature, time) AS "lastT" FROM measurements WHERE machine_id = 'C931baF7' AND time > now() - interval '150s' GROUP BY ten_second ORDER BY ten_second DESC; {code} -- This message was sent by Atlassian Jira (v8.3.2#803003)