Add unknown_entities method to PartialResponse

[vladcebo]

Introduces unknown_entities method to the PartialResponse.

Description of changes

Introduces a method uknown_entities on the PartialResponse.
During partial evaluation, knowing unknown entity ids can help clients to fetch the missing entities
and run the evaluation again.

This is achievable using the existing code via:

    let unknowns = partial_response
        .all_residuals()
        .fold(HashSet::new(), |mut unknowns, p| {
            unknowns.extend(p.unknown_entities().into_iter());
            unknowns
        });

Unfortunately this is slow for large policy sets due to Policy::from_ast relying on pretty printing.

so calling that in a loop doing entity discovery is just very expensive.

Issue #, if available

N/A

Checklist for requesting a review

The change in this PR is (choose one, and delete the other options):

• A backwards-compatible change requiring a minor version bump to cedar-policy (e.g., addition of a new API).

I confirm that this PR (choose one, and delete the other options):

• Updates the "Unreleased" section of the CHANGELOG with a description of my change (required for major/minor version bumps).

I confirm that cedar-spec (choose one, and delete the other options):

• Does not require updates because my change does not impact the Cedar formal model or DRT infrastructure.

I confirm that docs.cedarpolicy.com (choose one, and delete the other options):

• Does not require updates because my change does not impact the Cedar language specification.

[vladcebo]

Looking more at the API, there are many other methods that map the inner result viaPolicy::from_ast.

For example, say I'm interested in the policy ids that are definitely satisfied:

    let satisfied: HashSet<PolicyId> = cedar_response
        .definitely_satisfied()
        .map(|policy| policy.id().clone())
        .collect();

Well, you if you have lots of satisfied policies then there's lots of extra work in pretty printing that stuff.
Maybe can think of a more generic approach to expose that information in a cheaper way?

[adpaco-aws]

Hi @vladcebo , thanks for the PR!

Maybe can think of a more generic approach to expose that information in a cheaper way?

Yes, it sounds like at the very least we should test the assumption that pretty-printing is cheaper than building the est with est::Policy::from(ast.clone()). I'm also not sure if these interfaces will remain the same with Type-Aware Partial Evaluation. @shaobo-he-aws any thoughts?

[shaobo-he-aws]

cedar-policy/rfcs#95 should be an independent implementation in parallel to the existing PE. Besides, there's no explicit unknown in TPE. So, we should be fine here.

[github-actions]

Coverage Report

Head Commit: 53cf0490e051d45da55a651e5a7ab987f2f0611f

Base Commit: 5dd810be8cf74cf71500f4c2723751a66ff51f34

Download the full coverage report.

Coverage of Added or Modified Lines of Rust Code

Required coverage: 80.00%

Actual coverage: 65.22%

Status: FAILED ❌

Details

File	Status	Covered	Coverage	Missed Lines
cedar-policy-core/src/ast/policy.rs	🟢	12/13	92.31%	601
cedar-policy/src/api.rs	🔴	3/10	30.00%	1063-1069

Coverage of All Lines of Rust Code

Required coverage: 80.00%

Actual coverage: 81.38%

Status: PASSED ✅

Details

Package	Status	Covered	Coverage	Base Coverage
cedar-policy	🔴	3052/4464	68.37%	--
cedar-policy-cli	🔴	566/976	57.99%	--
cedar-policy-core	🟢	11668/13912	83.87%	--
cedar-policy-formatter	🟢	914/1043	87.63%	--
cedar-policy-validator	🟢	8513/9944	85.61%	--
cedar-wasm	🔴	0/29	0.00%	--

[vladcebo]

Added a test for coverage 👍

[github-actions]

Coverage Report

Head Commit: c9ac0c89b26403f537c1cb03e929c4f46ef9394c

Base Commit: 62d5394e600774f672af26ebd3c51a4a8c5d3129

Download the full coverage report.

Coverage of Added or Modified Lines of Rust Code

Required coverage: 80.00%

Actual coverage: 95.24%

Status: PASSED ✅

Details

File	Status	Covered	Coverage	Missed Lines
cedar-policy-core/src/ast/policy.rs	🟢	12/13	92.31%	601
cedar-policy/src/api.rs	🟢	8/8	100.00%

Coverage of All Lines of Rust Code

Required coverage: 80.00%

Actual coverage: 81.59%

Status: PASSED ✅

Details

Package	Status	Covered	Coverage	Base Coverage
cedar-policy	🔴	3071/4467	68.75%	68.66%
cedar-policy-cli	🔴	573/976	58.71%	58.71%
cedar-policy-core	🟢	11710/13911	84.18%	84.17%
cedar-policy-formatter	🟢	914/1043	87.63%	87.63%
cedar-policy-validator	🟢	8509/9940	85.60%	85.60%
cedar-wasm	🔴	0/29	0.00%	0.00%

[adpaco-aws]

Thanks for the contribution, @vladcebo ! I'll merge it now.