Operator Evaluation

The Python implementation separates expression evaluation from the semantics of individual operator categories.

Operator Vocabulary

constraint_handler.schemas.operators defines the built-in operator categories used in expression schemas:

Category	Operators
ComparisonOperator	eq, neq, max, min
StringOperator	concat, length
ConditionalOperator	if, ite, getOrElse, hasValue
ArithmeticOperator	Arithmetic, conversion, and ordering operators
LogicOperator	Boolean operators
SetOperator	Set operators
MultimapOperator	Multimap operators

constraint_handler.schemas.expression defines expression-tree nodes such as Operation, Variable, and Val. Its Operator type combines the built-in categories with Python and lambda operators.

Evaluation

constraint_handler.evaluator evaluates expression trees, propagates warnings and bad values, and dispatches evaluated operators by category. It also owns evaluator-specific recovery policy: selected operators can still produce a result when an argument is bad.

The operator semantics are implemented by category:

Module	Categories
constraint_handler.arithmetic	ArithmeticOperator
constraint_handler.comparison	ComparisonOperator
constraint_handler.conditional	ConditionalOperator
constraint_handler.logic	LogicOperator
constraint_handler.multimap	MultimapOperator
constraint_handler.set	SetOperator
constraint_handler.string	StringOperator

Each semantic module exposes evaluate_operator(operator, args) and returns an atom.EvalResult. New built-in operators should be added to their category in schemas.operators, implemented in the corresponding semantic module, and dispatched from evaluator.operator().