Graph schemas

[nichtich]

Graph schemas are required in practice for database management systems and for validation but they won't be part of version 1.0.0 of PG specification. A graph schema can be a graph of its own.

For instance GQL Example from this article

CREATE GRAPH TYPE socialNetworkSchema AS {
    (Person :Person&TaxPayer {name STRING, dob DATE, taxNo STRING}),
    (City :City {name STRING, state string, country STRING}),
    (Company :Company {name STRING, description STRING}),
    (Person)-[LivesIn :LIVES_IN {since DATE}]->(City),
    (Company)-[HeadquartersIn :HEADQUARTERS_IN {since DATE}]->(City),
    (Person)-[WorksFor :WORKS_FOR {since DATE}]->(Company),
    (Person)~[MarriedTo :MARRIED_TO {since DATE, until DATE}]~(Person)
}

as graph in PG format:

Person  :Person :TaxPayer name:STRING dob:DATE taxNo:STRING
City    :City             name:STRING state:STRING country:STRING
Company :Company          name:STRING description:STRING

# Edge identifiers are optional but included here
LivesIn:        Person  -> City :LIVES_IN        since:DATE
HeadquartersIn: Company -> City :HEADQUARTERS_IN since:DATE
WorksFor:       Person  -> Company :WORKS_FOR    since:DATE
MarriedTo:      Person -- Person :MARRIED_TO     since:DATE until:DATE

PG currently only supports data types STRING, BOOL and NUMBER. In practice INT and FLOAT should also be supported

Each property can be constraint as unique (-ID), required/optional (!/?) and/or repeatable ( */+). Examples:

title:STRING!   # required, non-repeatable
uri:STRING-ID!  # required, unique, non-repeatable
alias:STRING*   # repeatable, optional

Without cardinality and uniqueness indicator the constraint is unknown and thus may differ between implementations in different database management systems.

To specify schema for all nodes:

* key:INT-ID! # no labels => all nodes
* -> * date:STRING? # special node identifier "*"