Fixes 3 different bugs when making links to and from subgraph IO from
vue nodes
- When dragging a link from a node to a subgraph IO, there is no
feedback if a slot is not a valid connection target or if a slot is
actively hovered
- When a link is made from a subgraph IO to a node, the reactivity is
not triggered on the node to indicate a change of link state.
- When dragging a link from a subgraph IO to a node, the link would not
snap to the valid connection targets on nodes
- The fix for this one is not as thorough as I would like. It only
allows connections to the slot, not connections to the hovered widget.
We have two deeply disconnected linking systems and properly reconciling
them would be a multi-week project.
Resolves FE-561
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12281-Subgraph-io-fixes-3606d73d365081089f7ef19331c6d70a)
by [Unito](https://www.unito.io)
This is a targeted small scope change to improve the availability for
converting group nodes into a subgraph.
The prior implementation would only apply on the litegraph context menu
option for converting a node to a subgraph. It failed to apply on any of
the other more common methods. The code for unpacking group nodes has
been moved directly into the setup for converting a group of nodes into
a subgraph and drastically simplified.
Of note, several other long lived bugs were found while working on this
fix, but they are out of scope for this targeted PR.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12356-On-subgraph-conversion-always-unpack-group-nodes-3666d73d365081d09774c00a851b8198)
by [Unito](https://www.unito.io)
## Summary
Import published media assets for shared workflows before loading the
graph so the first missing-media scan sees the user's newly imported
references instead of surfacing a false missing asset error. cc FE-773
## Changes
- **What**: Moves the shared workflow import step ahead of
`loadGraphData` for the copy-and-open flow, while still allowing the
workflow to open with a warning path if asset import fails.
- **What**: Clears the shared workflow URL intent consistently on
failure paths, including graph load failure after an import attempt, so
reloads do not repeatedly replay the same shared workflow side effects.
- **What**: Invalidates the input asset cache after published asset
import so graph loading and missing-media resolution can observe the
refreshed media state.
- **What**: Adds a global loading spinner while shared workflow asset
import and graph load are in progress, with `role="status"`,
`aria-live`, reduced-motion-safe animation, and body teleporting so it
stays visible above blocking UI.
- **What**: Adds stable TestIds for the shared workflow dialog and
updates existing shared workflow E2E selectors away from copy-dependent
role text.
- **What**: Adds a cloud E2E regression fixture and spec covering the
critical flow: shared URL opens the dialog, the user confirms asset
import, published media is imported before the public-inclusive input
asset scan, the workflow loads, the share query is removed, and missing
media UI is not surfaced.
- **Breaking**: None.
- **Dependencies**: None.
## Root Cause
Shared workflow graph loading triggered the missing-media pipeline
before the user-selected published media import had completed. Because
`include_public=true` does not include published assets, the pre-import
scan could classify shared media as missing even when the user was about
to import those assets into their own library.
## Review Focus
- The ordering in `useSharedWorkflowUrlLoader`: import published assets
first, then load the graph, while keeping import failure non-fatal for
workflow opening.
- The failure cleanup behavior: the shared URL/preserved query intent is
now cleared for graph load failures too, avoiding repeated
reload-triggered imports.
- The spinner behavior in `App.vue`: it uses the existing
`workspaceStore.spinner` boolean and intentionally keeps broader
ref-counted spinner ownership as follow-up work.
- The E2E sentinel in `sharedWorkflowMissingMedia.spec.ts`: it asserts
no public-inclusive input asset scan occurs before `/api/assets/import`,
then waits for a settling window to ensure the missing-media overlay
does not appear.
## Validation
- `pnpm format`
- `pnpm lint` (passed with existing unrelated warnings only)
- `pnpm typecheck`
- `pnpm test:unit`
- Commit hook: lint-staged formatting/linting, `pnpm typecheck`, `pnpm
typecheck:browser`
- Push hook: `pnpm knip --cache` (passed with existing tag hint only)
## Follow-Up
- Consider a ref-counted or scoped global spinner API so long-running
flows do not directly toggle `workspaceStore.spinner`.
- Consider separating shared workflow load status into orthogonal result
fields instead of encoding partial success in a single string union.
- Consider moving published asset import/cache invalidation behind an
asset-service-owned API boundary.
- Backend follow-up remains needed for `include_public=true` not
including published assets; this PR only removes the frontend false
positive when the user explicitly imports the shared media.
## Screenshots
Before
https://github.com/user-attachments/assets/dc790046-237c-4dd8-b773-2507f9a66650
After
https://github.com/user-attachments/assets/6517cd38-2c3d-4bfe-a990-35892b7e50aehttps://github.com/user-attachments/assets/d89dc3d3-75d9-4251-998b-0c354414e25b
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12333-fix-avoid-false-missing-media-errors-after-importing-shared-workflow-assets-3656d73d365081b38634dcb7625cfc32)
by [Unito](https://www.unito.io)
## Summary
Avoid a second `comfyPage.setup()` in the job history browser tests by
registering the initial jobs route mocks before the normal page boot.
## Changes
- **What**: Adds an `initialJobsScenario` Playwright option with an auto
fixture for the job-history spec, moves QPOV2 setup into
`initialSettings`, and keeps the sidebar helper focused on UI
navigation.
- **Dependencies**: None
## Review Focus
Confirm the auto fixture ordering matches the intended browser-test
setup: initial `/api/jobs` mocks should be installed before the
`comfyPage` fixture performs its normal setup.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12324-test-avoid-job-history-double-setup-3656d73d365081778e24c11d3b65cbef)
by [Unito](https://www.unito.io)
## Summary
Stops the Shape submenu (and any other PrimeVue nested submenu) from
being clipped behind the node context menu when the menu fits in the
viewport.
## Changes
- **What**: `constrainMenuHeight` in `NodeContextMenu.vue` now applies
`max-height` + `overflow-y: auto` to the root `<ul>` only when
`scrollHeight > availableHeight`. The common case keeps `overflow:
visible`.
- Added `browser_tests/tests/nodeContextMenuShapeSubmenu.spec.ts`
regression spec.
## Review Focus
Root cause: setting only `overflow-y: auto` on a `<ul>` coerces
`overflow-x` to a non-visible value per CSS spec (`If one of
overflow-x/overflow-y is visible and the other isn't, the visible value
is computed as auto`). PrimeVue `ContextMenuSub` renders submenus
in-tree as a nested `<ul>` with `position: absolute; left: 100%`, so the
implicit horizontal clip hides them entirely.
The pre-existing overflow scenario (#10824 / #10854) is unchanged — when
the menu actually overflows, the clamp still applies and
`nodeContextMenuOverflow.spec.ts` continues to verify scroll. Submenu
clipping in that overflow case is a known limitation, not introduced by
this PR.
Fixes FE-570
## screenshot
### AS IS
<img width="788" height="505" alt="Screenshot 2026-05-07 at 12 43 26 PM"
src="https://github.com/user-attachments/assets/36d34070-0c57-4385-a130-0394f22f282e"
/>
### TO BE
<img width="779" height="627" alt="Screenshot 2026-05-07 at 12 42 44 PM"
src="https://github.com/user-attachments/assets/00956729-763b-4787-822f-209e8ea42331"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12035-fix-keep-node-context-menu-overflow-visible-when-content-fits-3586d73d365081ad9aaec82f220d401c)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
When the built-in logs terminal stayed open during a backend restart,
the buffer froze on pre-restart entries and live log streaming silently
stopped — only closing and reopening the panel resynced. Listen for the
api `reconnected` event and rebuild the terminal contents the same way a
fresh open would.
## Changes
- **What**:
- Extract `useLogsTerminal` composable. The SFC is now a thin shell
holding `terminal: shallowRef<Terminal>` and forwarding to the
composable, so `onMounted`/`onScopeDispose` no longer rely on the
child's emit callback timing.
- Subscribe to `api`'s `reconnected` event via `useEventListener`,
registered synchronously before any awaits. On reconnect:
`terminal.reset()` → refetch raw logs → `scrollToBottom()` →
`subscribeLogs(true)` (the backend loses the per-client subscription on
restart, so re-subscribe is required for live streaming to resume).
- Wrap in-flight resync/mount fetches in AbortControllers. Overlapping
reconnects abort the prior resync, and unmount mid-fetch suppresses
writes to the disposed xterm.
- Hide BaseTerminal whenever `errorMessage` is set so the error layout
doesn't expose an empty xterm container behind the message;
`loading=false` after both load failure and resync success so a later
successful reconnect can clear a stuck spinner.
- Migrate the load/resync error strings to vue-i18n
(`logsTerminal.loadError`, `logsTerminal.resyncError`).
## Review Focus
- **Re-subscribe is the non-obvious half of the fix** — without it, even
after the WebSocket reconnects the backend never resumes streaming logs
to this client because its subscription state was wiped on restart. The
visible "stale buffer" is only one symptom; the silent "no new logs"
symptom needed the explicit `subscribeLogs(true)` re-call in resync.
- `terminal.reset()` lives after a successful raw-logs fetch (not
before) so a failed resync leaves the prior buffer visible instead of
blanking it; resync errors surface via the same inline error message the
mount path uses.
- 8 unit tests around the composable: mount + subscribe, resync ordering
(reset → write → scroll → subscribe via `invocationCallOrder`),
in-flight resync abort on double reconnect, resync error surfacing,
mount-failure-then-recovery, unmount-mid-fetch terminal-write
suppression, listener cleanup on unmount.
- 2 E2E tests using `ws.close()` on the proxied WebSocket as the
reconnect trigger and `subscribeLogs` HTTP fetch count as the sync point
(same pattern as `wsReconnectStaleJob.spec.ts`). Red-checked: disabling
the `reconnected` listener fails exactly the two new tests, all 8
pre-existing tests stay green.
Fixes FE-712
## Screenshots
Before - (After rebooting, the console window does not update from its
state before the reboot must remount the console window for it to
resync.)
https://github.com/user-attachments/assets/b1e49c2c-89a4-4a4a-82b4-064412acee12
After - (The console window syncs automatically after a reboot.)
https://github.com/user-attachments/assets/54b582c5-ad42-41c0-9886-18f4495859da
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12270-fix-terminal-resync-logs-console-on-backend-reconnect-3606d73d3650812fb13fd1934c632344)
by [Unito](https://www.unito.io)
## Summary
Add the first product-area browser coverage on top of the merged typed
route mock foundation: the docked job history sidebar.
## Changes
- **What**: Adds `browser_tests/tests/sidebar/jobHistory.spec.ts` using
`jobsRouteFixture`.
- **What**: Covers direct sidebar entry, docked QPO history entry,
terminal history jobs, active queue jobs, tab filtering, search, clear
queue, and clear history.
- **What**: Adds typed `POST /api/queue` and `POST /api/history` route
helpers that validate request bodies with generated zod schemas.
- **What**: Adds stable test ids for the job history sidebar and queue
progress overlay so tests avoid structural CSS selectors.
- **Dependencies**: Builds on the typed route mock foundation merged in
#12267.
## Review Focus
Review the product assertions and whether this is the right first
coverage slice on top of the typed route mock foundation. This PR
intentionally avoids asset sidebar and floating QPO lifecycle coverage;
those should remain follow-up PRs.
## Screenshots (if applicable)
Not applicable.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12272-test-cover-job-history-sidebar-with-typed-route-mocks-3606d73d3650817481d5f9fac4bfc93c)
by [Unito](https://www.unito.io)
## Summary
Migrates the MaxHistoryItems browser coverage to the accepted jobs route
fixture pattern.
## Changes
- **What**: Composes `jobsRouteFixture` into the queue settings spec and
removes the old `AssetsHelper` route setup.
- **What**: Adds a `responseLimit` option to `jobsRouteFixture` so tests
can match a requested history limit while intentionally returning more
jobs.
- **Dependencies**: None.
## Review Focus
The key behavior is preserving both FE-501 acceptance cases: `/api/jobs`
still receives the configured `limit`, and the queue panel still caps
rendered history even if the mocked backend returns more rows than
requested.
Fixes FE-501
## Screenshots (if applicable)
Not applicable.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12298-test-migrate-MaxHistoryItems-browser-coverage-3616d73d365081d6bf77fb205fcd51d4)
by [Unito](https://www.unito.io)
## Summary
Migrate the default combo widget select from the PrimeVue `SelectPlus`
wrapper to a Reka `Combobox` implementation while preserving the
existing Comfy combo widget contract and the node-canvas dropdown
behavior.
## Changes
- **What**: Rewrites `WidgetSelectDefault.vue` on top of Reka
`ComboboxRoot`, `ComboboxTrigger`, `ComboboxInput`, `ComboboxContent`,
and `ComboboxItem`.
- **What**: Preserves the default combo widget surface: `v-model`,
`widget` prop, `aria-label` from the widget name/label,
`data-capture-wheel`, disabled state, placeholder/filter placeholder,
default slot controls, invalid current value display, array values,
dynamic/factory values, and `getOptionLabel` fallback behavior.
- **What**: Keeps dynamic `values` compatibility by refreshing
function-backed options when the dropdown opens, without re-evaluating
the factory on every search keystroke.
- **What**: Deletes the now-unused PrimeVue `SelectPlus.vue` wrapper and
removes the PrimeVue test plugin/stub path from the default widget
select tests.
- **What**: Updates App Mode dropdown clipping coverage and combo-widget
browser coverage to target the new Reka overlay/viewport structure.
- **Breaking**: No breaking change is intended for the documented Comfy
combo widget contract. This migration does not preserve incidental
PrimeVue `Select` prop pass-through from `widget.options`; that was a
side effect of wrapping PrimeVue rather than a stable widget API.
- **Dependencies**: No new dependencies.
## Review Focus
### Compatibility choices
The goal of this PR is a migration PR, not a broad behavior redesign.
The new implementation keeps the Comfy-specific combo contract rather
than attempting to emulate PrimeVue internals. In particular:
- `values` still accepts arrays and functions, and function values are
re-read on open to support dynamic/custom node option sources.
- `getOptionLabel(value) || value` is intentionally preserved to match
the sibling dropdown path and avoid turning an empty-string label into a
blank rendered option.
- Invalid/current values that are not present in the option list are
still rendered in the trigger instead of disappearing.
- `WidgetWithControl` continues to render its default slot in the
control area, with trigger text truncation preserved.
- App Mode `OverlayAppendToKey='body'` continues to map to a body portal
to avoid panel clipping.
### Visual alignment and screenshot updates
The previous PrimeVue implementation passed `size="small"`, which
injected internal `.p-select-sm .p-select-label` styling. That internal
PrimeVue style used its own small-select font size and padding,
overriding the surrounding widget sizing intent and making the select
trigger subtly taller with slightly larger text than nearby inline node
widget controls.
The Reka implementation intentionally keeps the normal widget styling
path instead of recreating that PrimeVue-specific internal override.
This means the trigger follows the same inline widget sizing direction
as neighboring controls rather than preserving the incidental PrimeVue
height/text-size delta. Because this is an expected visual difference
from the migration, the affected E2E screenshots should be recaptured
instead of treating the old PrimeVue select height as the target.
### Scrollbar and focus behavior
Reka provides the combobox/listbox semantics we want, including search,
arrow navigation, highlighted items, and Enter selection. The tricky
part is the canvas dropdown scrollbar behavior. The native Reka viewport
path hides/owns scrollbar behavior in a way that made it hard to
preserve the previous widget dropdown affordances, especially visible
scrollbars and mouse wheel capture over the node canvas.
To keep the previous behavior, this PR renders a dedicated scrollable
viewport inside `ComboboxContent` with the project scrollbar utilities
(`scrollbar-thin`, stable gutter, transparent track). That preserves
visible scroll affordance and allows wheel events over the dropdown to
scroll the list instead of zooming the canvas.
There was one Reka interaction to account for: pressing the native
scrollbar can be treated as a focus-outside event from the search input,
which previously closed the dropdown on mouse down or caused subsequent
wheel events to leak back to the canvas. The new
`useRestoreFocusOnViewportPointer` composable handles only that short
pointer gesture:
- viewport pointerdown marks a short-lived scrollbar/viewport
interaction,
- the next focus-outside event is prevented only if the search input can
be restored,
- the guard is cleared by `pointerup`, `pointercancel`, and a timeout so
normal outside clicks still close the dropdown.
### Tests and regression coverage
Unit coverage was updated around the new Reka implementation:
- option sources from arrays and functions,
- dynamic values refreshed on open but not on each search keystroke,
- selection updates and blank/undefined Reka emissions being ignored,
- search filtering and Reka keyboard selection behavior,
- disabled state, invalid current values, `getOptionLabel`, empty
results status, and WidgetWithControl slot preservation,
- composable coverage for pointerup, pointercancel, repeated pointerdown
listener cleanup, and no-input/no-op behavior.
Browser regression coverage now checks the canvas-specific interaction
surface:
- opening and selecting default combo widget options,
- wheel over the dropdown scrolls the list instead of zooming the
canvas,
- pressing the scrollbar does not close the dropdown,
- wheel capture still works after pressing the scrollbar,
- opening another node widget closes the previous dropdown,
- switching between node widgets preserves dropdown scroll capture,
- serialize/reload retains selected combo values.
## Screenshots (if applicable)
New
<img width="527" height="753" alt="스크린샷 2026-05-18 오전 1 36 27"
src="https://github.com/user-attachments/assets/2293d510-6965-4b84-9b12-b8528f8a734f"
/>
Old
<img width="496" height="473" alt="스크린샷 2026-05-18 오전 1 35 57"
src="https://github.com/user-attachments/assets/47c0e28a-27df-44a6-81a8-14fcc1f3bd8f"
/>
Reka Supports Auto highlight top item on search (Search -> Enter ->
Select 👍)
https://github.com/user-attachments/assets/9d633dfc-c23a-4e7a-8d39-b044c219f1f3
The default combo widget trigger has a small intentional visual delta
from the old PrimeVue path because the Reka implementation does not
recreate PrimeVue's internal `size="small"` label override.
https://github.com/user-attachments/assets/a9053a14-e39e-4d5e-a846-dcf9aeb0caed
## Validation
- `pnpm format`
- `pnpm lint` (passes; existing warning-only lint output remains in
unrelated tests)
- `pnpm typecheck`
- `pnpm typecheck:browser`
- `pnpm test:unit`
- `PLAYWRIGHT_LOCAL=1 PLAYWRIGHT_TEST_URL=http://127.0.0.1:5174 pnpm
exec playwright test
browser_tests/tests/vueNodes/widgets/combo/comboWidget.spec.ts
--project=chromium`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12288-refactor-migrate-default-combo-widget-select-to-Reka-3616d73d365081fd8742c038a7dc7851)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: github-actions <github-actions@github.com>
## Summary
Resolve the queue progress node label from queued prompt metadata so
subgraph execution IDs show the correct node name without depending on
the live canvas.
## Changes
- **What**: Store a prompt-scoped `executionId -> { title, type }`
lookup from `p.output` when queueing a job, and use that lookup for the
active job's executing node label.
- **What**: Reuse the same job-scoped node info for the browser tab
title so it stays aligned with the queue overlay.
- **What**: Add unit coverage for root and subgraph execution IDs, and
merge the branch forward to current `main`.
## Review Focus
This keeps the fix scoped to the existing singular `activeJobId` path.
It fixes subgraph labels and avoids the workflow-switching regression
from resolving against `app.rootGraph`, but it does not redesign
concurrent multi-job selection yet.
Longer term, the cleaner solution is still prompt-scoped execution
metadata from the backend rather than frontend reconstruction.
## Screenshots (if applicable)
N/A
---------
Co-authored-by: Alexander Brown <drjkl@comfy.org>
Co-authored-by: Benjamin Lu <benjaminlu1107@gmail.com>
## Summary
Centralize template route mocking by mirroring the existing
`AssetHelper` fixture pattern, so tests no longer hand-roll
`page.route('**/templates/index.json', ...)` and
`page.route('**/templates/**.webp', ...)` blocks.
## Changes
- **What**:
- Expand `browser_tests/fixtures/data/templateFixtures.ts` with stable
distribution exports (`STABLE_CLOUD_TEMPLATE`,
`STABLE_DESKTOP_TEMPLATE`, `STABLE_LOCAL_TEMPLATE`,
`STABLE_UNRESTRICTED_TEMPLATE`), an `ALL_DISTRIBUTION_TEMPLATES` set,
and a `generateTemplates(count, distribution?)` bulk generator.
`makeTemplate` and `mockTemplateIndex` are preserved.
- Add `browser_tests/fixtures/helpers/TemplateHelper.ts` modeled on
`AssetHelper`: an immutable `TemplateConfig` driven by composable
operators (`withTemplates`, `withTemplate`, `withCloudTemplates`,
`withDesktopTemplates`, `withLocalTemplates`,
`withUnrestrictedTemplates`, `withRawIndex`), a `TemplateHelper` class
with `mock()` / `mockIndex()` / `mockThumbnails()` / `clearMocks()`, and
a `createTemplateHelper(page, ...operators)` factory.
- Add `browser_tests/fixtures/templateApiFixture.ts` exposing
`templateApi: TemplateHelper` as a Playwright fixture with automatic
`clearMocks()` teardown (mirrors `assetApiFixture`).
- Migrate `browser_tests/tests/templateFilteringCount.spec.ts` to
`mergeTests(comfyPageFixture, templateApiFixture)` and
`templateApi.configure(withTemplates(...))` + `templateApi.mockIndex()`.
The webp thumbnail mock moves into the helper.
- **Breaking**: None. `makeTemplate` and `mockTemplateIndex` exports
remain.
## Review Focus
- `TemplateHelper.mock()` is split into `mockIndex()` and
`mockThumbnails()` so the spec can install the thumbnail handler in
`beforeEach` (where it has no per-test data) and the index handler
per-test (after `configure(...)`). Both still register through
`routeHandlers` so `clearMocks()` unroutes everything.
- Operators are pure (`(config) => config`) and the helper deep-copies
the resulting `templates` array, matching the AssetHelper immutability
pattern.
- The thumbnail route is `**/templates/**.webp` and serves
`browser_tests/assets/example.webp` — identical to the previous inline
behavior.
Fixes#11431
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-11837-refactor-browser-tests-centralize-template-mocking-via-TemplateHelper-3546d73d365081beac00d80c2ab5ea1c)
by [Unito](https://www.unito.io)
## Summary
Replace the merged stateful jobs API browser mock fixture with a small
declarative typed route-mock foundation.
## Changes
- **What**: Removes `JobsApiMock`, `jobsApiMockFixture`, and the old
shared `jobFixtures` helper.
- **What**: Adds a generic `RouteMocker` primitive for explicit typed
JSON route responses.
- **What**: Adds `jobsRouteFixture`, which registers explicit
`/api/jobs` list/detail responses without filtering, mutation handling,
or hidden in-memory backend behavior.
- **What**: Migrates the current queue overlay and missing-media runtime
specs onto the new jobs route fixture.
- **What**: Keeps `./browser_tests/tsconfig.json` in the ESLint
TypeScript resolver config.
- **Dependencies**: None.
## Review Focus
This is intended to be the foundation PR for the test-strategy reset:
old stateful helper out, typed declarative route mocks in. It
intentionally does not add the full asset sidebar, job history sidebar,
or floating QPO coverage suite; those should stack on top after this
fixture shape is accepted.
The boundary this PR is trying to preserve: route mocks may describe
frontend-visible API responses, but should not implement Core
queue/history mutation semantics.
Context:
https://www.notion.so/comfy-org/E2E-Test-Strategy-for-Assets-Job-History-and-Queue-Progress-Overlay-35f6d73d365081209bc5f10e6c7eb8de
## Screenshots (if applicable)
Not applicable.
*PR Created by the Glary-Bot Agent*
---
## Summary
Click vs drag is disambiguated by two thresholds: distance
(`Comfy.Pointer.ClickDrift`, 6px) and time
(`Comfy.Pointer.ClickBufferTime`). Distance does the real work — any
intentional drag immediately exceeds 6px on the first move. The time
threshold only matters in the corner case where the pointer is held
still then released without crossing the distance threshold.
The previous 150ms default forces every pointerdown to wait up to 150ms
before drag begins, even when the user is clearly dragging. This is
visible as lag when click+dragging an unselected node, where the wait
stacks on top of selection-state work that runs at drag start
([FE-558](https://linear.app/comfyorg/issue/FE-558/bug-delay-when-clickdragging-an-unselected-node)).
32ms (~2 frames at 60fps) is well below human-perceptible click latency
and still safely above incidental jitter on pointerdown.
## Changes
- `src/platform/settings/constants/coreSettings.ts` —
`Comfy.Pointer.ClickBufferTime`:
- `defaultValue: 150` → `32`
- `versionAdded: '1.4.3'` retained, add `versionModified: '1.44.19'`
- Slider `step: 25` → `1` (the old step made `32` unrepresentable on the
slider)
- Tooltip extended with rationale (why distance is the real
disambiguator, why this should be small)
- `src/lib/litegraph/src/CanvasPointer.ts` — Static `bufferTime = 150` →
`32` for consistency with the user-setting default; the runtime
`watchEffect` in `useLitegraphSettings.ts` continues to override at
boot. JSDoc explains the rationale and that the user setting overrides
at runtime.
## Verification
- `pnpm typecheck` clean.
- Litegraph + settings test suites: 998/998 pass.
- `pnpm format` clean (lint-staged enforced on commit).
## Why not just leave it user-overridable?
It already is. But the default is what every user — including ones who
never open settings — experiences. 150ms was never a justified value (no
comment explains it, no benchmark, and 6px distance covers the
disambiguation). Defaults should reflect best UX.
Refs FE-558.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12032-feat-settings-lower-Comfy-Pointer-ClickBufferTime-default-from-150ms-to-32ms-3586d73d365081f5a29cdc84b4b736e3)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Glary-Bot <glary-bot@users.noreply.github.com>
Co-authored-by: github-actions <github-actions@github.com>
*PR Created by the Glary-Bot Agent*
---
## Summary
- Fixes app mode bug where custom combo inputs (and other widget inputs)
unselect and show "Widget not visible" after refreshing or reopening a
saved app workflow
- Root cause: `resetSelectedToWorkflow()` loads from
`changeTracker.activeState` which may not have linearData yet after
refresh, and `pruneLinearData()` prunes valid entries during graph
loading when nodes aren't yet in the graph
- Two defensive guards: fallback to `initialState` for authoritative
data, and skip pruning during graph loading
## Changes
- `appModeStore.ts`: `resetSelectedToWorkflow()` now falls back to
`initialState.extra.linearData` when `activeState` has none
- `appModeStore.ts`: `pruneLinearData()` skips node-existence checks
when `ChangeTracker.isLoadingGraph` is true
- Unit tests: 4 new tests covering both fix paths (pruning during
loading, fallback to initialState)
- E2E test: Save-as → close → reopen → verify all inputs persist with no
"Widget not visible"
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-11422-fix-preserve-app-builder-inputs-through-graph-reconfiguration-3476d73d36508166a563f7df3967665c)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Glary-Bot <glary-bot@users.noreply.github.com>
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
*PR Created by the Glary-Bot Agent*
---
## Summary
The in-app cloud paywall modal shown on first install of ComfyUI Desktop
1.0 advertised "400 Free Credits Monthly" as a benefit, but the free
tier is currently disabled, making that copy misleading.
Per the thread direction, this skips the feature-flag plumbing (which
isn't available on Desktop anyway) and simply removes the line.
## Changes
- `CloudNotificationContent.vue`: change the feature loop from `n in 4`
to `n in [2, 3, 4]` so feature1 is skipped.
- `src/locales/*/main.json` (12 locales): remove the now-unused
`cloudNotification.feature1Title` key.
- `browser_tests/tests/dialog.spec.ts`: add a regression assertion that
"400" / "Free Credits" copy is no longer present in the modal.
## Test plan
- `pnpm typecheck` — passes
- `pnpm typecheck:browser` — passes
- `pnpm exec vitest run
src/platform/cloud/notification/components/DesktopCloudNotificationController.test.ts`
— 3/3 pass
- Pre-commit hooks (oxfmt, oxlint, eslint, stylelint, typecheck) — all
pass
- Manual: triggered `showCloudNotification()` against the dev server and
confirmed only 3 benefit rows render (screenshot attached); the "400
Free Credits Monthly" line is gone.
- Fixes DESK2-90
## Screenshots
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12251-fix-remove-400-Free-Credits-Monthly-line-from-cloud-paywall-modal-3606d73d365081eaa572e6cd995278d8)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Glary-Bot <glary-bot@users.noreply.github.com>
## Summary
Replaces #12164.
Right-clicking a Vue node, using the selection toolbox More Options
menu, or clicking the selection toolbox Node Info button now opens the
right-side Info tab only when the new-menu UI makes that panel
available. Legacy-menu contexts hide the no-op action even when the
legacy node library design is selected; node-library help remains
isolated to the node library itself. The existing
`selection_toolbox_node_info_opened` telemetry fires only after the
toolbox button successfully opens node info. No new context-menu
telemetry event is added in this PR.
## Changes
- **What**: Share the node-info availability/action path across the
context menu and selection toolbox, keep legacy-menu state out of the
right-side panel public store API, tighten node-info settings tests, and
add unit plus E2E regression coverage for new-menu and legacy-menu
modes.
- **Dependencies**: None
## Review Focus
Confirm the node context menu, selection toolbox direct Info button, and
selection toolbox More Options entry all respect right-side panel
availability, including legacy menu + legacy node library mode, while
node-library help behavior remains isolated to the node library.
## Validation
- Self-review: checked production path, unit mocks, and Playwright
coverage; only gap found was weak E2E coverage for the toolbox direct
Info path, now strengthened.
- `pnpm test:unit -- src/composables/graph/useSelectionState.test.ts
src/components/graph/SelectionToolbox.test.ts
src/components/graph/selectionToolbox/InfoButton.test.ts`
- `pnpm test:browser:local -- --project=chromium
browser_tests/tests/selectionToolboxActions.spec.ts
browser_tests/tests/selectionToolboxSubmenus.spec.ts
browser_tests/tests/vueNodes/interactions/node/contextMenu.spec.ts
--grep "info button opens the right-side info tab|info button is
hidden|hides Node Info|should open node info"`
- `pnpm typecheck:browser`
- `pnpm exec oxlint --type-aware
browser_tests/tests/selectionToolboxActions.spec.ts`
- `pnpm exec eslint --cache --no-warn-ignored
browser_tests/tests/selectionToolboxActions.spec.ts`
- `pnpm exec oxfmt --check
browser_tests/tests/selectionToolboxActions.spec.ts`
- `git diff --check`
- Commit hooks: lint-staged + `pnpm typecheck` + `pnpm
typecheck:browser`
- Push hook: `knip --cache` (existing tag hint only)
## Screenshots (if applicable)
Before
https://github.com/user-attachments/assets/4b1f6ddb-a01c-4958-81ab-36167f434e59https://github.com/user-attachments/assets/83433f0d-24f1-46b7-a81d-f0f065812496
After
https://github.com/user-attachments/assets/30bd61e5-f8d4-48b7-97e0-26c93e3cb362https://github.com/user-attachments/assets/afce9f51-a43d-434f-a006-6b357a61ac8f
---------
Co-authored-by: github-actions <github-actions@github.com>
## Summary
Clear missing media validation errors after paste/drop media uploads by
emitting the existing widget-change event path.
## Changes
- **What**: Emit `node.onWidgetChanged` after image/video upload
completion updates the file combo widget.
- **What**: Emit the same widget-change path after Load Audio upload
completion.
- **What**: Add unit coverage for upload completion emitting
`onWidgetChanged` and for missing media clearing through that existing
hook path.
- **What**: Add E2E coverage for Load Image drag/drop and paste clearing
validation rings, with red/green verified from a fresh `main` base.
- **Dependencies**: None.
## Review Focus
Please check that paste/drop upload paths now reuse the existing
widget-change error-clearing path instead of expanding `widget.callback`
patching.
Also check the Load Image E2E helper path for synthetic paste/drop
behavior.
Supersedes #12207.
Ref: FE-687
## Screenshots
Before
https://github.com/user-attachments/assets/2cee52bc-b1c8-4dff-8a02-5b18a69ae639
After
https://github.com/user-attachments/assets/e1ecd147-1d8a-470e-b77d-13345d473ef3
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12212-fix-clear-media-upload-errors-via-widget-change-35f6d73d365081bcb1a0dfc042d417eb)
by [Unito](https://www.unito.io)
## Summary
Adds browser coverage for the missing-media runtime paths introduced by
#12069 and #12111:
- OSS: annotated `[output]` media is resolved from job history.
- Cloud: compact `[output]` media is resolved from output assets.
- OSS and Cloud: dropped video uploads do not surface a missing-media
error while the upload is still in progress.
This PR is now rebased directly onto `main`; the parent fix PRs have
been squash-merged, so this branch only contains the E2E coverage
commit.
## Test Fixtures
- Adds workflow fixtures for OSS spaced output annotations and Cloud
compact output annotations.
- Adds a small plain MP4 fixture for video drag/drop upload coverage.
## Validation
- `pnpm exec oxfmt --check
browser_tests/tests/propertiesPanel/errorsTabMissingMediaRuntime.spec.ts
browser_tests/assets/missing/missing_media_cloud_output_annotation.json
browser_tests/assets/missing/missing_media_output_annotations.json`
- `pnpm typecheck:browser`
- `pnpm exec oxlint
browser_tests/tests/propertiesPanel/errorsTabMissingMediaRuntime.spec.ts
--type-aware`
- `git diff --check origin/main..HEAD`
Note: before the rebase, the Cloud project target for this spec passed
locally. Local OSS project execution against the currently running Cloud
dist did not reach the test body because `ComfyPage.waitForAppReady`
timed out in `beforeEach`.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12126-test-cover-missing-media-runtime-sources-35d6d73d365081f0a981c02f33c0ff84)
by [Unito](https://www.unito.io)
- Adds functions to SubgraphHelper to perform widget promotion by
standard user means
- Right Click -> Promote
- Properties Panel
- Adds new slot fixture code that works with simple `locator.dragTo`
operations.
- Adds multiple subgraph tests with a focus on historically difficult
operations.
- Fixes a bug where the litegraph `node.selected` state would not be
unset when switching graphs. This made it so 'Selecting a node ->
leaving subgraph -> re-enter subgraph -> right click on node' would fail
to select the node because it is marked as already selected.
┆Issue is synchronized with this [Notion
page](https://app.notion.com/p/PR-11806-Add-helper-functions-for-widget-promotion-3536d73d365081f58dd9cd730c1a91a9)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
API and other legacy JSON generated by python `json.dumps` can contain
`NaN` and `Infinity` which cannot be parsed with JS `JSON.parse`. This
adds regex to replace these invalid tokens with `null`.
## Changes
- **What**:
- add regex replace on bare NaN/infinity tokens after JSON.parse fails
- update call sites
- tests
## Review Focus
- The regex should only rewrite bare NaN/-Infinity/Infinity and not
touch string values or other invalid tokens.
- A small regex was chosen over JSON5 due to package size (30.3kB
Minified, 9kB Minified + Gzipped) or a manual parser due to the
unnecessarily complexity vs a single regex replace.
- The happy path is run first, the safe parse is only executed if that
failed, meaning no overhead the vast majority of the time and no
possiblity of corrupting valid workflows due to a bug in the fallback
parser
- Multiple call sites had to be updated due to pre-existing architecture
of the various parsers, an issue for unifying these is logged for future
cleanup
- New binary fixtures added for validating e2e import using real files
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12217-fix-add-support-for-parsing-python-generated-json-with-NaN-infinite-35f6d73d365081889fc7f4af823f29c1)
by [Unito](https://www.unito.io)
## Summary
The position of a link relative to its slot was able to drift on load,
due to widgets inside a node being able to resize without triggering an
node-level resize event (min-height node with space at the bottom could
have widgets expand into free space, causing misalignment).
Recreation:
1. Add KSampler
2. Add Float
3. Connect Float to KSamper.denoise
4. Reload workflow (F5)
5. Observe misalignment
## Changes
- **What**:
- track widget grid element as signal only that triggers resync
- node bound calculations skipped for widget signals
- prevent setDirty on non-graph nodes (e.g. LGraphNodePreview)
- tests
## Review Focus
This is a small focused approach to fix the reported issue - it does not
address the underlying issue of the layout not being a SSOT. This fix is
a small bandaid and investigation into resolving the layout SOT issue is
not impacted by this.
## Screenshots (if applicable)
Before:
<img width="673" height="374" alt="image"
src="https://github.com/user-attachments/assets/2d34b8e3-0731-4fd2-8553-4dd429010ced"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12214-fix-resolve-widget-input-link-position-drift-on-reload-35f6d73d3650814eb31bebb3042ff58b)
by [Unito](https://www.unito.io)
## Summary
When a user who has not used the app before first loads up, they are
presented with the template selection dialog. This conflicts when the
first-time user visits the app via a share link - both the share &
template dialog are triggered.
## Changes
- **What**:
- Skip the templates browser when share param is in URL
- Tests
- Add `url` to `setup`/`goto` to allow specifying the `share` parameter
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12024-fix-prevent-first-user-template-popup-when-following-shared-link-3586d73d365081cbbcecdba45a1ad1ea)
by [Unito](https://www.unito.io)
## Summary
Rework the painter always hands the backend a valid asset reference:
- Drop the `hasStrokes` flag and the `isCanvasEmpty` check.
- `serializeValue` falls back to the existing `modelValue` when the
canvas element is transiently unmounted, reuses the cached upload when
not dirty and a value is present, and otherwise uploads the current
canvas (a fully transparent PNG is a valid no-op mask, Painter's Python
`execute()` treats painter_alpha=0 the same as "no mask painted").
- `handleClear` now also clears `modelValue` so a user-initiated clear
doesn't resurrect a stale upload on the next serialize.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12196-FE-566-fix-Painter-mask-submission-edge-cases-on-cloud-35e6d73d365081dd8856ddb785952526)
by [Unito](https://www.unito.io)
## Summary
Sidebar buttons rendered literal i18n keys (e.g.
`sideToolbar.labels.assets`) on a fresh install when the user's
`navigator.language` base tag wasn't one of the 12 shipped locales —
German/Italian/Polish/Dutch/Brazilian-Portuguese users among others.
## Changes
- **What**: Add `resolveSupportedLocale()` that tries the full BCP-47
tag first (preserves `zh-TW`, `pt-BR`), then the base tag, then `'en'`.
Wire through both entry points (`createI18n`'s initial locale,
`Comfy.Locale`'s `defaultValue`) and clamp inside `loadLocale`,
propagating the resolved tag to `GraphView` so a stale stored
`Comfy.Locale='de'` from older builds also recovers.
- **Side benefit**: Brazilian Portuguese users were previously falling
through `pt-BR` → `pt` (unshipped) → broken. The full-tag-first lookup
now correctly lands them on the `pt-BR` bundle.
- **Breaking**: None.
- **Dependencies**: None.
## Root Cause
Three-link chain:
1. `Comfy.Locale`'s default was `() => navigator.language.split('-')[0]
|| 'en'`. German → `'de'` (unshipped).
2. `loadLocale('de')` silently `console.warn`'d and returned without
throwing.
3. `GraphView` then ran `i18n.global.locale.value = 'de'` anyway.
4. `st(key, fallback) = te(key) ? t(key) : fallback`. vue-i18n's `te()`
checks **only** the current locale and ignores `fallbackLocale` — every
key missed → `st()` returned the literal key string.
Two pathways reached the broken state (defaultValue path, and
unset-setting path through `createI18n`'s own `navigator.language`
snapshot); the new helper closes both.
## Review Focus
- `loadLocale` now returns `SupportedLocale` (was `void`). Old `void`
callers continue to compile; the only change is `GraphView` consuming
the return value.
- Unit-tested in `src/i18n.test.ts` (added `resolveSupportedLocale`
block + updated the `loadLocale` unsupported-locale case from "warn" to
"clamp to en").
- Self-reproduced via Playwright with `navigator.language='de-DE'` +
fresh-install state on both `main` (shows the bug) and this branch
(shows the fix). Spec saved at
`temp/scripts/issue-10563-locale-bug.spec.ts`.
Fixes#10563
FE-480 — https://linear.app/comfyorg/issue/FE-480
## Screenshots
**Before** (from #10563, on `main`):
<img width="258" height="399" alt="Sidebar with literal i18n keys"
src="https://github.com/user-attachments/assets/098d1d76-8e89-4237-813f-5f030b34e51e"
/>
**After** (this branch, same `navigator.language='de-DE'`):
<img width="367" height="793" alt="Screenshot 2026-04-28 at 2 07 38 PM"
src="https://github.com/user-attachments/assets/9d279de3-50a8-4774-999f-ab4c3018a9ef"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-11712-fix-i18n-clamp-unsupported-browser-locales-to-a-shipped-tag-3506d73d3650812f89d2f0fe3199de3a)
by [Unito](https://www.unito.io)
## Summary
When an image fails to load in the image preview, the context buttons
are still visible - clicking these does not work (Mask editor opens and
closes, download does nothing) - this hides the buttons if load fails.
## Changes
- **What**:
- hide buttons if load failed
- tests
## Review Focus
<!-- Critical design decisions or edge cases that need attention -->
<!-- If this PR fixes an issue, uncomment and update the line below -->
<!-- Fixes #ISSUE_NUMBER -->
## Screenshots (if applicable)
Current:
<img width="622" height="857" alt="image"
src="https://github.com/user-attachments/assets/26e391a0-5538-4c6c-ac8a-b6f2b6acabae"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12136-fix-hide-image-buttons-if-load-failed-35d6d73d365081579c71f1849b9ab1bd)
by [Unito](https://www.unito.io)
## Summary
Align the asset list pagination schema with generated ingest-types
metadata and remove the now-unneeded missing `has_more` fallback branch.
## Changes
- **What**: Reuse `zListAssetsResponse` for `total` and `has_more`, keep
the local loose `AssetItem` shape, and simplify `getAllAssetsByTag()` to
trust the required `has_more` contract.
- **Breaking**: None.
- **Dependencies**: None.
## Review Focus
This is PR 1 of 4 in the missing asset follow-up stack:
1. This PR - Asset schema / pagination cleanup
2. #11900 - Missing asset hash verification utility cleanup
3. #11901 - Browser regression coverage for public input assets
4. #11902 - TanStack Query public-input cache replacement
The key decision is intentionally narrow: pagination metadata now comes
from generated ingest-types, but asset item validation remains locally
loose to avoid changing UI/store synthetic asset shapes in this PR.
`asset_hash` nullability remains unchanged because absent-vs-null hash
semantics are still a backend/API contract follow-up.
Addresses #11894
## Screenshots (if applicable)
N/A
## Summary
The dirtyClose modal had three buttons (`Cancel | No | Save`) and the
sign-out flow collapsed two distinct outcomes (deny vs. dismiss) into a
single early return — so today clicking "No" *cancels* sign-out instead
of signing out without saving, and clicking "Save" never actually saves
before logging out. This PR drops `Cancel` for `dirtyClose`, gives each
caller a context-specific deny label, and fixes the sign-out 3-state
handling.
- Fixes
[FE-419](https://linear.app/comfyorg/issue/FE-419/unsaved-changes-modal-uses-confusing-button-labels)
## Changes
- **What**:
- `ConfirmationDialogContent.vue`: hide `Cancel` for
`type='dirtyClose'`; add `denyLabel?: string` prop; autofocus `Save`
(preserves work on Enter).
- `dialogService.confirm()`: accept and forward `denyLabel`.
- `useAuthActions.logout`: handle `null` (cancel) / `false` (sign out
anyway, no save) / `true` (save each modified workflow, then logout)
distinctly. Pass `denyLabel: 'Sign out anyway'`.
- `workflowService.closeWorkflow`: pass `denyLabel: 'Close anyway'`.
- i18n: add `auth.signOut.signOutAnyway` and
`sideToolbar.workflowTab.closeAnyway`.
- **Breaking**: none. The `denyLabel` prop is optional and falls back to
`g.no`.
## Review Focus
- The "Save" branch in `useAuthActions.logout` now iterates
`workflowStore.modifiedWorkflows` and awaits
`useWorkflowService().saveWorkflow(workflow)` for each before calling
`authStore.logout()`. The close-tab path
(`workflowService.closeWorkflow`) was already correct — only the
sign-out path needed the same shape.
- `ConfirmationDialogContent` autofocus moves from `Cancel` (gone for
`dirtyClose`) to `Save`. The dialog is still dismissable via ESC /
outside-click, which routes through `dialogComponentProps.onClose →
resolve(null)` — sign-out and close-tab both treat `null` as cancel.
- Out of scope: the native browser `beforeunload` warning
(`UnloadWindowConfirmDialog.vue`) is a separate flow and never reaches
the in-app modal.
## Tests
- Unit (`useAuthActions.test.ts`, new): logout handles `null` / `false`
/ `true` / no-modified-workflows; saves *every* modified workflow before
`authStore.logout`; passes `denyLabel='Sign out anyway'`.
- Unit (`ConfirmationDialogContent.test.ts`): Cancel hidden for
`dirtyClose`; custom `denyLabel` rendered; falls back to `g.no` when
omitted.
- E2E (`workflowTabs.spec.ts`): modified-tab close shows `Close anyway`
(not `No`) and no `Cancel`; clicking `Close anyway` removes the tab; ESC
keeps the tab.
## screenshot
### AS IS
<img width="816" height="379" alt="Screenshot 2026-04-27 at 5 40 19 PM"
src="https://github.com/user-attachments/assets/a8e39403-bf72-455a-8d86-6ceb1f94ac85"
/>
<img width="923" height="396" alt="Screenshot 2026-04-27 at 5 40 38 PM"
src="https://github.com/user-attachments/assets/08031c7c-b3a6-45d7-a4dc-5dcb4e63cfa0"
/>
### TO BE
<img width="1661" height="872" alt="Screenshot 2026-04-27 at 5 43 40 PM"
src="https://github.com/user-attachments/assets/b89d160b-be66-450e-981e-32b1591f6841"
/>
<img width="1488" height="584" alt="Screenshot 2026-04-27 at 5 44 21 PM"
src="https://github.com/user-attachments/assets/b3a141a7-1f3b-4f25-85a9-49529229c28b"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-11669-fix-clarify-unsaved-changes-modal-buttons-and-fix-sign-out-3-state-34f6d73d365081bf8afad8e146b3b990)
by [Unito](https://www.unito.io)
## Summary
Reuse the actionbar queue-button page object in the queue mode E2E tests
so dropdown selectors live in one helper.
## Changes
- **What**: Adds queue mode menu/item helpers to
`ComfyActionbar.queueButton` and updates `queueButtonModes.spec.ts` to
use them.
- **Dependencies**: None.
## Review Focus
This is stacked on #11209 and should be reviewed as a test-infra cleanup
only; the behavior covered by the spec is unchanged.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-11927-test-reuse-queue-button-page-object-in-e2e-3566d73d365081918d59c2d587c4c94a)
by [Unito](https://www.unito.io)
## Summary
Add a `@perf` test measuring the cost of entering a subgraph containing
80 interior nodes. Establishes a CI baseline for the synchronous
mount/unmount bottleneck.
## Changes
- **What**: Add `subgraph transition (enter and exit)` perf test to
`performance.spec.ts` and a test workflow asset
(`large-subgraph-80-nodes.json`) with a single subgraph node containing
80 Note nodes.
## Review Focus
This is PR 1 of 2. The test establishes a baseline on main so the
optimization PR (PR 2) can show a CI-proven delta for `taskDurationMs`
and `totalBlockingTimeMs`.
The test:
1. Loads the 80-node subgraph workflow
2. Enters and exits once to warm up
3. Measures a fresh enter transition (start → 80 nodes mounted → layout
settled)
4. Records `taskDurationMs`, `layouts`, and `TBT`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10480-test-add-perf-test-for-subgraph-transition-bottleneck-32d6d73d3650811b9b6eec03a9591f82)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Connor Byrne <c.byrne@comfy.org>
## Summary
Preview as Text (`PreviewAny`) nodes were re-executing on every prompt
submission because the rendered preview text was being echoed back to
the backend as input values, mutating the cache signature.
## Changes
- **What**: Set `widget.options.serialize = false` on the three widgets
the `Comfy.PreviewAny` extension adds (`preview_markdown`,
`preview_text`, `previewMode`) so they are excluded from the API prompt
sent to the backend.
## Root cause
The extension was setting `widget.serialize = false`, which only
controls **workflow JSON** persistence (checked in
`LGraphNode.serialize`). The **API prompt** serializer in
`executionUtil.graphToPrompt` checks `widget.options.serialize` instead
— a distinct property documented in litegraph's `WIDGET_SERIALIZATION`
convention.
After `onExecuted` writes the rendered text into the widget value, the
next `graphToPrompt` call serialized that text into
`inputs.preview_text` / `inputs.preview_markdown`. The backend cache
signature in `comfy_execution/caching.py` hashes all keys in
`node["inputs"]`, so the changing text invalidated the cache and forced
a redundant execution every time.
## Review Focus
Two commits, red-green TDD:
1. `test:` failing unit + e2e tests asserting the desired behavior.
2. `fix:` adds `options.serialize = false` to make them pass.
Tests verify the widgets are excluded from the API prompt; e2e
additionally simulates a prior execution populating widget values to
mirror the real bug condition.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-12010-fix-stop-PreviewAny-widgets-from-triggering-re-execution-3586d73d3650810585cdd077f3ac64f5)
by [Unito](https://www.unito.io)
## Changes
Add Playwright E2E tests for queue button modes in the topbar (6 tests):
- Run button visibility in topbar
- Queue mode trigger menu visibility
- Opening the mode menu via trigger click
- Verifying available modes are shown as menu item radios
- Menu closes after selecting a mode
- Run button sends prompt when clicked (via intercepted `/api/prompt`
route)
## Test
```bash
pnpm test:browser:local -- browser_tests/tests/queueButtonModes.spec.ts
```
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-11209-test-add-E2E-tests-for-queue-button-modes-3416d73d365081a5bf10f8b9c6bdc2a7)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Benjamin Lu <benjaminlu1107@gmail.com>
