Skip to main content
Version: Next



summarize perform aggregations


[summarize] [<field>:=]<agg> [where <expr>][, [<field>:=]<agg> [where <expr>] ...] [by [<field>][:=<expr>] ...]


The summarize operator consumes all of its input, applies an aggregate function to each input value optionally organized with the group-by keys specified after the by keyword, and at the end of input produces one or more aggregations for each unique set of group-by key values.

The summarize keyword is optional since it is an implied operator.

Each aggregate function may be optionally followed by a where clause, which applies a Boolean expression that indicates, for each input value, whether to deliver it to that aggregate. (where clauses are analogous to the where operator.)

The output field names for each aggregate and each key are optional. If omitted, a field name is inferred from each right-hand side, e.g, the output field for the count aggregate function is simply count.

A key may be either an expression or a field. If the key field is omitted, it is inferred from the expression, e.g., the field name for by lower(s) is lower.

If the cardinality of group-by keys causes the memory footprint to exceed a limit, then each aggregate's partial results are spilled to temporary storage and the results merged into final results using an external merge sort. The same mechanism that spills to storage can also spill across the network to a cluster of workers in an adaptive shuffle, though this is not yet implemented.


Average the input sequence:

echo '1 2 3 4' | zq -z 'summarize avg(this)' -



Sum the input sequence, leaving out the summarize keyword:

echo '1 2 3 4' | zq -z 'sum(this)' -



Create integer sets by key and sort the output to get a deterministic order:

echo '{k:"foo",v:1}{k:"bar",v:2}{k:"foo",v:3}{k:"baz",v:4}' | zq -z 'set:=union(v) by key:=k' - | sort



Use a where clause:

echo '{k:"foo",v:1}{k:"bar",v:2}{k:"foo",v:3}{k:"baz",v:4}' | zq -z 'set:=union(v) where v > 1 by key:=k' - | sort