Catch an exit half-way on a slow machine, not only on a fast one
tests / feature (8.4) (push) Successful in 1m45s
tests / feature (8.5) (push) Successful in 2m0s
tests / browser (chrome, chromium) (push) Successful in 8m6s
tests / browser (firefox, firefox) (push) Failing after 12m52s
tests / browser (safari, webkit) (push) Failing after 13m1s

The browser suite passes on every engine here and fails on the runner,
which takes three times as long: each failing test triggers a close and
then samples for an in-between state — a rail part-way out, a scrim
part-way faded, a menu's exit copy part-way sunk — and a starved runner
takes its one sample after the 150-650ms exit has already finished.

So the tests that assert *that* something animates now stretch every
motion duration token to three seconds first (`slowMotion()`, beside
`ready()` in tests/Pest.php; ActionsTest's own copy of it goes). The two
polling helpers grew their budget to match: the rail panel and the sheet
slide on emphasized-accelerate, which is under 1% of its travel at a
quarter of the way through, so a 300ms window no longer reached the
threshold once the exit itself was three seconds long.

The two full-screen date picker tests waited for a resize through
click()'s own retry, which ate the whole 15s budget on Firefox; they now
wait for the new width and a settled document first. The bottom sheet's
preset test waits for its entry to finish before pressing the grip.

`hold()` in search.js reads the view's animations a frame after the
closed state, as the rail's settle() does, but one frame is not always
enough: an engine that starts them on its next tick shows none, and the
full-screen layout would end at once, mid-exit. An empty list is now
asked again on the following frame.

Feature 1159 passed. Browser 299 passed on Chrome, Firefox and WebKit,
and again on WebKit under ten spinning cores.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Andreas Reinhold / reini
2026-09-17 23:57:43 +02:00
co-authored by Claude Opus 5
parent 145f60aa11
commit 94d6e31432
7 changed files with 82 additions and 21 deletions
+14 -1
View File
@@ -162,15 +162,28 @@ document.addEventListener('alpine:init', () => {
this.leaving = true
requestAnimationFrame(() => {
// A frame on, the closed state has met the style and the exit's transitions exist, as
// the rail's own settle() reads them. One frame is not always enough: an engine that
// starts them on its next refresh tick would show none here, and holding for nothing
// would end the full-screen layout at once, mid-exit. So an empty list is asked again
// on the following frame before it counts as "nothing to wait for".
const hold = (frame) => requestAnimationFrame(() => {
const animations = this.$refs.view?.getAnimations() ?? []
if (animations.length === 0 && frame === 0) {
hold(1)
return
}
Promise.allSettled(animations.map((animation) => animation.finished)).then(() => {
if (leaving === this.leavings) {
this.leaving = false
}
})
})
hold(0)
},
clear() {