Repository navigation
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
AI review status for this pull request.
|
meta-cla
Bot
added
the
CLA Signed
![astracat-bot[bot]](https://avatars.githubusercontent.com/in/4796231?s=60&v=4){kind=small}
There was a problem hiding this comment.
Approval waits on an engineering owner's decision on the draft spec:AST-094, which this diff adds as the only authority for the gesture and motion system: an engineering owner records it by commenting /approve-spec <full head SHA> on this pull request (the author may do so as an engineering owner), and should also settle whether FR4's "physics are not themable" or the new theme motion roles win. Independently of that decision, these changes are needed:
- Rebase onto
mainonce the predecessor pull requests land, so this pull request carries only its 15-file hook fix instead of 29,286 changed lines. - Build the compact-touch query from the theme's exported compact definition (
spec:AST-012FR9) instead of the hardcoded 744px width intouchPresentation.ts. - Drop the
platformadaptation condition, or amend the closed condition vocabulary ofspec:AST-012FR2 first, and if the theme motion roles stay, refuse role ids the motion tables do not define. - End an in-flight press with
touchCancelwhenusePressTrackerreplaces its tracker or element, and add null-detachment and StrictMode tests for both hooks.
[Automated review]
|
|
||
| /** Both: presentations such as a sheet in place of a popover. */ | ||
| export const COMPACT_TOUCH_QUERY = | ||
| '(width < 744px) and (pointer: coarse), (height < 500px) and (pointer: coarse)'; |
There was a problem hiding this comment.
This keeps its own width (744px). Current spec:AST-012 FR9 defines compact as width below the theme's effective md point (768 by default) or height below 500, and says no resolver keeps size numbers of its own; main now exports that definition as compactDefinition(). After rebasing, build these queries from it, or get FR9 amended first.
| * is everything else. Pair `ios` with `pointer: 'coarse'` for a touch | ||
| * profile, such as touch motion tuned per platform. | ||
| */ | ||
| platform?: 'ios' | 'other'; |
There was a problem hiding this comment.
spec:AST-012 FR2 closes the when vocabulary (width, height, pointer, contrast, motion) and says future environment fields need an explicit addition to it. A platform axis needs that amendment before it ships as a stable theme field.
| ) { | ||
| return; | ||
| } | ||
| attached?.detach(); |
There was a problem hiding this comment.
When this runs mid-press (an element swap or a changed profile option), the old tracker's dispose() clears touchId without emitting, so the consumer sees touchDown but never touchUpInside, touchUpOutside or touchCancel. Ending a tracked press with touchCancel before disposing (the unmount flag already silences it on unmount) would keep the control-event stream balanced.
| `${location}: a motion role is named by its table path, such as "sheet.present" or "springs.navigationPush".`, | ||
| ); | ||
| } | ||
| const registered = motionRole(id); |
There was a problem hiding this comment.
motionRole(id) only finds roles whose table module was imported, and no theme module imports one, so in defineTheme this check never fires; together with the shape-only isMotionRoleId, a misspelled role ('springs.sheetPresnt') is accepted and writes a variable nothing reads. Validating against a static list of the tables' roles would turn that into an error.
PR Analysis Report📚 Storybook Preview🧪 Sandbox PreviewComponent Size
Sizes are gzip. Import counts the JavaScript the import loads eagerly; CSS counts the package stylesheet rules referenced by the modules it can load from its own package, lazily loaded ones included. Bundle Size SummaryGzip figures measure each package's entry file, not the code an import pulls in.
Accessibility AuditStatus: No accessibility violations detected. Visual Regression55 of 502 shot(s) changed. A change here is a question, not a failure: check whether the after is the
and 35 more. Generated by PR Enrichment workflow | Storybook | Sandbox | View full report |
37ce331 to
7cedc85
Compare
github-actions
Bot
added
the
needs:design-review
There was a problem hiding this comment.
Approval waits on engineering-owner decisions: the draft spec:AST-094, which this diff adds as the only authority for the gesture and motion system, needs an engineering owner to record a decision by commenting /approve-spec <full head SHA> on this pull request (the author may do so as an engineering owner), including whether AST-094 FR4 and DEC-1 or the new theme motion roles win, and the new stable @astryxdesign/core/motion/reducedMotion.stylex subpath needs an engineering owner to name its governing spec, or it should move behind the experimental boundary. Independently of those decisions, these changes are needed:
- Rebase onto
mainonce the predecessor pull requests land, so this pull request carries only its 15-file hook fix instead of 31,705 authored lines. - Keep a
usePressTrackerpress alive when its tracker is rebuilt (an element swap, an option change, or a new ref object on a re-render), or end it withtouchCancel, and add re-render, null-detachment and StrictMode tests for both hooks. - Keep the text-selection and iOS callout lock out of the stable press controller that Link, Button and other components use, or file it as a stable change, since an
[experimental]changeset cannot change stable behaviour (spec:AST-017FR22, FR27). - Drop the
platformadaptation condition or amendspec:AST-012FR2 first, read the theme's effectivemdpoint in the touch resolver (FR9), and correct the motion-roles and experimental-subpaths changesets.
[Automated review]
| @@ -283,6 +301,9 @@ function onPointerDown(event: PointerEvent): void { | |||
| gesture: idlePressGesture(), | |||
| pointerId: event.pointerId, | |||
| timer: null, | |||
| // A held finger never starts text selection on the surface (UIKit's | |||
| // controls and cells never do), on any coarse pointer. | |||
| unlock: lockSelection([element], event.target), | |||
There was a problem hiding this comment.
This makes the stable press controller set -webkit-touch-callout, -webkit-user-select and user-select to none on every usePressFeedback surface (Link, Button, Item, ClickableCard, …) for any touch or pen press, so on iOS a long press on an Astryx Link no longer offers the link callout, and text in a pressable row can no longer be selected with a long press. That is a stable runtime change, which spec:AST-017 FR22 and FR27 keep out of an [experimental] changeset. Keep the lock to the experimental gesture host's press-tracked controls, or file it as a stable change under owner-approved authority.
| @@ -177,6 +190,13 @@ const STATIC_EXPORTS = { | |||
| types: './dist/utils/interactionOverlay.stylex.d.ts', | |||
| default: './dist/utils/interactionOverlay.stylex.js', | |||
| }, | |||
| // Proposed; pending owner review (spec:AST-094 OQ4): the Reduce Motion | |||
There was a problem hiding this comment.
AST-094 at this head has only OQ1 and OQ2, so this OQ4 citation points nowhere, and ./motion/reducedMotion.stylex ships as a stable public subpath (no /experimental, no @experimental marker, under a [fix] changeset). Either give it a governing record an engineering owner approves, or move it behind the experimental boundary (spec:AST-017 FR21, FR27) or keep it internal to core.
| }, | ||
| }), | ||
| [ | ||
| ref, |
There was a problem hiding this comment.
With ref in these dependencies, a caller that passes a new ref object on each render (the case the new "a new ref object" test covers) gets a new tracker on every render. If it paints from onHighlightChange with setState, the re-render after touchDown disposes the tracker that owns the touch, the arena hands the lift to that disposed tracker, and no touchUpInside arrives. Keying the tracker on the element (or updating one tracker in place) and ending a tracked press with touchCancel before disposal would keep taps working; a re-render-mid-press test would catch it.
| '@astryxdesign/cli': patch | ||
| --- | ||
|
|
||
| [feat] Themes can retune touch motion roles. `defineTheme({motion})` takes `springs`, `durations` and `curves` keyed by motion role, the table path of a measured touch motion value (`springs.sheetPresent`, `timing.reducedMotionCrossFadeMs`), at the root or in an adaptation rule, and writes one `--motion-*` variable per role it sets. A role the theme leaves out keeps its measured default, an unknown role id is refused, and the duration scale (`fast`, `medium`, `ratio`) is optional when a theme sets only roles. No shipped component reads the roles yet. Adaptation rules gain a `platform` condition: `ios` for the browsers of iPhone and iPad, or `other`. `astryx docs motion` lists every motion role with its CSS variable and default, and the theme authoring guide and template describe the new fields. |
There was a problem hiding this comment.
timing.reducedMotionCrossFadeMs is not in MOTION_ROLE_KINDS at this head (reduced-motion-record.md says it is no longer a theme role), so a theme that follows this note throws "is not a motion role". Name a role the generated list has, such as navigation.transitionMs.
7cedc85 to
79fbdd8
Compare
… other native-scroll cases
There was a problem hiding this comment.
Holding for an engineering-owner decision on actionGroup: it lets an action group's recognizers take mouse, pen and finger presses where the primary pointer is fine, which spec:AST-094 FR19 and architecture:react-component-runtime INV13 currently rule out, and the family:action-groups FR5 it cites is not on main yet (#8622). Approval needs an engineering owner to provide or identify the governing spec for that addition, either by landing #8622 with an FR19/INV13 exception or by commenting /approve-spec <full head SHA> on this pull request; the author may do so as an engineering owner. Independently of that decision, please land #8309 first or rebase onto it, so this pull request carries only its own hook change (about 1,400 lines) instead of the whole stack (about 20,200), and correct the gestures README and barrel, which still send Lab through workspace imports that INV13 and the stack's own Lab-import guard rule out.
[Automated review]
| * primary pointer is fine too (`AttachOptions.actionGroup`). | ||
| * @default false | ||
| */ | ||
| actionGroup?: boolean; |
There was a problem hiding this comment.
actionGroup turns on recognizers and listeners where the primary pointer is fine, which spec:AST-094 FR19 and architecture:react-component-runtime INV13 currently rule out, and the family:action-groups FR5 it cites is still proposed in #8622. It needs that record (with the exception) or an engineering owner's decision before it can merge.
| A gesture recognizer system with UIKit's semantics, for the touch | ||
| presentations of Astryx core components. Components import it from | ||
| `../gestures`, and Lab through internal workspace imports; nothing here is | ||
| exported from the package. |
There was a problem hiding this comment.
architecture:react-component-runtime INV13 says Core and Lab reach these modules through @astryxdesign/core/internal/touch, and scripts/lab-core-imports.test.mjs fails any Lab import around core's exports, so "Lab through internal workspace imports" points Lab authors at an import the guard rejects. The barrel's header (index.ts:7-8) says the same.
An attach reads the element's own will-change in a task of its own. A shared animator (animatorFor) is never detached, so its node can leave the document with that read still waiting, and the read then forces a style pass on a node with no style: a test run that mounts and unmounts many pagers piled up such reads until a later wait timed out. The read now skips a node out of the document; a promotion that needs it reads it then, as one on a node attached in the same task does. A detach and dispose already cancel it. The new test fails on the previous code (one read of a removed node); the in-task read test now puts its element in the document, where the read happens.
There was a problem hiding this comment.
Holding for an engineering-owner decision on actionGroup: it lets an action group's recognizers take mouse, pen and finger presses where the primary pointer is fine, which spec:AST-094 FR19 and architecture:react-component-runtime INV13 currently rule out, and the family:action-groups FR5 it cites is not on main yet (#8622). Approval needs an engineering owner to provide or identify the governing spec for that addition, either by landing #8622 with an FR19/INV13 exception or by commenting /approve-spec <full head SHA> on this pull request; the author may do so as an engineering owner. Independently of that decision:
- Land #8309 first or rebase onto it, so this pull request carries only its own hook change (about 1,400 lines) instead of the whole stack (about 19,700).
- If
actionGroupstays, keep an element a group when a second hook on it leaves the option out; today the last attachment wins (GestureHost.ts:382). - Correct the gestures README and barrel, which still send Lab through workspace imports that INV13 and the stack's own Lab-import guard rule out.
[Automated review]
| ): () => void { | ||
| const registration = this.registrationFor(element); | ||
| registration.manageStyle = options.manageStyle ?? registration.manageStyle; | ||
| registration.actionGroup = options.actionGroup ?? registration.actionGroup; |
There was a problem hiding this comment.
actionGroup is stored once per element, but each attachment sets it: useGestureRecognizer always passes actionGroup (default false), so a second hook on the same element that leaves the option out overwrites a group's true, and nothing recomputes it on detach. Under a fine pointer the group then stops serving presses, depending only on hook order. If the option stays, pass it only when the caller sets it and keep the element a group while any attachment asks for it.
| A gesture recognizer system with UIKit's semantics, for the touch | ||
| presentations of Astryx core components. Components import it from | ||
| `../gestures`, and Lab through internal workspace imports; nothing here is | ||
| exported from the package. |
There was a problem hiding this comment.
architecture:react-component-runtime INV13 routes Lab to these modules through core's @astryxdesign/core/internal/touch subpath, and scripts/lab-core-imports.test.mjs fails any Lab import of core that core's exports do not list, so "Lab through internal workspace imports" points Lab authors at an import the guard rejects. The barrel header (index.ts:7-8) says the same.
…of the document) into meta#8270
…sture system (spec:AST-094 FR24, FR43) utils/pressFeedback.ts reaches gestures/coarsePointer.ts, so a page whose only touch work is press feedback carries a gesture module, and a bundle check that counts gestures/ code finds it in the root and hooks entries. The coarse-pointer query belongs with the environment queries. This test walks the controller's import graph and fails until it reaches no module under gestures/.
…ries in utils/ (spec:AST-094 FR43) gestures/coarsePointer.ts becomes utils/coarsePointer.ts, beside utils/touchQueries.ts, with its test; the gesture host's selection lock, the press controller and the pointer test helper import it from there. The press controller's import graph now reaches no module under gestures/, so a page whose only touch work is press feedback carries none of the gesture system. No behavior changes.
There was a problem hiding this comment.
Holding for an engineering-owner decision on actionGroup: it lets an action group's recognizers take mouse, pen and finger presses where the primary pointer is fine, which spec:AST-094 FR19 and architecture:react-component-runtime INV13 currently rule out, and the family:action-groups FR5 it cites is not on main yet (#8622). Approval needs an engineering owner to provide or identify the governing spec for that addition, either by landing #8622 with an FR19/INV13 exception or by commenting /approve-spec <full head SHA> on this pull request; the author may do so as an engineering owner. Independently of that decision:
- Land #8309 first or rebase onto it, so this pull request carries only its own hook change (about 1,400 lines) instead of the whole stack (about 19,800).
- If
actionGroupstays, keep an element a group while any of its attachments asks for it; today hook order decides, and the group outlives the hook that asked (GestureHost.ts:382). - Correct the gestures README and barrel, which send Lab through workspace imports that INV13 and the stack's own Lab-import guard rule out.
[Automated review]
| * primary pointer is fine too (`AttachOptions.actionGroup`). | ||
| * @default false | ||
| */ | ||
| actionGroup?: boolean; |
There was a problem hiding this comment.
family:action-groups FR5 is not on main yet (#8622), and current spec:AST-094 FR19 and architecture:react-component-runtime INV13 say that no recognizer attaches where the primary pointer is fine. This option turns on exactly that path (mouse, pen and finger presses served to the group's recognizers, claimed clicks swallowed), so it needs an engineering owner's decision before it lands.
| ): () => void { | ||
| const registration = this.registrationFor(element); | ||
| registration.manageStyle = options.manageStyle ?? registration.manageStyle; | ||
| registration.actionGroup = options.actionGroup ?? registration.actionGroup; |
There was a problem hiding this comment.
actionGroup lives on the element's one registration, but every attachment overwrites it: useGestureRecognizer always passes actionGroup (default false), and nothing recomputes it on detach. Under a fine pointer, a group whose actionGroup hook attaches before another hook on the same element stops taking presses, and one whose hook attaches last keeps serving the other recognizers after that hook detaches. Passing the option only when the caller sets it, and keeping the element a group while any of its attachments asks for it (recomputed on detach), would fix both.
|
|
||
| A gesture recognizer system with UIKit's semantics, for the touch | ||
| presentations of Astryx core components. Components import it from | ||
| `../gestures`, and Lab through internal workspace imports; nothing here is |
There was a problem hiding this comment.
architecture:react-component-runtime INV13 routes Lab to these modules through core's @astryxdesign/core/internal/touch subpath, and scripts/lab-core-imports.test.mjs fails any Lab import that goes around core's exports, so “internal workspace imports” points Lab authors at an import the guard rejects. The barrel's header (index.ts:7-8) says the same.
| * @input The source of utils/pressFeedback.ts and every module it imports | ||
| * @output The press feedback controller's import graph reaches no module | ||
| * under gestures/: a page whose only touch work is press feedback carries | ||
| * none of the gesture system (`spec:AST-094` FR24, FR43: pointer queries |
There was a problem hiding this comment.
Current spec:AST-094 has no FR43 (its requirements end at FR41). Citing FR24 alone, or the record that states where pointer queries live, keeps this header accurate.
…(spec:AST-094 FR18, DEC-13) A screen-edge pan attached to a stack container, as wide as the screen, with a touch 4 px from its edge: inside the 40.2 px band the pan decides at UIKit's 15 px whatever the element's width, and then tracks the travel past it. From several edges too. Both fail while a wider element decides at FR19's 10 px.
…'s width (spec:AST-094 FR18, DEC-13) FR18: inside the 40.2 px band a screen-edge pan decides at UIKit's 15 px screen-edge hysteresis; outside it the author's scrolling is untouched and FR19's 10 px governs every other claim. The pan decided at 10 px on any element wider than the band, so NavigationStack's edge pan, attached to the stack container, began 5 px early. The edge zone still decides only the touch-action; a touch outside every band still fails at once. The two cases that pinned the 10 px decision go.
…er's move keeps that finger until it lifts) into the experimental touch exports
…than the band (spec:AST-094 FR18, DEC-22) FR18 as amended in meta#8631 (DEC-22): the band's pan-x claim and 15 px start hold only where the screen-edge pan is attached to an element within the band; on an element that reaches beyond it, whose touch-action cannot claim per touch, a touch in the band decides at FR19's 10 px like every other claim. This restores the 10 px decision 3d85c30 removed and keeps 15 px on an element within the band. The cases pin DEC-22 both ways (a wide element with a touch at x=4 begins at 10, an in-band strip at 15, and from several edges at 10 on a wide element); the 10 px cases fail on 3d85c30.
There was a problem hiding this comment.
Holding for an engineering-owner decision on actionGroup: it lets an action group's recognizers take mouse, pen and finger presses where the primary pointer is fine, which spec:AST-094 FR19 and architecture:react-component-runtime INV13 currently rule out, and the family:action-groups FR5 it cites is not on main yet (#8622). Approval needs an engineering owner to provide or identify the governing spec for that addition, either by landing #8622 with an FR19/INV13 exception or by commenting /approve-spec <full head SHA> on this pull request; the author may do so as an engineering owner. Independently of that decision:
- Land #8309 first or rebase onto it, so this pull request carries only its own hook change (about 1,400 lines) instead of the whole stack (about 19,900).
- If
actionGroupstays, keep an element a group while any of its attachments asks for it; today hook order decides, and the group outlives the hook that asked (GestureHost.ts:382). - Correct the gestures README and barrel, which send Lab through workspace imports that INV13 and the stack's own Lab-import guard rule out, and the FR43 citation in
pressFeedbackImports.test.ts, which current AST-094 does not have.
[Automated review]
| ): () => void { | ||
| const registration = this.registrationFor(element); | ||
| registration.manageStyle = options.manageStyle ?? registration.manageStyle; | ||
| registration.actionGroup = options.actionGroup ?? registration.actionGroup; |
There was a problem hiding this comment.
useGestureRecognizer always passes actionGroup (default false, useGestureRecognizer.ts:109, :154), so this ?? lets the last attachment decide whether the element is a group, and the detach closure (:395-415) never recomputes it. With two hooks on one element and only one asking for a group, a fine-pointer press either reaches no recognizer or reaches the one that never asked, and keeps doing so after the group hook detaches, depending only on hook order. If the option stays: pass it only when the caller sets it, and keep the element a group while any attachment asks for it, recomputed on detach.
|
|
||
| A gesture recognizer system with UIKit's semantics, for the touch | ||
| presentations of Astryx core components. Components import it from | ||
| `../gestures`, and Lab through internal workspace imports; nothing here is |
There was a problem hiding this comment.
architecture:react-component-runtime INV13 says Core and Lab reach these modules through @astryxdesign/core/internal/touch, and scripts/lab-core-imports.test.mjs fails a Lab import of core outside core's exports, which has no ./internal/* subpath yet. Suggest saying Lab reaches them only through that subpath once it exists; index.ts:7-8 repeats the same route.
| * @input The source of utils/pressFeedback.ts and every module it imports | ||
| * @output The press feedback controller's import graph reaches no module | ||
| * under gestures/: a page whose only touch work is press feedback carries | ||
| * none of the gesture system (`spec:AST-094` FR24, FR43: pointer queries |
There was a problem hiding this comment.
Current spec:AST-094 has no FR43 (its requirements end at FR41). Cite FR24 alone, or the record that states where pointer queries live.
…xperimental touch exports The motion README's Files table is the union: main's rows, with meta#8663's resolveSnapPoint.ts and selectPage.ts, plus this branch's, in main's order. # Conflicts: # packages/core/src/motion/README.md
…on/README.md) into facebook#8309
There was a problem hiding this comment.
Holding for an engineering-owner decision on actionGroup: it lets an action group's recognizers take mouse, pen and finger presses where the primary pointer is fine, which spec:AST-094 FR19 and architecture:react-component-runtime INV13 currently rule out, and the family:action-groups FR5 it cites is not on main yet (#8622). Approval needs an engineering owner to provide or identify the governing spec for that addition, either by landing #8622 with an FR19/INV13 exception or by commenting /approve-spec <full head SHA> on this pull request; the author may do so as an engineering owner. Independently of that decision:
- Land #8309 first or rebase onto it, so this pull request carries only its own hook change (about 1,400 lines) instead of the whole stack (about 19,900).
- If
actionGroupstays, keep an element a group while any of its attachments asks for it; today hook order decides, and the group outlives the hook that asked (GestureHost.ts:382). - Point Lab at
@astryxdesign/core/internal/touchin the gestures README and barrel, as INV13 and the stack's own Lab-import guard require, and fix the FR43 citation inpressFeedbackImports.test.ts, which current AST-094 does not have.
[Automated review]
| * primary pointer is fine too (`AttachOptions.actionGroup`). | ||
| * @default false | ||
| */ | ||
| actionGroup?: boolean; |
There was a problem hiding this comment.
family:action-groups FR5 is not on main yet (#8622), and current spec:AST-094 FR19 and architecture:react-component-runtime INV13 say that where the primary pointer is fine no recognizer attaches and no touch listener is registered. With actionGroup: true the host serves mouse, pen and finger presses to the group's recognizers there, so this needs an engineering owner's decision (landing #8622 with an FR19/INV13 exception, or /approve-spec <full head SHA> on this pull request) before it lands.
| ): () => void { | ||
| const registration = this.registrationFor(element); | ||
| registration.manageStyle = options.manageStyle ?? registration.manageStyle; | ||
| registration.actionGroup = options.actionGroup ?? registration.actionGroup; |
There was a problem hiding this comment.
useGestureRecognizer always passes actionGroup (default false, useGestureRecognizer.ts:109, :154), so this ?? lets the last attachment decide whether the element is a group, and the detach closure (:395-415) never recomputes it. With two hooks on one element under a fine pointer, attach order decides whether a mouse press reaches any recognizer, and a recognizer that never asked keeps taking presses after the group hook's attachment is released. Keeping the element a group while any of its attachments asks for it (recomputed on detach), and passing the option only when the caller sets it, would fix both.
|
|
||
| A gesture recognizer system with UIKit's semantics, for the touch | ||
| presentations of Astryx core components. Components import it from | ||
| `../gestures`, and Lab through internal workspace imports; nothing here is |
There was a problem hiding this comment.
INV13 says Core and Lab reach these modules through @astryxdesign/core/internal/touch, which main now ships (packages/core/src/internal/touch.ts re-exports the arena, the recognizer base, tap and long press from this directory), and scripts/lab-core-imports.test.mjs fails a Lab import of core outside core's exports. Naming that entry here, and in index.ts's header, keeps Lab authors on the route the guard allows; “nothing here is exported from the package” is no longer accurate either.
| * @input The source of utils/pressFeedback.ts and every module it imports | ||
| * @output The press feedback controller's import graph reaches no module | ||
| * under gestures/: a page whose only touch work is press feedback carries | ||
| * none of the gesture system (`spec:AST-094` FR24, FR43: pointer queries |
There was a problem hiding this comment.
Current spec:AST-094 has no FR43 (its requirements end at FR42). Cite FR24 alone, or the record that states where pointer queries live.
|
Closing: carried by #8360 (this head, 2e77071, is in it by hash). The review items:
|
Depends on #8309. Review only this pull request's change: vjeux/astryx@58dbb1a...2e77071
Why
Astryx's gesture hooks called a component's handlers at the wrong times. A recognizer unmounted mid-gesture reported
cancelledfrom React's cleanup, and a press tracker's timer could call back after unmount, so a component could set state on something already gone. A touch that stopped a flicking list still counted as a press by default, and a component that changed a press tracker's options after mount was ignored.What led to it
This is part of the UIKit-parity work on Astryx's touch components. Reviews of #8073, #8076, a context menu and a number input found these. React's rules for Astryx say a component's callback never fires from an Effect or from cleanup (
architecture:react-component-runtimeINV2, INV3).Options and tradeoffs
Why this choice
The hooks are the only place every component passes through, so one fix covers all of them, and a press in progress survives an option change. Each recognizer's internal plumbing becomes
protected, so components can no longer call it by accident; the arena and the host reach it through one internal accessor. A press stopping a flick no longer presses, because the brake now defaults to the host's.How to check it
Evidence