Aggregation

Aggregate functions compute summary values over matched results. They appear in the return block and automatically group by any non-aggregated return keys.

Aggregate functions

Function	Description
`count($var)`	Number of matched bindings
`sum($var.prop)`	Sum of a numeric property
`avg($var.prop)`	Arithmetic mean of a numeric property
`min($var.prop)`	Minimum value
`max($var.prop)`	Maximum value

Count

Count the number of matched nodes:

query total_people() {
    match {
        $p: Person
    }
    return { count($p) as total }
}

[
  { "total": 847 }
]

Sum and average

Compute totals and means over numeric properties:

query age_stats() {
    match {
        $p: Person
    }
    return {
        count($p) as total,
        sum($p.age) as total_age,
        avg($p.age) as average_age
    }
}

Min and max

Find the extreme values of a property:

query age_range() {
    match {
        $p: Person
    }
    return {
        min($p.age) as youngest,
        max($p.age) as oldest
    }
}

Aliasing with `as`

Every aggregate should be aliased with as to give the output column a readable name:

query company_headcount() {
    match {
        $p: Person
        $p worksAt $c
    }
    return { $c.name, count($p) as headcount }
}

Grouping

When you mix plain properties and aggregates in return, the plain properties become grouping keys. Results are grouped by those keys automatically:

query employees_per_company() {
    match {
        $p: Person
        $p worksAt $c
    }
    return { $c.name, count($p) as employees }
}

[
  { "name": "Acme Corp", "employees": 42 },
  { "name": "Globex", "employees": 17 }
]

Multiple aggregates

Return several aggregates in a single query:

query company_age_stats() {
    match {
        $p: Person
        $p worksAt $c
    }
    return {
        $c.name,
        count($p) as headcount,
        avg($p.age) as avg_age,
        min($p.age) as youngest,
        max($p.age) as oldest
    }
}

Ordering by aggregated values

Use the aggregate alias in the order block:

query largest_companies() {
    match {
        $p: Person
        $p worksAt $c
    }
    return { $c.name, count($p) as headcount }
    order { headcount desc }
    limit 10
}

omnigraph read ./repo.omni \
    --query queries.gq \
    --name largest_companies \
    --json

[
  { "name": "Acme Corp", "headcount": 42 },
  { "name": "Globex", "headcount": 17 },
  { "name": "Initech", "headcount": 12 }
]

Aggregation

On this page