deps: update vitest to v3

[FedericoBiccheddu]

Type

• Refactor
• Feature
• Bug Fix
• Optimization
• Documentation Update

Description

Update vitest to latest available major version, 3.0.

Related

[FedericoBiccheddu]

Open as draft as some test are failing, one of them (test/CommandExecutor.test.ts) fails even on main and need to understand why.

[changeset-bot]

🦋 Changeset detected

Latest commit: f61cfb6

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
@effect/vitest	Minor

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[FedericoBiccheddu]

Open as draft as some test are failing, one of them (test/CommandExecutor.test.ts) fails even on main and need to understand why.

The problem was trivial: I don't have bash installed; trying with sh made the test pass!

[mikearnaldi]

wondering if this should be a major given the 3.x bump on vitest

[FedericoBiccheddu]

wondering if this should be a major given the 3.x bump on vitest

I initially thought a minor bump was ok for two reasons:

• AFAIK @effect/vitest is not stable; other packages, like platform, rpc, etc have undergone massive refactors that resulted in BC, which were expected. Unless the approach has changed or I'm missing important information about the effect team's stance on BC, I'm personally fine w/ this being a minor bump instead of a v1.*, as it aligns with the principles of other effect packages from a consumer PoV;
• the consumer API has not changed.

However, upon further reflection, @effect/vitest has a peer dep on vitest and users may have tests running on 2.x version without relying on @effect/test; this minor bump could be a BC for them hindering upgrades to other effect packages, right?

[codingismy11to7]

However, upon further reflection, @effect/vitest has a peer dep on vitest and users may have tests running on 2.x version without relying on @effect/test; this minor bump could be a BC for them hindering upgrades to other effect packages, right?

i thought about that too, but when i looked at the vitest 3.0 release notes it didn't look like a particularly onerous upgrade?

also, people who are depending on < 1.0 stuff (like myself) kinda need to be prepared for things like that, imo

[IMax153]

i thought about that too, but when i looked at the vitest 3.0 release notes it didn't look like a particularly onerous upgrade?

It doesn't matter if the upgrade is "onerous" or not - if there are breaking changes in Vitest 3.0 that will affect the external API or functionality of @effect/vitest, a minor version bump of @effect/vitest (which allows for breaking changes before there is a major) or a major version bump is most appropriate.

I don't have a strong preference and am fine with just a minor.