Skip to content

fix(validate): skipped is not failed; per-project framework; skip templates — readiness gate + cwd flake - #317

Merged
IanFrelinger merged 1 commit into
masterfrom
fix/readiness-gate-and-cwd-flake
Aug 16, 2026
Merged

fix(validate): skipped is not failed; per-project framework; skip templates — readiness gate + cwd flake#317
IanFrelinger merged 1 commit into
masterfrom
fix/readiness-gate-and-cwd-flake

Conversation

@IanFrelinger

Copy link
Copy Markdown
Owner

Summary

Two things that were red for reasons unrelated to any recent change, both fixed at the root rather than papered over.

1. Full Platform Readiness Gate — red on every master commit (60+ runs) with the same three items

Every one was the validation adapter (nexo ci verifyValidationServiceAdapter) misreading the world, not a real regression:

ItemReal causeFix
SelfExtendInvariantDRecursionCeilingTests.Rejection_extension_past_configured_recursion_ceiling_is_refused "failed"It is a [Fact(Skip = "GAP: …")] marker → TRX outcome="NotExecuted" → parser mapped everything ≠ Passed to failedTrxTestResultParser reports Skipped=true (reason preserved); new TestResult.Skipped / ValidationResult.TestsSkipped; adapter keeps skipped out of TestsRun/TestsFailed and out of the per-test list (consumers render every !Passed as a failure); summary says "N skipped"
Nexo.Transport.A2A.Server.Tests.csproj: Test execution failedProject is net9.0-only (TestHost 8 lacks PipeWriter.UnflushedBytes); validate forced --framework net8.0 on every project → VSTest: "The argument …/net8.0/….dll is invalid" for an output never builtSelectTestFramework reads the project: multi-target → net8.0 if listed (one host, not two), else first TFM; single-target → what it declares
__BrickName__Brick.Tests.csproj: dotnet build failedsamples/templates/brick/… is a template stamped by nexo new; discovery swept it upIsDiscoverableTestProject skips __Placeholder__ names, templates/ segments, hidden segments (.git, .claude worktrees — a second checkout would double every run) and bin/obj — only below the validation root

Verified in a clean Linux container from this branch: the shipped parser over the real gap-marker TRX → executed=3 failed=0 skipped=1; A2A.Server.Tests 13/13 under --framework net9.0 --no-build (the shape validate now issues).

2. kernel-coverage flake — FileBarrierAuditSinkGapCoverageTests.WriteAsync_normalizes_blank_directory_and_prefix

A fixed Task.Delay(100) waiting for a background drain task, under a starved thread pool (coverlet on a loaded runner). Now polls for the observable effect (bounded, 5 s). Separately, every xUnit class that sets the process cwd lives in one non-parallel collection (ProcessCwd, replacing the two ad-hoc ones) so a cwd flip can't reach parallel neighbours — the sink test was bare before.

No application/** changes; the CLI's renderer needs nothing (skips never appear in its !Passed list).

Test plan

  • Nexo.Tests.Infrastructure parser + adapter + sink classes: 42/42 in a Linux SDK container (net9.0)
  • New: NotExecuted → skipped-not-failed (parser + adapter aggregation), discovery matrix (theory), SelectTestFramework matrix (theory), hidden-above-root allowed
  • CI gates green on this PR
  • After merge: Full Platform Readiness Gate on master — the failure list should be empty (its ci verify step is what these fixes target; it does not run on PRs)

🤖 Generated with Claude Code

…lares; skip templates and hidden trees; deflake cwd sink test
The Full Platform Readiness gate has been red on every master commit with the same three items.
None was a real regression; each was the validation adapter misreading the world:
- A `[Fact(Skip = "GAP: ...")]` marker (SelfExtendInvariantDRecursionCeilingTests) landed in TRX as
NotExecuted and was counted as a FAILED test. TrxTestResultParser now reports it with
Skipped=true (reason preserved); TestResult.Skipped and ValidationResult.TestsSkipped carry it;
ValidationServiceAdapter keeps skipped out of TestsRun/TestsFailed and out of the per-test list
(consumers render every !Passed result as a failure), and says "N skipped" in the summary.
- Nexo.Transport.A2A.Server.Tests is net9.0-only (TestHost 8 lacks PipeWriter.UnflushedBytes);
validate forced `--framework net8.0` on every project and VSTest answered "The argument <dll> is
invalid" for an output that was never built. SelectTestFramework now reads the project: a
multi-target project runs net8.0 when it lists it (one host, not two), else its first TFM; a
single-target project runs what it declares.
- samples/templates/brick/__BrickName__Brick.Tests is a template stamped by `nexo new`, swept up
by discovery and unbuildable in place. IsDiscoverableTestProject now skips placeholder-named
projects, `templates/` segments, hidden segments (.git, .claude worktrees — a second checkout
would double every run) and bin/obj — only below the validation root.
Separately, kernel-coverage flaked on FileBarrierAuditSinkGapCoverageTests: a fixed 100 ms wait for
a background drain task under a starved thread pool. The test now polls for the observable effect
(bounded), and every xUnit class that sets the process cwd lives in one non-parallel collection
("ProcessCwd", replacing the two ad-hoc ones) so a cwd flip cannot reach parallel neighbours.
Tests: NotExecuted -> skipped-not-failed (parser + adapter aggregation), discovery matrix (theory),
SelectTestFramework matrix (theory), hidden-above-root allowed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@cursor

cursorBot commented Aug 16, 2026

Copy link
Copy Markdown
Contributor

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.

@IanFrelinger
IanFrelinger merged commit 0ee2294 into masterAug 16, 2026
14 checks passed
@IanFrelinger
IanFrelinger deleted the fix/readiness-gate-and-cwd-flake branch August 16, 2026 06:03
IanFrelinger pushed a commit that referenced this pull request Aug 17, 2026
…rograms historical; document spikes/
docs/DocsIndex.md: Start Here #1 is the tester quickstart, README #2,
GettingStarted #3; IntegratorGuide and consumer-template/CONSUMING.md are
listed for the first time; RELEASE.md, the setup-matrix scripts and the
bootstrap helpers move out of the numbered list into an "escape hatches"
paragraph (RELEASE stays under Operator); a new "Trust loop / certification
(experimental, hold-mode)" group makes docs/trust-loop/*, the evidence
ledger, governed-pipeline, SELF-EXTEND-AUDIT, the sample objective and
spikes/ reachable; "Planning & Roadmap" becomes "Planning history
(historical as of 2026-08-16)" because it described a finished program.
Stale status docs are date-stamped rather than deleted:
- ProtocolIntegration-MCP-A2A.md said the Nexo.API wiring was "planned"; it
landed in #269/#270 (Program.cs registers all four adapters and maps
/api/mcp and /api/a2a/{agentId}; IngressCatalog rows McpServer/A2AAgents).
- ci-pre-existing-failures.md said the Full Platform Readiness Gate was RED
as of 06-21; it went green with #317-#320/#335. Kept because
docs/planning/* still cite it.
- GapAnalysis.md and NorthStarGapAnalysis.md predate the trust loop; both get
a historical banner naming their still-open rows and pointing at the
ledger, ProjectTiers and CHANGELOG.
GettingStarted.md #5 contradicted DocsIndex on "installer" scripts (the
scripts/install/* helpers exist but bootstrap the container lane); it now
says so and points first-run readers at the quickstart. The pointer to a
README section "First Successful Pipeline Run" (which does not exist) now
names the sections that do. IntegratorGuide.md recommended `dotnet build
Nexo.sln` (the docs guard forbids it in README); it now builds
Nexo.Kernel.sln and the CLI project, with hello-brick as the smallest test
example, and the compatibility matrix names the solutions that actually
apply.
spikes/README.md: the 27 tracked files under spikes/ were cited as PASS
evidence by the ledger but undocumented. It says what each spike is, every
flag of the flight runner, which ledger rows cite each spike, and that none
of it is a supported entry point.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants

@IanFrelinger@PlzTouchGrass