Kaleidoscope (programming language)

From Wikipedia, the free encyclopedia
Jump to: navigation, search

The Kaleidoscope programming language is a constraint programming language embedding constraints into an imperative object-oriented language. It adds keywords always, once, and assert..during (formerly while..assert) to make statements about relational invariants. Objects have constraint constructors, which are not methods, to enforce the meanings of user-defined datatypes.

There are three versions of Kaleidoscope which show an evolution from declarative to an increasingly imperative style. Differences between them are as follows.[1]

Kaleidoscope'90 Kaleidoscope'91 Kaleidoscope'93
Constraint Evaluation Lazy Eager Eager
Variables Hold streams Hold streams Imperative
Concurrent Constraints Strict Strict Non-strict
Syntax Smalltalk-like Algol-like Algol-like
Constraint Model Refinement Refinement Perturbation
Method Dispatching Single Multiple Multiple
Assignment As a constraint As a constraint Destructive

Example[edit]

Compare the two code segments, both which allow a user to drag the level of mercury in a simple graphical thermometer with the mouse.

Without constraints:

while mouse.button = down do
 old <- mercury.top;
 mercury.top <- mouse.location.y;
 temperature <- mercury.height / scale;
 if old < mercury.top then
  delta_grey( old, mercury.top );
  display_number( temperature );
 elseif old > mercury.top then
  delta_white( mercury.top, old );
  display_number( temperature );
 end if;
end while;

With constraints:

always: temperature = mercury.height / scale;
always: white rectangle( thermometer );
always: grey rectangle( mercury );
always: display number( temperature );
while mouse.button = down do
 mercury.top = mouse.location.y;
end while;

References[edit]

  • Lopez, Gus; Bjorn Freeman-Benson; Alan Borning (1994). "Kaleidoscope: A Constraint Imperative Programming Language". Constraint Programming. Springer-Verlag. pp. 313–329. 
  • Marriot, Kim; Peter J. Stuckey (1998). Programming with constraints: An introduction. MIT Press.  ISBN 0-262-13341-5