chore: optimise effects

[Rich-Harris]

Noticed this while reviewing recently merged PRs. @dummdidumm realised in https://github.com/sveltejs/svelte/pull/11560/files#r1597502682 that we're doing unnecessary work in a loop.

It turns out we can go further — we don't need the filter_flags & EFFECT stuff at all, because effects only contains effects with that flag. And since the only time we don't want to process effects recursively is inside flush_local_render_effects (when EFFECT effects are excluded), these two conditions are in fact synonymous:

if ((filter_flags & EFFECT) !== 0) {...}
if (!shallow) {...}

And shallow is only used negatively, so rather than doing !shallow everywhere we should save ourselves the unnecessary instruction by doing recursive instead.

The resulting code is somewhat clearer in my eyes.

[changeset-bot]

🦋 Changeset detected

Latest commit: 691145c

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

This PR includes changesets to release 1 package

Name	Type
svelte	Patch

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

[Rich-Harris]

It looks like all the filter_flags stuff is unnecessary? At the very least, removing it doesn't cause any tests to fail

[Rich-Harris]

In fact, if we remove flush_local_render_effects (which is only used inside the legacy beforeUpdate) altogether, the tests still pass. So one of two things is true:

• there are some opportunities for further simplification
• we are lacking some tests

[Rich-Harris]

Opened #11570 on top of this