test: fix repo-wide pytest failure — reconcile validate-command fixtures with the no-outputs check (BE-4097)

[mattmillerai]

ELI-5

Two recently merged PRs each work fine alone but broke the test suite together: #551 taught comfy validate to reject workflows with no output node (matching what the ComfyUI server does), and #553 added tests whose sample workflows… have no output node. Every PR's build CI has been red since. This PR fixes the two stale test fixtures so they match the intended behavior — no product code changes.

What

Repo-wide pytest has been failing on main since 09de2e5 (#553 / BE-3359) landed on top of 354113a (#551 / BE-3357): two fixtures in tests/comfy_cli/command/test_validate_command.py trip the new prompt_no_outputs server-parity check.

• test_api_format_unchanged — fixture was a bare EmptyLatentImage (not an output node), so the CLI now correctly exits 1. Replaced with the complete, server-valid sd15 txt2img graph (same graph test_engine.py::TestValidateServerParity._sd15_full already proves clean against the same sd15_object_info.json catalog), which carries the SaveImage output node every real API-format export has. The test still proves exactly what it did: an API-format file is validated directly, exit 0, no converted_from_ui key.
• test_empty_dict_payload_unchanged → test_empty_dict_payload_not_converted_and_rejected — fixture is {}; rejecting a zero-output (including node-less) prompt is the intended behavior of fix(validate): presence-check required inputs, no-outputs check, hard-error range violations (BE-3357) #551, per the check's own code comment in comfy_cli/cql/engine.py (mirroring ComfyUI server execution.py:1155-1162). Updated to assert exit 1 + prompt_no_outputs, while keeping the test's original point: the empty dict is not treated as UI format (no conversion, no converted_from_ui key). Renamed since "unchanged" no longer describes it.

Why this isn't a new dead-end

The flipped assertion (exit_code == 1 for {}) does not introduce a capability denial — it aligns a stale test with behavior already merged and intentionally designed in #551, asserted identically by the engine's own unit test (tests/comfy_cli/cql/test_engine.py::test_empty_workflow_is_no_outputs) and grounded in real server behavior (ComfyUI execution.py rejects zero-output prompts with prompt_no_outputs). The ticket calls this reconciliation "mechanical, no design call."

Verification

• uv run --extra dev pytest . — 2640 passed, 37 skipped, 0 failed (main currently fails 2)
• ruff check + ruff format --check with the CI-pinned ruff==0.15.15 — clean

Judgment calls

• Inlined the sd15 graph rather than importing _sd15_full from test_engine.py — cross-test-module imports are worse coupling than one duplicated literal fixture, and this file already inlines all its fixtures.
• Renamed the empty-dict test (ticket offered "update or rename/repurpose"); nothing references the old name.

Note: the "Windows Specific Commands" smoke workflow is red on all PRs from an unrelated pre-existing issue (unpinned mixpanel) — not addressed here, out of scope.

Overlap flag

Open PR #573 (BE-3358, feature work) incidentally bundles fixes to these same two tests. This PR is the standalone, minimal unbreaker so main's CI goes green without waiting on #573's feature review. Whichever lands second takes a small conflict in test_validate_command.py — after this merges, #573 should rebase and keep this branch's version of the two tests (they're equivalent in intent; this one also renames the empty-dict test).

[coderabbitai]

Warning

Review limit reached

You’ve reached a temporary PR review limit under our Fair Usage Limits Policy.

Your recent review volume is higher than typical usage, so adaptive limits are currently applied.

Next review available in: 11 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 1b780751-fa06-44bd-8061-e2210289ebe8

📥 Commits

Reviewing files that changed from the base of the PR and between d4ff53e and b3cd13e.

📒 Files selected for processing (1)

• tests/comfy_cli/command/test_validate_command.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch matt/be-4097-fix-validate-no-outputs-tests

✨ Simplify code

• Create PR with simplified code
• Commit simplified code in branch matt/be-4097-fix-validate-no-outputs-tests

---

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

[github-actions]

🔍 Cursor Review — Consolidated panel

Triggered by @mattmillerai.

✅ No high-signal findings.

Panel: 6/8 reviewers contributed findings.

Reviewers that did not contribute: kimi-k2.5:adversarial (empty), kimi-k2.5:edge-case (empty)

[mattmillerai]

Unreviewed admin-merge — justification

This was merged via admin override (gh pr merge --admin --squash), bypassing branch protection, with no human code review (reviewDecision was empty at merge). Recording the rationale for the audit trail.

Justification: this is a no-blast-radius test change.

• Diff is test-only: a single file, tests/comfy_cli/command/test_validate_command.py (+36 / −5). Nothing under the shipped/importable comfy_cli/ package was touched.
• No production code path affected: the change reconciles test fixtures with the validate-command no-outputs check (BE-4097); it cannot alter runtime behavior for any user of the CLI.
• All required CI passed before merge: build, ruff, CodeQL, Socket Security, codecov, and the full cursor-review adversarial/edge-case panel — all green.

Because the blast radius is limited to the test suite and CI was fully green, the unreviewed admin merge carries no user-facing or runtime risk. Merge commit: 18314a2.