fix(authz): reuse policy and entity resolution within bulk requests

[strantalis]

Proposed Changes

Repeated entries in GetDecisionBulk fetched and compiled the same policy and resolved the same entity again. Add bounded memo tables scoped to one sequential bulk RPC: eight policy sets and 32 successful entity resolutions.

Policy keys normalize and deduplicate FQNs. Resolution keys include the complete token or entity-chain request, resource context, and environment-entity mode. Separate RPCs get separate tables; failures are not cached. This does not introduce parallel evaluation.

Layer 5 of 7 in the authorization performance stack. Review and merge from the bottom upward.

Checklist

• Added or updated unit or integration coverage appropriate to this layer
• Ran focused verification for this layer
• Documented behavior and validation limits below

Testing Instructions

• Race tests verify repeated policy/entity reuse, capacity limits, fresh state for a new RPC, and cache misses when tokens, resource context, or environment-entity mode change.
• Access v2 and authorization v2 focused tests passed.

Stack-wide validation:

• Focused race tests, the full verbose policy integration suite with the race detector, SDK README code-block tests, formatting, and service lint restricted to changes since the stack base passed with zero new issues.
• make test was attempted with Colima configured. The round-trip suite failed because its required platform server at 127.0.0.1:8080 was not running.
• make lint stopped because the configured Buf API token is invalid. Separate make go-lint reported existing repository issues; the formatting issue in the new hierarchy test was corrected.
• make govulncheck reported vulnerabilities in the existing dependencies and the installed Go 1.26.3 standard library. This stack changes no dependency versions.
• GitHub CI is still running. Keep this stack in draft pending review and CI results.

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
  auto_review:
    drafts: true

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

Benchmark results, click to expand

Benchmark authorization.GetDecisions Results:

Metric	Value
Approved Decision Requests	1000
Denied Decision Requests	0
Total Time	220.906729ms

Benchmark authorization.v2.GetMultiResourceDecision Results:

Metric	Value
Approved Decision Requests	1000
Denied Decision Requests	0
Total Time	115.681097ms

Benchmark Statistics

Name	№ Requests	Avg Duration	Min Duration	Max Duration

Bulk Benchmark Results

Metric	Value
Total Decrypts	100
Successful Decrypts	100
Failed Decrypts	0
Total Time	405.41829ms
Throughput	246.66 requests/second

TDF3 Benchmark Results:

Metric	Value
Total Requests	5000
Successful Requests	5000
Failed Requests	0
Concurrent Requests	50
Total Time	41.572117428s
Average Latency	414.538408ms
Throughput	120.27 requests/second

[github-actions]

⚠️ Govulncheck found vulnerabilities ⚠️

The following modules have known vulnerabilities:

• otdfctl
• service
• tests-bdd

See the workflow run for details.

[github-actions]

X-Test Results

✅ go@main-main
bats-test-results
✅ go@pull-3990-v0.9.0
✅ go@main-v0.9.0
✅ java@main-main
✅ go@pull-3990-main
✅ go@pull-3990-pull-3990
cukes-report
✅ java@main-pull-3990
✅ go@main-pull-3990
✅ js@main-main
✅ java@main-v0.9.0
✅ js@main-v0.9.0
✅ js@main-pull-3990
opentdfplatformR34QSR.dockerbuild
govulncheck-failure-3
govulncheck-failure-8
govulncheck-failure-1