test(drive-abci): gate PR runs to one comprehensive chain simulation

[QuantumExplorer]

Issue being fixed or feature implemented

The drive-abci strategy suite runs 92 whole-chain simulations on every Rust PR — 261s of test CPU, 18% of the entire workspace suite — and its 7–13s simulations set the wall-clock floor of the test phase. Most PRs get no signal from running all of them.

What was done?

New comprehensive simulation (comprehensive_tests.rs): one 30-block, ~6s run that packs every subsystem that composes deterministically:

• per-block identity creation, top-ups, key additions, credit withdrawals, credit transfers
• document create/replace/delete on dashpay, random contract creation
• token minting against a precomputed token contract
• address funding from core asset locks, address-to-address transfers, identity top-ups from address balances
• random core height increases with validator quorum rotation
• two epoch changes with masternode payouts

with verify_state_transition_results (per-transition proof verification), verify_sum_trees, and a final grovedb integrity check. Verified via the run's state-transition stats that all ten transition types actually execute (98 IdentityCreate, 207 Batch, 41 TopUp, 15 CreditTransfer, 11 Withdrawal, 12 IdentityUpdate, 9 ContractCreate, plus address ops).

Determinism trick: a single hard-coded start identity means contract deployment (block 2, before identity inserts begin at block 3) has exactly one owner candidate, so the token contract's id — and the token id the mint op signs against — is precomputable.

CI gating (tests-rs-workspace.yml): on pull_request events the nextest filter excludes binary_id(drive-abci::strategy_tests) except this test. Push, nightly, and dispatch runs keep the full suite — the same post-merge safety-net pattern as the shielded phase, so a regression in an excluded simulation surfaces minutes after merge. Filter verified with cargo nextest list: exactly 1 of 97 strategy-binary tests selected, lib/bin tests unaffected.

Latent panic fix (strategy-tests): IdentityTransfer(None) drew a recipient from an empty range when exactly one identity existed (gen_range(0..0)); the guard now requires two identities.

Not covered by the PR-path simulation (nightly/push still cover them): shielded operations (separate CI phase with its own cache), contested-resource voting, protocol upgrade forks, failure injection, and masternode list mutations.

Expected saving: the strategy tail currently bounds the test phase's 117s wall time; with it gone the tail is the 19s wallet-storage rekey test, so roughly ~60s off every Rust PR job. PR coverage on codecov will dip slightly relative to base since the excluded simulations' coverage only appears on push runs.

How Has This Been Tested?

• New simulation passes locally in 5.8s, deterministic across runs (fixed seeds).
• State-transition stats confirm every operation type executes rather than silently no-oping.
• cargo nextest list with the exact CI filter expression selects 1 strategy test + all non-strategy tests.
• cargo fmt and cargo clippy --all-features clean on both changed crates.

Breaking Changes

None.

Checklist:

• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have added or updated relevant unit/integration/functional/e2e tests
• I have added "!" to the title and described breaking changes in the corresponding section if my code contains any
• I have made corresponding changes to the documentation if needed

For repository code-owners and collaborators only

• I have assigned this pull request to a milestone

🤖 Generated with Claude Code

Summary by CodeRabbit

• Bug Fixes

  • Prevented identity-to-identity transfers from failing when only one identity is available.
  • Improved reliability of immutable-structure checks in pull request validation.

• Tests

  • Added comprehensive coverage for mixed platform operations across 30 blocks.
  • Expanded verification of state transitions, payouts, asset operations, and database consistency.
  • Improved strategy test execution and coverage reporting, including carryforward coverage for skipped areas.

[coderabbitai]

📝 Walkthrough

Walkthrough

The workflows now resolve immutable-structure baselines from the merge commit’s first parent. Workspace CI runs a comprehensive strategy test, separates coverage phases, and uploads coverage by category. Identity transfers now require at least two identities.

Changes

Strategy validation

Layer / File(s)	Summary
Merge-base immutable checks .github/workflows/tests-rs-wallet.yml, .github/workflows/tests-rs-workspace.yml	The workflows derive changed Rust files and baseline contents from the merge commit’s first parent.
Comprehensive mixed-operations strategy test packages/rs-drive-abci/tests/strategy_tests/test_cases/comprehensive_tests.rs, packages/rs-drive-abci/tests/strategy_tests/test_cases/mod.rs	A deterministic 30-block strategy test configures mixed platform operations, mock RPC responses, verification, and final state assertions.
Strategy test selection and transfer guard .github/workflows/tests-rs-workspace.yml, packages/strategy-tests/src/lib.rs	Workspace CI retains the comprehensive strategy test, runs remaining strategy tests separately on push and nightly events, and requires at least two identities for identity transfers.
Coverage reuse and uploads .github/workflows/tests-rs-workspace.yml, .codecov.yml	Coverage reuse and completion tracking now distinguish strategy and shielded phases. Codecov uploads use separate rust, rust-strategy, and rust-shielded files with carryforward enabled.

Estimated code review effort: 4 (Complex) | ~60 minutes

Sequence Diagram(s)

sequenceDiagram
participant comprehensive_mixed_operations_test
participant NetworkStrategy
participant Platform
participant MockCoreRpc
participant Grovedb
comprehensive_mixed_operations_test->>NetworkStrategy: configure mixed platform operations
NetworkStrategy->>Platform: execute 30 blocks
Platform->>MockCoreRpc: broadcast asset-unlock transactions
MockCoreRpc-->>Platform: return unknown unlock statuses
Platform->>Grovedb: verify state transitions and sum trees
Grovedb-->>comprehensive_mixed_operations_test: return zero reported issues

Loading

Possibly related PRs

• dashpay/platform#4293: Both PRs modify Rust workspace coverage reuse, carryforward, and split uploads.

Suggested reviewers:vivekgsharma, shumkov

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly describes the main change: limiting pull-request drive-abci runs to one comprehensive chain simulation.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch claude/pr-strategy-test-gate

---

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.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.04%. Comparing base (5bbd7c9) to head (c1e9a00).

Additional details and impacted files

@@ Coverage Diff @@## v4.2-dev #4297 +/- ##
============================================
- Coverage 87.61% 87.04% -0.57% 
============================================
Files 2704 2704 Lines 345206 345206 ============================================
- Hits 302443 300483 -1960 - Misses 42763 44723 +1960 

Components	Coverage Δ
dpp	88.61% <ø> (-0.23%)	⬇️
drive	85.86% <ø> (-0.40%)	⬇️
drive-abci	88.33% <ø> (-1.33%)	⬇️
sdk	∅ <ø> (∅)
dapi-client	∅ <ø> (∅)
platform-version	∅ <ø> (∅)
platform-value	92.88% <ø> (ø)
platform-wallet	∅ <ø> (∅)
drive-proof-verifier	48.02% <ø> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[QuantumExplorer]

Pushed two additions since the last review pass:

• c1e9a00e40 — codecov carryforward flags. Coverage now uploads as three flags: rust (always), rust-strategy (chain simulations, push/nightly only), rust-shielded (when the shielded phase runs). With carryforward: true in .codecov.yml, codecov reuses the base commit's coverage for any flag a PR doesn't upload, so the ~0.57% spurious project-coverage drop this PR previously showed disappears from future PR reports. This also replaces the shielded coverage cache from ci: skip shielded Rust tests on PRs without shielded changes #4293 entirely (content-hash/restore/save steps deleted) — not uploading the flag is strictly simpler than restoring a verified cached lcov, and the skip decision itself still comes from the shielded-change detector. The complete-suite tree-hash marker now requires the chain-simulation phase to have succeeded too, closing the same marker hole for the strategy suite that was previously closed for shielded.
• 839d6c99ff — immutable-structure check over git transport. The step's gh pr view call failed three consecutive runs on a runner with broken api.github.com connectivity; the changed-file list and base contents now come from git diff/git show against the PR merge commit's first parent (verified against the live merge ref that the first parent equals the base tip).

🤖 Generated with Claude Code

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In @.codecov.yml:
- Around line 9-15: Bootstrap the new Codecov flags before enabling
carryforward: update .codecov.yml for rust-strategy and rust-shielded and adjust
.github/workflows/tests-rs-workspace.yml at lines 326-328 so the migration PR
uploads both reports, or configure carryforward to use the last complete commit
containing each flag rather than the combined-only baseline.

🪄 Autofix

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: deb98897-adb3-4e92-a278-69cddc917456

📥 Commits

Reviewing files that changed from the base of the PR and between 839d6c9 and c1e9a00.

📒 Files selected for processing (3)

• .codecov.yml
• .github/workflows/tests-rs-wallet.yml
• .github/workflows/tests-rs-workspace.yml

🚧 Files skipped from review as they are similar to previous changes (1)

• .github/workflows/tests-rs-wallet.yml