test: Mock spinner

[snocorp]

Description

Mock spinner allows tests to avoid sending output to stderr.

Testing

Fixed the bootstrapLocalTemplate tests and added some assertions as an example.

---

Note

Low Risk
Test-only and test-helper changes with no production runtime behavior.

Overview
Adds createMockSpinner to @sanity/cli-test: a Vitest mock factory for @sanity/cli-core/ux’s spinner that returns a full SpinnerInstance stub (with optional overrides) so CLI tests avoid real terminal output.

bootstrapLocalTemplate integration tests now mock spinner from @sanity/cli-core/ux via that helper, dynamically import the module under test after mocks are registered, and drop the old output.spinner / print / clear stubs on the fake Output. Tests assert spinner usage (e.g. initial message, three start/succeed cycles per bootstrap).

@sanity/cli-test tsconfig gains a @sanity/cli-core/* path alias so the new helper can import SpinnerInstance from the ux subpath.

^{Reviewed by Cursor Bugbot for commit 064a258. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

📦 Bundle Stats — @sanity/cli

Compared against main (01d2f80d)

@sanity/cli

Metric	Value	vs main (01d2f80)
Internal (raw)	2.7 KB	-
Internal (gzip)	1.0 KB	-
Bundled (raw)	11.16 MB	-
Bundled (gzip)	2.10 MB	-
Import time	868ms	+5ms, +0.6%

bin:sanity

Metric	Value	vs main (01d2f80)
Internal (raw)	782 B	-
Internal (gzip)	423 B	-
Bundled (raw)	9.87 MB	-
Bundled (gzip)	1.78 MB	-
Import time	2.21s	+4ms, +0.2%

🗺️ View treemap · Artifacts

Details

• Import time regressions over 10% are flagged with ⚠️
• Sizes shown as raw / gzip 🗜️. Internal bytes = own code only. Total bytes = with all dependencies. Import time = Node.js cold-start median.

📦 Bundle Stats — @sanity/cli-core

Compared against main (01d2f80d)

Metric	Value	vs main (01d2f80)
Internal (raw)	106.7 KB	-
Internal (gzip)	26.7 KB	-
Bundled (raw)	21.72 MB	-
Bundled (gzip)	3.46 MB	-
Import time	765ms	-7ms, -0.9%

🗺️ View treemap · Artifacts

Details

• Import time regressions over 10% are flagged with ⚠️
• Sizes shown as raw / gzip 🗜️. Internal bytes = own code only. Total bytes = with all dependencies. Import time = Node.js cold-start median.

📦 Bundle Stats — create-sanity

Compared against main (01d2f80d)

Metric	Value	vs main (01d2f80)
Internal (raw)	908 B	-
Internal (gzip)	483 B	-
Bundled (raw)	931 B	-
Bundled (gzip)	491 B	-
Import time	❌ ChildProcess denied: node	-

Details

• Import time regressions over 10% are flagged with ⚠️
• Sizes shown as raw / gzip 🗜️. Internal bytes = own code only. Total bytes = with all dependencies. Import time = Node.js cold-start median.

[github-actions]

Coverage Delta

No covered files changed in this PR.

Overall Coverage

Metric	Coverage
Statements	74.5% (±0%)
Branches	64.3% (±0%)
Functions	69.0% (±0%)
Lines	75.1% (±0%)

[runeb]

@snocorp Looks good, just a question. I see this is in integration tests. We do want to catch issues with spinners and for example non-interactive users like agents. Will we still be able to ensure progress indication doesnt break any of the flows we care about?

[snocorp]

@snocorp Looks good, just a question. I see this is in integration tests. We do want to catch issues with spinners and for example non-interactive users like agents. Will we still be able to ensure progress indication doesnt break any of the flows we care about?

Good question. I think this generally makes it easier to verify the spinner is used as expected. I suppose it's possible that a new version of the spinner could break something so we shouldn't mock all the places we use it. In many existing test cases the output would get swallowed by the output capture that happens when you use testCommand and we can assert the output there so that should cover enough to be safe.