Postgres support (user defined types)

[nene]

Domains

• CREATE DOMAIN
  • name [AS] data_type
  • COLLATE collation
  • DEFAULT expression
  • domain_constraint [ ... ]:
    • CONSTRAINT constraint_name
    • NOT NULL
    • NULL
    • CHECK (expression)
• DROP DOMAIN
  • IF EXISTS
  • name, ...
  • CASCADE | RESTRICT
• ALTER DOMAIN
  • SET DEFAULT expression | DROP DEFAULT
  • { SET | DROP } NOT NULL
  • ADD domain_constraint
    • CONSTRAINT name
    • CHECK (...)
    • NOT NULL
    • [ NOT VALID ]
  • DROP CONSTRAINT [ IF EXISTS ] constraint_name [ RESTRICT | CASCADE ]
  • RENAME CONSTRAINT constraint_name TO new_constraint_name
  • VALIDATE CONSTRAINT constraint_name
  • OWNER TO { new_owner | CURRENT_ROLE | CURRENT_USER | SESSION_USER }
  • RENAME TO new_name
  • SET SCHEMA new_schema

Types

• CREATE TYPE
  • AS ( attribute_name data_type [ COLLATE collation ] [, ... ] )
  • AS ENUM ( 'label' [, ... ] )
  • AS RANGE ( ... ):
    • SUBTYPE = subtype
    • SUBTYPE_OPCLASS = subtype_operator_class
    • COLLATION = collation
    • CANONICAL = canonical_function
    • SUBTYPE_DIFF = subtype_diff_function
    • MULTIRANGE_TYPE_NAME = multirange_type_name
  • ( ... ):
    • INPUT = input_function
    • OUTPUT = output_function
    • RECEIVE = receive_function
    • SEND = send_function
    • TYPMOD_IN = type_modifier_input_function
    • TYPMOD_OUT = type_modifier_output_function
    • ANALYZE = analyze_function
    • SUBSCRIPT = subscript_function
    • INTERNALLENGTH = { internallength | VARIABLE }
    • PASSEDBYVALUE
    • ALIGNMENT = alignment
    • STORAGE = storage
    • LIKE = like_type
    • CATEGORY = category
    • PREFERRED = preferred
    • ELEMENT = element
    • DELIMITER = delimiter
    • COLLATABLE = collatable
  • CREATE TYPE name;
• DROP TYPE
  • IF EXISTS
  • name, ...
  • CASCADE | RESTRICT
• ALTER TYPE
  • OWNER TO { new_owner | CURRENT_ROLE | CURRENT_USER | SESSION_USER }
  • RENAME TO new_name
  • SET SCHEMA new_schema
  • RENAME ATTRIBUTE attribute_name TO new_attribute_name
    • CASCADE | RESTRICT
  • action, ...:
    • ADD ATTRIBUTE attribute_name data_type
      • COLLATE collation
      • CASCADE | RESTRICT
    • DROP ATTRIBUTE attribute_name
      • IF EXISTS
      • CASCADE | RESTRICT
    • ALTER ATTRIBUTE attribute_name
      • [ SET DATA ] TYPE data_type
      • COLLATE collation
      • CASCADE | RESTRICT
  • ADD VALUE new_enum_value
    • IF NOT EXISTS
    • { BEFORE | AFTER } neighbor_enum_value
  • RENAME VALUE existing_enum_value TO new_enum_value
  • SET ( property = value [, ... ] )

[maheshsundaram]

Thanks for documenting this! I would be glad to help out if I can figure out how. If you have any time and is convenient for you, can you point me to where I can begin?

[nene]

I guess you can look at existing commits to see how. This one might be a good start: 00adcae

The trickiest bit is figuring out the structure of the resulting syntax tree. For CREATE TYPE I guess it should be something like:

export interface CreateTypeStmt extends BaseNode {
  type: "create_type_stmt";
  createKw: Keyword<"CREATE">;
  typeKw: Keyword<"TYPE">;
  name: EntityName;
  // ENUM definition is probably the simplest, though others should fit here as well
  definition?: AsClause<EnumTypeDefinition>;
}

export interface EnumTypeDefinition extends BaseNode {
  type: "enum_type_definition";
  enumKw: Keyword<"ENUM">;
  values: ParenExpr<ListExpr<StringLiteral>>;
}