Struct cedar_policy_core::evaluator::Evaluator

pub struct Evaluator<'e> { /* private fields */ }

Evaluator object.

Conceptually keeps the evaluation environment as part of its internal state, because we will be repeatedly invoking the evaluator on every policy in a Slice.

Implementations

impl<'q, 'e> Evaluator<'e>

pub fn new( q: &'q Request, entities: &'e Entities, extensions: &'e Extensions<'e>, ) -> Result<Self, EvaluationError>

Create a fresh Evaluator for the given request, which uses the given Entities to resolve entity references. Use the given Extensions when evaluating the request.

(An Entities is the entity-hierarchy portion of a Slice, without the policies.)

Can throw an error, eg if evaluating attributes in the context throws an error.

pub fn evaluate(&self, p: &Policy) -> Result<bool, EvaluationError>

Evaluate the given Policy, returning either a bool or an error. The bool indicates whether the policy applies, ie, “is satisfied” for the current request. This is different than “if the current request should be allowed” – it doesn’t consider whether we’re processing a Permit policy or a Forbid policy.

pub fn partial_evaluate( &self, p: &Policy, ) -> Result<Either<bool, Expr>, EvaluationError>

Partially evaluate the given Policy, returning one of:

1. A boolean, if complete evaluation was possible
2. An error, if the policy is guaranteed to error
3. A residual, if complete evaluation was impossible The bool indicates whether the policy applies, ie, “is satisfied” for the current request. This is different than “if the current request should be allowed” – it doesn’t consider whether we’re processing a Permit policy or a Forbid policy.

pub fn run_to_error( &self, e: &Expr, slots: &SlotEnv, ) -> (PartialValue, Option<EvaluationError>)

Run an expression as far as possible. however, if an error is encountered, instead of error-ing, wrap the error in a call the error extension function.

pub fn interpret( &self, e: &Expr, slots: &SlotEnv, ) -> Result<Value, EvaluationError>

Interpret an Expr into a Value in this evaluation environment.

Ensures the result is not a residual. May return an error, for instance if the Expr tries to access an attribute that doesn’t exist.

pub fn partial_interpret( &self, e: &Expr, slots: &SlotEnv, ) -> Result<PartialValue, EvaluationError>

Interpret an Expr into a Value in this evaluation environment.

May return a residual expression, if the input expression is symbolic. May return an error, for instance if the Expr tries to access an attribute that doesn’t exist.

Trait Implementations

impl<'e> Debug for Evaluator<'e>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.