The TABLE statement1

Purpose

This statement provides syntactic sugar for defining tabular information conveniently. It introduces a number of relations and rules in a single statement, to simplify a script.

Syntax

`TABLE` <Concept> `(` (<expression> `:` <Concept> <Multiplicities>?)* `)`

where:

  • <label> is the name of the rule. It can be a single word or a string (enclosed by double brackets). It is followed by a colon (:) to distinguish the label from the concept that follows.
  • <Concept> is the name of the Concept for atoms of which the rule specifies an identity
  • Between brackets are expressions whose source concept must be <Concept>. This is enforced by the type system.

Semantics by example

TABLE  C (e1:C1, e2:C2)

translates into the following declarations:

RELATION E1[C*C1]
RELATION E2[C*C2]

Multiplicity annotations are allowed. For example:

TABLE  C (e1:C1[UNI,TOT], e2:C2, e3:C[ASY], ...)

translates into the following declarations:

RELATION E1[C*C1] [UNI,TOT]
RELATION E2[C*C2]
RELATION E3[C*C] [ASY]
...

Practice

This statement makes nice combinations with the IDENT statement. For example to define two identities for persons:

TABLE Person(name:String, ssn:SSN, birthplace:City, birthdate:Date)
IDENT Person(ssn)
IDENT Person(name,birthplace,birthdate)

This states that a person is uniquely defined by ssn, but also by the combination of name, birthplace, and birthdate. This statement can also be used to objectify (reify) an expression e

TABLE T(pi:P, rho:R)
IDENT T(pi,rho)
RULE "Create T" : e |- pi~;rho
RULE "Delete T" : pi~;rho |- e

If a user is tempted to replace the Create/Delete pair with a single equivalence, this becomes:

TABLE T(pi:P, rho:R)
IDENT T(pi,rho)
RULE e = pi~;rho

1 This statement will be available in Ampersand vs. 4.

results matching ""

    No results matching ""