fix: only call onchange once for array mutations

[Rich-Harris]

small addition to #15069 — only call onchange once for array mutations

• #15069
• this PR

This isn't just about avoiding unnecessary work — it's about correctness. Without something like this, the callback will fire when the array mutation is incomplete, which could lead to all sorts of bad things.

I thought we were already doing something like this in the context of $inspect, but it turns out we weren't. Perhaps we should?

[changeset-bot]

⚠️ No Changeset found

Latest commit: de827fb

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

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

[github-actions]

Playground

pnpm add https://pkg.pr.new/svelte@15073

[Rich-Harris]

(I didn't add any tests yet, was just noodling around. Also: maybe we would want to expose the batching function at some point, though I didn't do that preemptively)

[paoloricciuti]

(I didn't add any tests yet, was just noodling around. Also: maybe we would want to expose the batching function at some point, though I didn't do that preemptively)

I can add tests in my pr, I was waiting to find a solution for the tests...looks good to me!

[dummdidumm]

Will this have a non-trivial impact on performance? We're now doing array.includes everytime we get anything from a proxied array

[szymeo]

@dummdidumm wdyt about adding performance tests? I'm not that deep in ecosystem yet, so maybe it got addressed somehow already? If not I can have a look on poc

[paoloricciuti]

@dummdidumm wdyt about adding performance tests? I'm not that deep in ecosystem yet, so maybe it got addressed somehow already? If not I can have a look on poc

We have a series of benchmarks already that runs on CI. But i refer to @trueadm for that 😄

[trueadm]

Our benchmarks doesn't cover deep reactivity.