Grouping
This page covers grouping of TypeQL Get query results.
Behavior
A Get query without aggregation or grouping returns a stream of ConceptMaps, each representing a matched solution for the query pattern. With grouping, the results of the same query are divided into groups based on a specified variable. Each group contains all results with the selected variable having the same value or concept.
Grouping can be used alongside aggregation to return an aggregated value for every group.
Usage
For this example, use a database with the IAM schema and sample data loaded.
Let’s match all person entities that have a full-name and email attributes, retrieve them all,
and group by the instance of entity.
If any person has more than one attribute of the full-name or email types, they are placed in its respective group.
The grouped results are easier to navigate and process this way.
match
$p isa person, has full-name $n, has email $e;
get;
group $p;TypeDB Studio produces the following or similar textual output:
iid 0x826e800a8000000000000003 isa person => {
{
$n Jacky isa full-name;
$p iid 0x826e800a8000000000000003 isa person;
$e jk@typedb.com isa email;
} {
$p iid 0x826e800a8000000000000003 isa person;
$n Jack Keeper isa full-name;
$e jk@typedb.com isa email;
}
}Usage with polymorphism
For this example, use a database with the IAM schema and sample data loaded.
Now let’s query for any attributes of any person, grouped by the person entity.
match
$p isa person, has $a;
get;
group $p;TypeDB Studio produces the following or similar textual output:
iid 0x826e800a8000000000000003 isa person => {
{
$a Jacky isa full-name;
$p iid 0x826e800a8000000000000003 isa person;
} {
$a Jack Keeper isa full-name;
$p iid 0x826e800a8000000000000003 isa person;
} {
$a jk@typedb.com isa email;
$p iid 0x826e800a8000000000000003 isa person;
}
}Usage with aggregation
For this example, use a database with the IAM schema and sample data loaded.
Now let’s see how many people own every attribute by grouping the same query by the $a variable
and counting the results in every group.
match
$p isa person, has $a;
get;
group $a;
count;TypeDB Studio produces the following or similar textual output:
kevin.morrison@typedb.com isa email => Optional[1]
pearle.goodman@typedb.com isa email => Optional[1]
masako.holley@typedb.com isa email => Optional[1]
Jack Keeper isa full-name => Optional[1]
Pearle Goodman isa full-name => Optional[1]
Jacky isa full-name => Optional[1]
Masako Holley isa full-name => Optional[1]
jk@typedb.com isa email => Optional[1]
Kevin Morrison isa full-name => Optional[1]