Overview

In Ultipa Graph, a schema represents a node or edge type. When modeling a real-world scenario into a graph, node schemas are generally used to depict different types of entities (such as Account, Company, Card), and edge schemas represent types of relations between entities (such as Follow, Hold, Transfer).

The definition of schemas is a prerequisite for creating any node or edge data. Each node or edge is designated to one and only one schema. A graphset usually contains multiple node schemas and edge schemas.

A node schema and an edge schema both named default are automatically created during the creation of a graphset. They can be freely utilized. However, the two default schemas are not allowed to be altered or dropped.

The symbol @ is used in UQL to denote a schema. The expression @<schema> specifies a certain schema, such as @Account.

Show Schema

// Show all schemas in the graphset
show().schema()

// Show all node schemas in the graphset
show().node_schema()

// Show all edge schemas in the graphset
show().edge_schema()

// Show the node schema named movie in the graphset
show().node_schema(@movie)

// Show the edge schema named filmedIn in the graphset
show().edge_schema(@filmedIn)

Example result:

_nodeSchema

_edgeSchema

The properties only contains custom properties; system properties are not included in the results.

Create Schema

// Create a node schema named movie in the graphset, and provide description
create().node_schema("movie", "The movies added by the admin")

// Create an edge schema named filmedIn in the graphset
create().edge_schema("filmedIn")

// Create multiple node/edge schemas at one time
create()
  .node_schema("movie", "The movies added by the admin")
  .node_schema("country")
  .edge_schema("filmedIn")

Naming Conventions

Here are the naming conventions for schemas:

• Contains 2 to 64 characters.
• Not allowed to start with a tilde symbol ~.
• Not allowed to contain backquote symbol `.
• Not allowed to use any reserved words.

All node schemas in a graphset must have distinct names, and the same applies for edge schemas. A node schema and an edge schema may share the same name.

When the schema name contains characters other than letters (A-Z, a-z), numbers (0-9) and underscores (_), the schema name must be wrapped with a pair of backquotes (`) when being used.

find().nodes({@`movie*`}) as n
return n

Use TRY

Create three node schemas at the same time, but one of the names (default) is duplicated with an existing node schema.

create().node_schema("new_1").node_schema("default").node_schema("new_2")

The creation of the node schema new_1, which was specified before the duplicated schema, succeeds. However, the one (new_2) specified after the duplicated schema fails, with the error message Schema already exist! returned.

TRY create().node_schema("new_1").node_schema("default").node_schema("new_2")

The creation of the schemas is the same as above, though the error message is shielded by the TRY prefix, while returning the message SUCCEED.

Alter Schema

// Alter name and description of the node schema currently named movie
alter().node_schema(@movie)
  .set({name: "Adm_movie", description: "Movies added by the admin"})

// Alter description of the edge schema named filmedIn
alter().edge_schema(@filmedIn).set({description: "The country where a movie is filmed"})

Drop Schema

Dropping a schema means to delete the schema, along with any nodes or edges belonging to that schema. The two default schemas cannot be dropped.

// Drop the node schema named movie
drop().node_schema(@movie)

// Drop the edge schema named filmedIn
drop().edge_schema(@filmedIn)
                   
// Drop multiple node/edge schemas at one time
drop()
  .node_schema(@movie)
  .node_schema(@country)
  .edge_schema(@filmedIn)