74147d7ee2
## Summary Backend validation errors (`node_errors`) are reported against the **flattened** prompt, so an error whose real fix lives on a subgraph host node gets attached to an interior node the user may never open — and, after ADR 0009, often *cannot* meaningfully fix there. This PR re-surfaces a validation error onto the subgraph host node **when — and only when — the error's subject (the specific input/widget named by `extra_info.input_name`) is exposed through the subgraph boundary**. ## Why this is needed Two concrete situations motivated this, both observed in real workflows: 1. **Broken link at the host.** Root node A should feed subgraph host B, whose boundary input is linked to interior node C. If the A→B link is missing, the backend flattens the prompt, sees C with no resolved input, and raises `required_input_missing` **on C** (`"B:C"`). The actual fix — connect B's input — is one level up, on a node the error never points at. 2. **Host-owned widget values.** Per ADR 0009 (subgraph promoted widgets use linked inputs), a promoted widget's value is owned by the host `SubgraphNode`; the interior widget only supplies schema/defaults. When the backend raises `value_not_in_list` (or min/max violations) for that value, attributing it to the interior node is factually wrong — the value that failed validation *is the host's value*. This continues the direction of #13059, which moved **missing-model** detection identity to `{hostExecutionId, hostWidgetName}` with the interior path kept as diagnostics. That was possible in the FE pre-scan; this PR applies the same ownership principle to **backend-received** errors via a receive-side mapping, since the backend cannot know about subgraph boundaries in a flattened prompt. ## The rule (design) > Lift an error from interior node N to host H **iff** N's input slot named by the error is linked to the containing subgraph's boundary (`SubgraphInput`). Apply the same test again at H (boundary-by-boundary, matching ADR 0009's chaining principle) and stop at the first level where the subject is no longer boundary-linked — that node is where the user can actually fix it. The predicate is **structural (boundary exposure), not data-flow**. ### In scope — examples - `required_input_missing` on interior `"12:5"` whose input is fed by the boundary → surfaces on host `12`'s input slot (red slot ring on the host, errors-tab card titled/located at the host, message names the host's `SubgraphInput.name`). - `value_not_in_list` / `value_smaller_than_min` / `value_bigger_than_max` on a promoted interior widget → surfaces on the host's promoted widget. Nested hosts chain: `"1:2:3"` lifts to `"1:2"`, and further to `"1"` only if `1:2`'s own slot is boundary-linked too. - Clearing follows the surface: connecting the highlighted host input or fixing the host widget clears the underlying interior (raw) error — range-guarded per target, so a still-out-of-range host value does **not** clear. ### Out of scope — examples - **No value-flow ancestry.** All in the root graph: A's widget links to B, B's to C, and C rejects the value that originated at A → the error **stays on C**. Following same-graph links to a "root cause" node is explicitly not this feature. - Errors without an `input_name` subject, node-level types (`exception_during_validation`, `dependency_cycle`, image-not-loaded), and unknown validation types — never lifted. Unknown types stay node-scoped to match how the error catalog renders them (the shared `isNodeLevelValidationError` in `executionErrorUtil` encodes this, and the catalog derives its node-level rules from the same set). - Runtime execution errors (exceptions during a run) — validation responses only. - Interior errors whose input is fed by another interior node — fixable in place, stay in place. - Fan-out display dedupe: when one boundary input feeds multiple interior nodes and the host slot is unconnected, each interior error lifts to the same host slot as a separate panel line. A single fix (connecting the host input) clears all of them — the clearing translation already fans out — so the duplication is cosmetic. Display-level dedupe is a follow-up; deduping inside the lift would break the one-source-per-error clearing contract. - Reactive re-lifting on graph topology edits while errors are displayed (invariant documented on the computed; follow-up), and deriving the error catalog's full validation rule table from the shared classification (follow-up; the node-level type set and the image-not-loaded predicate are already single-sourced in `executionErrorUtil` and consumed by both the lift and the catalog). ## Changes - **What**: New pure module `core/graph/subgraph/liftNodeErrorsToBoundary.ts` — per-error, fail-open record transform (unresolvable ids/slots/links leave the error where the backend put it; raw payload is never mutated). `executionErrorStore` derives `surfacedNodeErrors` from it and display consumers switch over (errors tab grouping, canvas node/slot flags, Vue node badges, app-mode/linear hints); raw `lastNodeErrors` remains the source of truth for mutation. Host-side clearing translates through the lift's diagnostics fields (`source_execution_id` / `source_input_name`) with a per-target range guard. - **Breaking**: None. No persistence/serialization changes; interior identity survives as diagnostics metadata only (ADR 0009 language). ## Review Focus - The lift predicate lives entirely on link topology (`LLink.originIsIoNode` → `SubgraphInput`) — no `proxyWidgets`/promotion-store style source authority is reintroduced. - `clearSlotErrorsWithRangeCheck` now resolves clear targets first and range-checks each target's raw errors; the lifted-path twin of the existing range-retention test pins this. - `useProcessedWidgets` deliberately stays on the raw record: host promoted widgets already map to interior errors via `widget.sourceExecutionId`, so an interior widget keeps its local red hint when the user opens the subgraph (hint layer vs surface layer). - Unit coverage is carried by the pure module (real litegraph subgraph fixtures, incl. nested recursion, promoted widgets via `promoteValueWidgetViaSubgraphInput`, ordering, fail-open/no-mutation); one e2e pins the user-visible contract (host ring + host slot dot + interior clean). ## Screenshots ### Before https://github.com/user-attachments/assets/81e5c4db-515d-4f1f-8f8a-e07ac490510f ### After https://github.com/user-attachments/assets/2949da06-a049-41c1-a480-98ee28333bf2
Release Schedule
The project follows a structured release process for each minor version, consisting of three distinct phases:
-
Development Phase - 2 weeks
- Active development of new features
- Code changes merged to the development branch
-
Feature Freeze - 2 weeks
- No new features accepted
- Only bug fixes are cherry-picked to the release branch
- Testing and stabilization of the codebase
-
Publication
- Release is published at the end of the freeze period
- Version is finalized and made available to all users
Nightly Releases
Nightly releases are published daily at https://github.com/Comfy-Org/ComfyUI_frontend/releases.
To use the latest nightly release, add the following command line argument to your ComfyUI launch script:
--front-end-version Comfy-Org/ComfyUI_frontend@latest
Overlapping Release Cycles
The development of successive minor versions overlaps. For example, while version 1.1 is in feature freeze, development for version 1.2 begins simultaneously. Each feature has approximately 4 weeks from merge to ComfyUI stable release (2 weeks on main, 2 weeks frozen on RC).
Example Release Cycle
| Week | Date Range | Version 1.1 | Version 1.2 | Version 1.3 | Patch Releases |
|---|---|---|---|---|---|
| 1-2 | Mar 1-14 | Development | - | - | - |
| 3-4 | Mar 15-28 | Feature Freeze | Development | - | 1.1.0 through 1.1.13 (daily) |
| 5-6 | Mar 29-Apr 11 | Released | Feature Freeze | Development | 1.1.14+ (daily) 1.2.0 through 1.2.13 (daily) |
| 7-8 | Apr 12-25 | - | Released | Feature Freeze | 1.2.14+ (daily) 1.3.0 through 1.3.13 (daily) |
Release Summary
Major features
v1.5: Native translation (i18n)
ComfyUI now includes built-in translation support, replacing the need for third-party translation extensions. Select your language
in Comfy > Locale > Language to translate the interface into English, Chinese (Simplified), Russian, Japanese, Korean, or Arabic. This native
implementation offers better performance, reliability, and maintainability compared to previous solutions.
More details available here: https://blog.comfy.org/p/native-localization-support-i18n
v1.4: New mask editor
https://github.com/Comfy-Org/ComfyUI_frontend/pull/1284 implements a new mask editor.
v1.3.22: Integrated server terminal
Press Ctrl + ` to toggle integrated terminal.
https://github.com/user-attachments/assets/eddedc6a-07a3-4a83-9475-63b3977f6d94
v1.1.0: Node search box
Fuzzy search & Node preview
Release link with shift
https://github.com/user-attachments/assets/a1b2b5c3-10d1-4256-b620-345de6858f25
QoL changes
v1.3.32: **Litegraph** Nested group
https://github.com/user-attachments/assets/f51adeb1-028e-40af-81e4-0ac13075198a
v1.3.24: **Litegraph** Group selection
https://github.com/user-attachments/assets/e6230a94-411e-4fba-90cb-6c694200adaa
v1.3.6: **Litegraph** Toggle link visibility
v1.3.4: **Litegraph** Auto widget to input conversion
Dropping a link of correct type on node widget will automatically convert the widget to input.
v1.3.4: **Litegraph** Canvas pan mode
The canvas becomes readonly in pan mode. Pan mode is activated by clicking the pan mode button on the canvas menu or by holding the space key.
v1.3.1: **Litegraph** Shift drag link to create a new link
v1.2.62: **Litegraph** Show optional input slots as donuts
v1.2.44: **Litegraph** Double click group title to edit
https://github.com/user-attachments/assets/5bf0e2b6-8b3a-40a7-b44f-f0879e9ad26f
v1.2.39: **Litegraph** Group selected nodes with Ctrl + G
https://github.com/user-attachments/assets/7805dc54-0854-4a28-8bcd-4b007fa01151
v1.2.38: **Litegraph** Double click node title to edit
https://github.com/user-attachments/assets/d61d5d0e-f200-4153-b293-3e3f6a212b30
v1.2.7: **Litegraph** drags multiple links with shift pressed
https://github.com/user-attachments/assets/68826715-bb55-4b2a-be6e-675cfc424afe
https://github.com/user-attachments/assets/c142c43f-2fe9-4030-8196-b3bfd4c6977d
v1.2.2: **Litegraph** auto connects to correct slot
Before
https://github.com/user-attachments/assets/c253f778-82d5-4e6f-aec0-ea2ccf421651
After
https://github.com/user-attachments/assets/b6360ac0-f0d2-447c-9daa-8a2e20c0dc1d
v1.1.8: **Litegraph** hides text overflow on widget value
https://github.com/user-attachments/assets/5696a89d-4a47-4fcc-9e8c-71e1264943f2
Developer APIs
v1.6.13: prompt/confirm/alert replacements for ComfyUI desktop
Several browser-only APIs are not available in ComfyUI desktop's electron environment.
window.promptwindow.confirmwindow.alert
Please use the following APIs as replacements.
// window.prompt
window['app'].extensionManager.dialog
.prompt({
title: 'Test Prompt',
message: 'Test Prompt Message'
})
.then((value: string) => {
// Do something with the value user entered
})
// window.confirm
window['app'].extensionManager.dialog
.confirm({
title: 'Test Confirm',
message: 'Test Confirm Message'
})
.then((value: boolean) => {
// Do something with the value user entered
})
// window.alert
window['app'].extensionManager.toast.addAlert('Test Alert')
v1.3.34: Register about panel badges
app.registerExtension({
name: 'TestExtension1',
aboutPageBadges: [
{
label: 'Test Badge',
url: 'https://example.com',
icon: 'pi pi-box'
}
]
})
v1.3.22: Register bottom panel tabs
app.registerExtension({
name: 'TestExtension',
bottomPanelTabs: [
{
id: 'TestTab',
title: 'Test Tab',
type: 'custom',
render: (el) => {
el.innerHTML = '<div>Custom tab</div>'
}
}
]
})
v1.3.22: New settings API
Legacy settings API.
// Register a new setting
app.ui.settings.addSetting({
id: 'TestSetting',
name: 'Test Setting',
type: 'text',
defaultValue: 'Hello, world!'
})
// Get the value of a setting
const value = app.ui.settings.getSettingValue('TestSetting')
// Set the value of a setting
app.ui.settings.setSettingValue('TestSetting', 'Hello, universe!')
New settings API.
// Register a new setting
app.registerExtension({
name: 'TestExtension1',
settings: [
{
id: 'TestSetting',
name: 'Test Setting',
type: 'text',
defaultValue: 'Hello, world!'
}
]
})
// Get the value of a setting
const value = app.extensionManager.setting.get('TestSetting')
// Set the value of a setting
app.extensionManager.setting.set('TestSetting', 'Hello, universe!')
v1.3.7: Register commands and keybindings
Extensions can call the following API to register commands and keybindings. Do note that keybindings defined in core cannot be overwritten, and some keybindings are reserved by the browser.
app.registerExtension({
name: 'TestExtension1',
commands: [
{
id: 'TestCommand',
function: () => {
alert('TestCommand')
}
}
],
keybindings: [
{
combo: { key: 'k' },
commandId: 'TestCommand'
}
]
})
v1.3.1: Extension API to register custom topbar menu items
Extensions can call the following API to register custom topbar menu items.
app.registerExtension({
name: 'TestExtension1',
commands: [
{
id: 'foo-id',
label: 'foo',
function: () => {
alert(1)
}
}
],
menuCommands: [
{
path: ['ext', 'ext2'],
commands: ['foo-id']
}
]
})
v1.2.27: Extension API to add toast message
iExtensions can call the following API to add toast messages.
app.extensionManager.toast.add({
severity: 'info',
summary: 'Loaded!',
detail: 'Extension loaded!',
life: 3000
})
Documentation of all supported options can be found here: https://primevue.org/toast/#api.toast.interfaces.ToastMessageOptions
v1.2.4: Extension API to register custom sidebar tab
Extensions now can call the following API to register a sidebar tab.
app.extensionManager.registerSidebarTab({
id: 'search',
icon: 'pi pi-search',
title: 'search',
tooltip: 'search',
type: 'custom',
render: (el) => {
el.innerHTML = '<div>Custom search tab</div>'
}
})
The list of supported icons can be found here: https://primevue.org/icons/#list
We will support custom icons later.
v1.10.9: Selection Toolbox API
Extensions can register commands that appear in the selection toolbox when specific items are selected on the canvas.
app.registerExtension({
name: 'TestExtension1',
commands: [
{
id: 'test.selection.command',
label: 'Test Command',
icon: 'pi pi-star',
function: () => {
// Command logic here
}
}
],
// Return an array of command IDs to show in the selection toolbox
// when an item is selected
getSelectionToolboxCommands: (selectedItem) => ['test.selection.command']
})
The selection toolbox will display the command button when items are selected:
Contributing
We welcome contributions to ComfyUI Frontend! Please see our Contributing Guide for:
- Ways to contribute (code, documentation, testing, community support)
- Development setup and workflow
- Code style guidelines
- Testing requirements
- How to submit pull requests
- Backporting fixes to release branches
Development
For detailed development setup, testing procedures, and technical information, please refer to CONTRIBUTING.md.
i18n
See locales/README.md for details.
Storybook
See .storybook/README.md for component development and visual testing documentation.
Troubleshooting
For comprehensive troubleshooting and technical support, please refer to our official documentation:
- General Troubleshooting Guide - Common issues, performance optimization, and reporting bugs
- Custom Node Issues - Debugging custom node problems and conflicts
- Desktop Installation Guide - Desktop-specific installation and troubleshooting