Skip to main content
Since 1.2.0 Group data into buckets based on time interval, while filling in gaps of missing data. If you don’t provide a gapfilling algorithm, such as locf or interpolate, gaps are left as NULL in the returned data.

​
Arguments

NameTypeDefaultRequiredDescription
bucket_widthINTERVAL | INTEGER-✔A time interval to specify the length of each bucket. For example, use 1 day to get daily buckets. Use INTEGER only if your time column is integer-based.
timeTIMESTAMPTZ | INTEGER-✔The timestamp on which to base the bucket
timezoneTEXT-❌The timezone to use for bucketing. For example, Europe/Berlin. Available in 2.9 or later. Does not work for integer-based time. If you have an untyped start or finish argument and a timezone argument, you might run into a problem where you are not passing your arguments for the parameter that you expect. To solve this, either name your arguments or explicitly type cast them.
startTIMESTAMPTZ | INTEGER-❌The start of the period to gapfill. Values before start are passed through, but no gapfilling is performed. Use INTEGER only if your time column is integer-based. Best practice is to use the WHERE clause. Specifying start is legacy. The WHERE is more performant, because the query planner can filter out s by constraint exclusion.
finishTIMESTAMPTZ | INTEGER-❌The end of the period to gapfill. Values after finish are passed through, but no gapfilling is performed. Use INTEGER only if your time column is integer-based. Best practice is to use the WHERE clause. Specifying finish is legacy. The WHERE is more performant, because the query planner can filter out s by constraint exclusion.

​
Returns

The start time of the time bucket.