concepts.dsl.constraint.ConstraintSatisfactionProblem#

class ConstraintSatisfactionProblem[source]#

Bases: object

A constraint satisfaction problem.

A constraint satisfaction problem is a set of constraints, and a set of variables. The solution to a constraint satisfaction problem is to find a set of values for the variables that satisfy all the constraints.

Methods

`add_constraint`(c[, note])	Add a constraint to the constraint satisfaction problem.
`add_domain_value`(identifier, value)	Add a value to the domain of a variable.
`add_equal_constraint`(left[, right, note])	Add an equality constraint.
`clone`([constraints])	Clone the constraint satisfaction problem.
`empty`()
`get_domain`(identifier)	Get the domain of a variable.
`get_type`(identifier)	Get the type of a variable.
`ground_assignment_value`(assignments, identifier)
`ground_assignment_value_partial`(assignments, ...)
`increment_state_timestamp`()
`new_actionable_var`(dtype[, wrap])	Create a new actionable variable.
`new_var`(dtype[, domain, actionable, name, ...])	Create a new variable.
`print_assignment_dict`(assignments)
`with_group`(expression[, note])	Add a group constraint.

Attributes

`index2record`	A mapping from variable indices to the variable objects.
`index_order`	A list of tuples (index, index), indicating that the first index should be solved earlier than the second index.
`constraints`	A list of constraints.

__init__(index2record=None, index_order=None, constraints=None, counter=0, state_timestamp=0)[source]#

Initialize a constraint satisfaction problem.

Parameters:

index2record (Dict[int, OptimisticValueRecord] | None) – a mapping from variable indices to the variable objects.
index_order (List[Tuple[int, int]] | None) – a list of tuples (index, index), indicating that the first index should be solved earlier than the second index.
constraints (List[Constraint | GroupConstraint] | None) – a list of constraints.
counter (int) – the counter for generating new variable indices.
state_timestamp (int)

__new__(**kwargs)#

add_constraint(c, note=None)[source]#

Add a constraint to the constraint satisfaction problem.

Parameters:

c (Constraint) – the constraint to be added.
note (Any | None) – the note of the constraint.

add_domain_value(identifier, value)[source]#

Add a value to the domain of a variable.

Parameters:

identifier (int)
value (Any)

add_equal_constraint(left, right=None, note=None)[source]#

Add an equality constraint.

Parameters:

left (TensorValue | OptimisticValue) – the left hand side of the equality constraint.
right (TensorValue | OptimisticValue | bool | None) – the right hand side of the equality constraint.
note (Any | None) – the note of the constraint.

clone(constraints=None)[source]#

Clone the constraint satisfaction problem.

Parameters:: constraints (List[Constraint] | None) – the constraints to be replaced into the cloned constraint satisfaction problem. If None, the constraints of the original constraint satisfaction problem will be used.
Return type:: ConstraintSatisfactionProblem

empty()[source]#

get_domain(identifier)[source]#

Get the domain of a variable.

Parameters:: identifier (int)
Return type:: Set[Any] | None

get_type(identifier)[source]#

Get the type of a variable.

Parameters:: identifier (int)
Return type:: TensorValueTypeBase | PyObjValueType

ground_assignment_value(assignments, identifier)[source]#

Parameters:

assignments (Dict[int, Assignment])
identifier (int)

Return type:

Any

ground_assignment_value_partial(assignments, identifier)[source]#

Parameters:

assignments (Dict[int, Assignment])
identifier (int)

Return type:

Any

increment_state_timestamp()[source]#

new_actionable_var(dtype, wrap=False, **kwargs)[source]#

Create a new actionable variable.

Parameters:

dtype (TensorValueTypeBase | PyObjValueType) – the type of the variable.
wrap (bool) – whether to wrap the variable index into an OptimisticValue.

Return type:

int | OptimisticValue

new_var(dtype, domain=None, actionable=False, name=None, is_constant=False, wrap=False)[source]#

Create a new variable.

Parameters:

dtype (TensorValueTypeBase | PyObjValueType) – the type of the variable.
domain (Set[Any] | None) – the domain of the variable. If None, it will be assumed to the full domain of the type.
actionable (bool) – whether the variable is actionable.
name (str | None) – the name of the variable.
is_constant (bool) – whether the variable is constant.
wrap (bool) – whether to wrap the variable index into an OptimisticValue.

Returns:

The index of the new variable (int) if wrap is False, or the wrapped OptimisticValue if wrap is True.

Return type:

int | OptimisticValue

print_assignment_dict(assignments)[source]#

Parameters:: assignments (Dict[int, Assignment])

with_group(expression, note=None)[source]#

Add a group constraint.

Parameters:

expression (ValueOutputExpression)
note (Any)

constraints: List[Constraint | GroupConstraint]#: A list of constraints.

index2record: Dict[int, OptimisticValueRecord]#: A mapping from variable indices to the variable objects.

index_order: List[Tuple[int, int]]#: A list of tuples (index, index), indicating that the first index should be solved earlier than the second index.