## Cause
When graphs are actually exported, several layers of cleanup are
applied. Among these is link compression. Any widgets with inputs that
aren't used do not have inputs stored in the workflow. This was
implemented for backwards compatibility with the old "convert to input"
system for widgets. As part of this process, the target_slots on links
are rewritten such that they point to the index of the widget as if
unconnected widgets did not exist.
This "incorrect" state for links is only corrected AFTER a workflow has
loaded because the 'fix' method needs nodes to be initialized in order
to calculate the correct target_slot
This becomes a problem when subgraphs are introduced. SubgraphInputs
need to resolve a link to its target slot in order to construct a clone
of the linked widget DURING the loading process. Since this target slot
is not accurate, this can result in the cloned widget having the wrong
type.
For a minimal reproduction:
- Create a subgraph with an Empty Latent Image with batch_size linked to
the Subgraph Input
- Export the workflow
- On load, the batch_size has step and min attributes which incorrectly
correspond to width
## Fix
There's multiple possible ways to address this and input on direction is
appreciated
- Fix links before loading graph
- Likely to break with any dynamic state
- Fix links, then load graph again
- Ugly, bad performance, dynamic state may require multiple passes to
correctly ripple
- In the Subgraph code, ignore target_slot and instead `.find()` input
with matching linkId (proposed)
- Promising, but means accepting that state is just wrong sometimes.
Another forever footgun.
- Entirely remove the input compression
- Some people may complain, and old workflows still need to be supported
- Only remove target_slot redirection inside subgraphs
- Creates ugly logical difference between what happens inside and
outside subgraphs.
- Still leaves old workflows broken
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7388-Remove-target_slot-compression-from-subgraph-exports-2c66d73d3650815d8c96c5047958ab67)
by [Unito](https://www.unito.io)
## Summary
Fixes the issue where unpacking a subgraph containing missing nodes
causes those nodes to disappear. Missing nodes are now automatically
restored as placeholder nodes that preserve their original data,
allowing them to be recovered when the node types are installed later.
## Changes
- **What**:
- Modified `multiClone()` to preserve missing nodes as serialized data
when creating subgraphs
- Added `skipMissingNodes` option to `unpackSubgraph()` method to
restore missing nodes as placeholder nodes instead of throwing errors
- Updated `useSubgraphOperations.unpackSubgraph()` to automatically
restore missing nodes as placeholders (removed confirmation dialog)
- Replaced deprecated `LiteGraph.cloneObject()` with `structuredClone()`
- Removed unused i18n keys and debugging logs
## Review Focus
- **Placeholder node restoration**: Missing nodes are restored using the
same mechanism as `LGraph.configure()` (creating `LGraphNode` with
`last_serialization` and `has_errors` flags). This ensures compatibility
with the existing missing node manager.
- **Performance**: Optimized `getMissingNodeTypes()` to check
`registered_node_types` first before attempting node creation, and uses
Set for O(1) duplicate checking.
- **Data preservation**: Missing nodes preserve their original type,
title, and serialized data in `last_serialization`, allowing automatic
recovery when node types are installed.
- **Backward compatibility**: The `skipMissingNodes` option defaults to
`false`, maintaining original behavior for other code paths. Only the
UI-level `unpackSubgraph()` always uses `skipMissingNodes: true`.
<!-- If this PR fixes an issue, uncomment and update the line below -->
<!-- Fixes#7279 -->
## Demo
Before:
https://github.com/user-attachments/assets/e0327d05-802d-4a64-a9db-4d174e185d82
After:
https://github.com/user-attachments/assets/37ab3140-0ada-480e-b9d5-fef8856f8b27
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7341-fix-unpacking-a-missing-node-causes-it-to-disappear-2c66d73d36508151ac6be70a7b2bc56d)
by [Unito](https://www.unito.io)
## Summary
Reduce lower level font definitions in most places. Default to Inter.
See #6912
## Review Focus
Comic Sans is still an option...
## Screenshots (if applicable)
<!-- Add screenshots or video recording to help explain your changes -->
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7220-Style-Font-Consistency-2c26d73d365081348f2dd8909dd9bb8f)
by [Unito](https://www.unito.io)
---------
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
Co-authored-by: github-actions <github-actions@github.com>
When moving an existing link with subgraphInput as source to a new node,
the prior link is removed, but the previous target node would not have
it's link property cleared.
Resolves#7225
Also re-enables several functional skipped tests.
It feels like I'm having to play whack-a-mole reimplementing the same
fixes on every permutation of subgraphIO links. I'd like to setup up a
unified test set that covers them all, but wouldn't want the added work
to further delay this fix.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7229-When-moving-subgraphInput-link-properly-disconnect-old-link-2c36d73d36508149aca0ce477fee5c9e)
by [Unito](https://www.unito.io)
Previously the color of a link would simply use the type of the target
slot and fallback to the type of the origin slot. When a connection is
made to a node that accepts the any type ('*'), the link has the green
color of an unknown type.
Instead, when a connection is made, the type of a link is now calculated
as the greatest common type of the source and destination. This means
that connections to reroutes are correctly colored.
| Before | After |
| ------ | ----- |
| <img width="360" alt="before"
src="https://github.com/user-attachments/assets/a5544730-e69a-4c85-af33-b303bb30ae71"
/>| <img width="360" alt="after"
src="https://github.com/user-attachments/assets/7d7b59fd-1b79-440b-a97d-a1657313c484"
/>|
The code for calculating common types already exists, it has simply been
moved into litegraph and given a more descriptive name.
Resolves#7196
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7211-Color-links-as-common-type-2c16d73d365081188460f6b5973db962)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Alexander Brown <drjkl@comfy.org>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
Before node v2, you could use getInputPos, getOutputPos. In node v2,
that is not possible.
## Summary
Want to get the position of the output / inputs on the graph like before
node v2.
## Changes
- **What**: <!-- Core functionality added/modified --> Added
LGraphNode.getSlotPosition
## Review Focus
## Screenshots (if applicable)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7042-Expose-LGraphNode-getSlotPosition-2ba6d73d36508195a0fed2d1fe2b64f6)
by [Unito](https://www.unito.io)
## Summary
Changes the layout store to treat node sizes as body-only measurements
while LiteGraph continues to reason about full heights. DOM-driven
updates are tagged with `LayoutSource.DOM`, which lets the store strip
the title height exactly once before persisting. That classification (a
new mutation source - `LayoutSource.DOM`) is accurate because those
mutations are triggered by the browser’s layout engine via
ResizeObserver, rather than by direct calls into the layout APIs (e.g.,
`moveNodeTo`, `useNodeDrag`). So all sources are:
- `LayoutSource.DOM`: browser layout/ResizeObserver measurements that
include the title bar
- `LayoutSource.Vue`: direct Vue-driven mutations routed through the
layout store
- `LayoutSource.Canvas`: legacy LiteGraph/canvas updates that will be
phased out over time
- `LayoutSource.External`: for multiplayer or syncing with a when going
online after making changes offline (in teams/workspace)
When layout state flows back into LiteGraph we add the title height just
in time for `liteNode.setSize`, so LiteGraph’s rendering stays
unchanged. This makes Vue node resizing and workflow persistence
deterministic - multiline widgets hold their dimensions across reloads
because every path that crosses the layout/LiteGraph boundary performs
the same normalization.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6966-normalize-height-at-sites-2b76d73d365081b6bcb4f4ce6a27663a)
by [Unito](https://www.unito.io)
---------
Co-authored-by: github-actions <github-actions@github.com>
This pull request introduces improvements to widget customization and UI
consistency in the application. The most notable changes are the
addition of support for icon classes in widget options, updates to
button rendering logic, and enhanced visual consistency for button
components.
Widget customization enhancements:
* Added an optional `iconClass` property to the `IWidgetOptions`
interface in `widgets.ts`, allowing widgets to specify custom icons.
UI and rendering updates:
* Updated `WidgetButton.vue` to render the widget label and, if
provided, an icon using the new `iconClass` option. Also standardized
button styling and label usage.
* Improved button styling in `WidgetRecordAudio.vue` for better visual
consistency with other components.
<img width="662" height="534" alt="Screenshot 2025-11-25 at 01 36 45"
src="https://github.com/user-attachments/assets/43bbe226-07fd-48be-9b98-78b08a726b1b"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6900-Style-button-widgets-2b66d73d3650818ebeadd9315a47ba0f)
by [Unito](https://www.unito.io)
## Summary
**Problem:** ensureCorrectLayoutScale scales up LG -> Vue. But doesn't
scale down from Vue -> LG.
**Solution:** Bi directional scaling.
**Bonus:** fix edge cases such as subgraphs, groups, and reroutes. Also,
set auto scale: true now that we 'preserve' LG scale.
**IMPORTANT:** useVueNodeResizeTracking.ts sets vue node height -
Litegraph.NODE_TITLE_HEIGHT on workflow load using a resize observer.
Reloading the page (loading a workflow) in Vue mode, will subtract
height each time. This can look like a problem caused by
ensureCorrectLayoutScale. It is not. Need to fix. Here was an attempt by
[removing the Litegraph.NODE_TITLE_HEIGHT
entirely](https://github.com/Comfy-Org/ComfyUI_frontend/pull/6643).
## Review Focus
Full lifecycle of loading workflows and switching between vue and lg.
Race conditions could be present. For example switching the mode using
keybind very fast.
## Screenshots (if applicable)
https://github.com/user-attachments/assets/5576b760-13a8-45b9-b8f7-64e1caf443c1https://github.com/user-attachments/assets/46d6f870-df76-4084-968a-53cb629fc123
---------
Co-authored-by: github-actions <github-actions@github.com>
## Summary
- Adds [Oxc linter](https://oxc.rs/docs/guide/usage/linter) as a dev
dependency
- Creates minimal `.oxlintrc.json` configuration file
- Integrates oxlint into the lint workflow (runs before ESLint)
- Adds `pnpm oxlint` script for standalone usage
- **NEW**: Adds
[eslint-plugin-oxlint](https://github.com/oxc-project/eslint-plugin-oxlint)
to disable redundant ESLint rules
- Updates `CLAUDE.md` documentation with oxlint command
## Motivation
Oxc is a high-performance Rust-based linter that is 50-100x faster than
ESLint. By integrating it into our lint workflow, we get:
- **Faster CI/CD pipelines** (5% improvement in this codebase)
- **Quicker local development feedback**
- **Additional code quality checks** that complement ESLint
- **Reduced duplicate work** by disabling ESLint rules that oxlint
already checks
## Changes
- **package.json**: Added `oxlint` and `eslint-plugin-oxlint` to
devDependencies, integrated into `lint`, `lint:fix`, and `lint:no-cache`
scripts
- **pnpm-workspace.yaml**: Added `eslint-plugin-oxlint` and
`mixpanel-browser` to catalog
- **eslint.config.ts**: Integrated `eslint-plugin-oxlint` to
automatically disable redundant ESLint rules
- **.oxlintrc.json**: Created minimal configuration file with schema
reference
- **CLAUDE.md**: Added `pnpm oxlint` to Quick Commands section
- **.gitignore**: Added `core` dump files
## CI/CD Performance Benchmark
Real-world CI/CD timing from GitHub Actions workflow runs:
### Baseline (ESLint only) - [Run
#18718911051](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18718911051)
- Run ESLint with auto-fix: **125s**
- Final validation (lint + format + knip): **16s**
- **Total: 141s**
### With Oxlint (oxlint + ESLint) - [Run
#18719037963](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18719037963)
- Run ESLint with auto-fix (includes oxlint): **118s**
- Final validation (includes oxlint + lint + format + knip): **16s**
- **Total: 134s**
### Results
✅ **7 seconds faster (5.0% improvement)** despite running an additional
linting pass
### Analysis
The oxlint integration actually **improves** CI/CD performance by ~5%.
This unexpected improvement is likely because:
1. **Oxlint catches issues early**: Some code that would have slowed
down ESLint's parsing/analysis is caught by oxlint first
2. **ESLint cache benefits**: The workflow uses `--cache`, and oxlint's
fast execution helps populate/validate the cache more efficiently
3. **Parallel processing**: Modern CI runners can overlap some of the
I/O operations between oxlint and ESLint
Even if oxlint added overhead, the value proposition would still be
strong given its additional code quality checks and local development
speed benefits. The fact that it actually speeds up the pipeline is a
bonus.
## eslint-plugin-oxlint Performance Impact
Benchmark comparing ESLint performance with and without
eslint-plugin-oxlint:
### Baseline (ESLint without plugin) - [Run
#18723242157](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18723242157)
- Run ESLint with auto-fix: **122s** (2m 2s)
- Final validation: **17s**
### With eslint-plugin-oxlint - [Run
#18723675903](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18723675903)
- Run ESLint with auto-fix: **129s** (2m 9s)
- Final validation: **12s**
### Results
**Performance: +7 seconds ESLint, -5 seconds validation (net +2
seconds)**
The eslint-plugin-oxlint integration has a **minimal performance
impact** (+2 seconds total). The slight increase in ESLint time is
likely due to the additional plugin configuration overhead, while the
validation step is faster because fewer redundant lint warnings need to
be processed.
### Benefits
The small performance cost is outweighed by important benefits:
1. **Prevents duplicate work**: Disables ~50 ESLint rules that oxlint
already checks (e.g., `no-constant-condition`, `no-debugger`,
`no-empty`, etc.)
2. **Reduces noise**: Eliminates redundant lint warnings from two tools
checking the same thing
3. **Cleaner workflow**: One authoritative source for each type of lint
check
4. **Best practice**: Recommended by the Oxc project for ESLint + oxlint
integration
5. **Consistent results**: Ensures both tools don't conflict or give
contradictory advice
## Usage
```bash
# Run oxlint standalone
pnpm oxlint
# Run full lint workflow (oxlint + ESLint)
pnpm lint
pnpm lint:fix
```
## Notes
- Oxlint now runs as part of the standard `pnpm lint` workflow
- The configuration uses minimal rules by default (Oxc's philosophy is
"catch erroneous or useless code without requiring any configurations by
default")
- Oxlint provides fast feedback while ESLint provides comprehensive
checks
- eslint-plugin-oxlint automatically manages rule conflicts between the
two tools
- Both tools complement each other in the linting pipeline
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6197-chore-Add-Oxc-linter-to-project-2946d73d3650818cbb55ef9c0abdb9b9)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: DrJKL <DrJKL0424@gmail.com>
## Summary
1. Add a `getOptionLabel` option to `ComboWidget` so users of it can map
of custom labels to widget values. (e.g., `"My Photo" ->
"my_photo_1235.png"`).
2. Utilize this ability in Cloud environment to map user uploaded
filenames to their corresponding input asset.
3. Copious unit tests to make sure I didn't (AFAIK) break anything
during the refactoring portion of development.
4. Bonus: Scope model browser to only show in cloud distributions until
it's released elsewhere; should prevent some undesired UI behavior if a
user accidentally enables the assetAPI.
## Review Focus
Widget code: please double check the work there.
## Screenshots (if applicable)
https://github.com/user-attachments/assets/a94b3203-c87f-4285-b692-479996859a5a
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6585-Feat-input-mapping-2a26d73d365081faa667e49892c8d45a)
by [Unito](https://www.unito.io)
- DOMWidgets have ownership reset after a `subgraphNode.clone()`.
- Fixes Ctrl+C on a subgraphNode with a prompted prompt making the
prompt disappear.
- alt + drag uses the copy/paste pathway that deeply clones subgraphs.
- Fixed dangling references on nodes in subgraphs by updating subgraph
ids before configuration.
- Attempt to recursively resolve disconnected proxyWidgets (Can matter
when subgraphs load out of order).
- Fix Right click -> clone creating linked copies of subgraphs.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6383-Multiple-fixes-related-to-copying-subgraphs-29b6d73d365081819671ced440dde327)
by [Unito](https://www.unito.io)
This pull request refactors how context menu items are contributed by
extensions in the LiteGraph-based canvas. The legacy monkey-patching
approach for adding context menu options is replaced by a new, explicit
API (`getCanvasMenuItems` and `getNodeMenuItems`) for extensions. A
compatibility layer is added to support legacy extensions and warn
developers about deprecated usage. The changes improve maintainability,
extension interoperability, and migration to the new context menu
system.
### Context Menu System Refactor
* Introduced a new API for extensions to contribute context menu items
via `getCanvasMenuItems` and `getNodeMenuItems` methods, replacing
legacy monkey-patching of `LGraphCanvas.prototype.getCanvasMenuOptions`.
Major extension files (`groupNode.ts`, `groupOptions.ts`,
`nodeTemplates.ts`) now use this new API.
[[1]](diffhunk://#diff-b29f141b89433027e7bb7cde57fad84f9e97ffbe5c58040d3e0fdb7905022917L1779-R1771)
[[2]](diffhunk://#diff-91169f3a27ff8974d5c8fc3346bd99c07bdfb5399984484630125fdd647ff02fL232-R239)
[[3]](diffhunk://#diff-04c18583d2dbfc013888e4c02fd432c250acbcecdef82bf7f6d9fd888e632a6eL447-R458)
* Added a compatibility layer (`legacyMenuCompat` in
`contextMenuCompat.ts`) to detect and warn when legacy monkey-patching
is used, and to extract legacy-added menu items for backward
compatibility.
[[1]](diffhunk://#diff-2b724cb107c04e290369fb927e2ae9fad03be9e617a7d4de2487deab89d0d018R2-R45)
[[2]](diffhunk://#diff-d3a8284ec16ae3f9512e33abe44ae653ed1aa45c9926485ef6270cc8d2b94ae6R1-R115)
### Extension Migration
* Refactored core extensions (`groupNode`, `groupOptions`, and
`nodeTemplates`) to implement the new context menu API, moving menu item
logic out of monkey-patched methods and into explicit extension methods.
[[1]](diffhunk://#diff-b29f141b89433027e7bb7cde57fad84f9e97ffbe5c58040d3e0fdb7905022917L1633-L1683)
[[2]](diffhunk://#diff-91169f3a27ff8974d5c8fc3346bd99c07bdfb5399984484630125fdd647ff02fL19-R77)
[[3]](diffhunk://#diff-04c18583d2dbfc013888e4c02fd432c250acbcecdef82bf7f6d9fd888e632a6eL366-R373)
### Type and Import Cleanup
* Updated imports for context menu types (`IContextMenuValue`) across
affected files for consistency with the new API.
[[1]](diffhunk://#diff-b29f141b89433027e7bb7cde57fad84f9e97ffbe5c58040d3e0fdb7905022917R4-L7)
[[2]](diffhunk://#diff-91169f3a27ff8974d5c8fc3346bd99c07bdfb5399984484630125fdd647ff02fL1-R11)
[[3]](diffhunk://#diff-04c18583d2dbfc013888e4c02fd432c250acbcecdef82bf7f6d9fd888e632a6eL2-R6)
[[4]](diffhunk://#diff-bde0dce9fe2403685d27b0e94a938c3d72824d02d01d1fd6167a0dddc6e585ddR10)
### Backward Compatibility and Migration Guidance
* The compatibility layer logs a deprecation warning to the console when
legacy monkey-patching is detected, helping developers migrate to the
new API.
---
These changes collectively modernize the context menu extension
mechanism, improve code clarity, and provide a migration path for legacy
extensions.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5993-Contextmenu-extension-migration-2876d73d3650813fae07c1141679637a)
by [Unito](https://www.unito.io)
---------
Co-authored-by: github-actions <github-actions@github.com>
## Summary
The private fields triggered an error in intitializing the
linkLayoutSync. Turns out that wasn't necessary anymore.
> [!NOTE]
> Edit: Doing some more investigation, it looks like the slot sync can
also be removed?
## Changes
- **What**: Converts JS private fields to typescript private, adds some
readonly declarations
- **What**: Removes the useLinkLayoutSync usage in useVueNodeLifecycle
- **What**: Removes the useSlotLayoutSync usage in useVueNodeLifecycle
## Review Focus
Was the sync doing something that wouldn't be caught in normal
usage/testing?
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6330-Fix-Vue-Litegraph-conversion-bug-with-Reroutes-2996d73d3650819ebb24e4aa2fc51c65)
by [Unito](https://www.unito.io)
`LLink.disconnect` is intended to cleanup only the link itself. #4800
mistakenly assumed that it would perform all required steps for
disconnection. Later, #5015 would partially resolve this by adding some
of the missing functionality into `LLink.disconnect`, but this still
left output cleanup unhandled and failed to call
`node.onConnectionsChanged`.
This PR instead moves the disconnection code to call the function that
already has robust handling for these items and removes the
no-longer-needed and potentially misleading workaround.
Resolves#6247
Also un-skipped several SubgraphIO tests. They appear to function fine.
I'm assuming the reasons for them being skipped have been resolved.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6258-Fix-disconnection-of-subgraphInput-links-2966d73d36508112ad1fe602cdcf461b)
by [Unito](https://www.unito.io)
Some virtual nodes (like get/set nodes) perform link redirection at
prompt resolution. The prior implementation incorrectly tried to return
the source of the virtual link after resolution, but this causes things
to break when the source of the virtual link is a subgraph IO.
Instead, this PR changes the code section to restart resolution from the
destination of the virtual link so that the existing subgraph boundary
resolution code is applied.
Also fix a bug with reconnection of complex/any types on
conversion to subgraph.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6135-Fix-link-resolution-of-virtual-nodes-2916d73d36508183b891ef6eb39bad4c)
by [Unito](https://www.unito.io)
Fixed the AUDIO_RECORD widget to display correctly in both rendering
modes:
- Removed conflicting AUDIO_RECORD registration from ComfyWidgets that
was blocking the custom widget implementation in uploadAudio.ts
extension
- Changed canvasOnly flags from true to false on both audioUIWidget and
recordWidget to enable Vue nodes rendering
- Added type override (recordWidget.type = 'audiorecord') after widget
creation to enable Vue component lookup while preserving LiteGraph
button rendering
- Removed unused IAudioRecordWidget type definition
The widget now works correctly:
- LiteGraph mode: Displays as a functional button
- Vue nodes mode: Displays full recording UI with waveform visualization
## Summary
<!-- One sentence describing what changed and why. -->
## Changes
- **What**: <!-- Core functionality added/modified -->
- **Breaking**: <!-- Any breaking changes (if none, remove this line)
-->
- **Dependencies**: <!-- New dependencies (if none, remove this line)
-->
## 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)
<!-- Add screenshots or video recording to help explain your changes -->
https://github.com/user-attachments/assets/cf6513d4-0a4b-4210-88e2-a948855b5206
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6094-fix-Enable-AUDIO_RECORD-widget-in-both-LiteGraph-and-Vue-nodes-modes-28e6d73d365081faa879f53e3e2dddad)
by [Unito](https://www.unito.io)
---------
Co-authored-by: bymyself <cbyrne@comfy.org>
## Summary
Added Vue mode guards to LiteGraph canvas to prevent dual event handling
between Vue node components and LiteGraph node event handlers. Fixes a
bug where the litegraph and vue positions become out of sync and then
there are effectively dead spots on the canvas (the user should still be
able to interact with the graph even if the positions desync - the only
real downside of desync should just be mispositioned nodes in the
serialized state).
Returns early only in `processNodeClick` and the node-related
(alt+click+drag node cloning) canvas handlers. In general, the LiteGraph
canvas still owns some events/interactions - but the node interactions
are fully owned by Vue when in Vue nodes mode.
## Changes
- **What**: Added `LiteGraph.vueNodesMode` checks in
[LGraphCanvas.ts](src/lib/litegraph/src/LGraphCanvas.ts) to skip native
event processing when Vue components handle interactions
- **Breaking**: This could have some side effects or break some
extensions if anything was relying on these. However, prior to this
change, things like `processNodeClick` should only have been getting the
click events in de-sync scenarios anyway (described in
[Summary](##Summary) section)
## Review Focus
Any possible side-effects you can think of.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6058-fix-LiteGraph-capturing-node-pointer-events-if-Vue-and-LG-node-positions-become-desynced-28c6d73d365081f389f8c72299c31b0b)
by [Unito](https://www.unito.io)
## Summary
Enhancing and further modernizing the UI, giving users more usable area
whilst keeping farmiliar positioning and feel of elements.
## Changes
- **What**: Significant restructure of the UI elements, changing
elements from large blocks to floating elements, updating:
- Side toolbar menu (floating style, supports small/normal mode,
combines to scroll on height overflow)
- Bottom tabs panel (floating style, tabs redesigned)
- Action bar (support for docking/undocking menu)
- Added login/user menu button to top right
- Restyled breadcrumbs (still collapse when overflows)
- Add litegraph support for fps info position (so it isn't covered by
the sidebar)
- **Breaking**:
- Removed various elements and added new ones, I have tested custom
sidebars, custom actions, etc but if scripts are inserting elements into
"other" elements they may have been (re)moved.
- Remove support for bottom menu
- Remove support for 2nd-row tabs
## Screenshots
<img width="1116" height="907" alt="ui"
src="https://github.com/user-attachments/assets/b040a215-67d3-4c88-8c4d-f402a16a34f6"
/>
https://github.com/user-attachments/assets/571dbda5-01ec-47e8-b235-ee1b88c93dd0
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5980-Floating-Menus-UI-rework-2866d73d3650810aac60cc1afe979b60)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: github-actions <github-actions@github.com>
This pull request introduces a new audio playback widget for node UIs
and integrates it into the node widget system. The main changes include
the implementation of the `WidgetAudioUI` component, its registration in
the widget registry, and updates to pass node data to the new widget.
Additionally, some logging was added for debugging purposes.
**Audio Widget Implementation and Integration:**
* Added a new `WidgetAudioUI.vue` component that provides audio playback
controls (play/pause, progress slider, volume, options) and loads audio
files from the server based on node data.
* Registered the new `WidgetAudioUI` component in the widget registry by
importing it and adding an entry for the `audioUI` type.
[[1]](diffhunk://#diff-c2a60954f7fdf638716fa1f83e437774d5250e9c99f3aa83c84a1c0e9cc5769bR21)
[[2]](diffhunk://#diff-c2a60954f7fdf638716fa1f83e437774d5250e9c99f3aa83c84a1c0e9cc5769bR112-R115)
* Updated `NodeWidgets.vue` to pass `nodeInfo` as the `node-data` prop
to widgets of type `audioUI`, enabling the widget to access
node-specific audio file information.
**Debugging and Logging:**
* Added logging of `nodeData` in `LGraphNode.vue` and
`WidgetAudioUI.vue` to help with debugging and understanding the data
structure.
[[1]](diffhunk://#diff-a7744614cf842e54416047326db79ad81f7c7ab7bfb66ae2b46f5c73ac7d47f2R188-R189)
[[2]](diffhunk://#diff-71cce190d74c6b5359288857ab9917caededb8cdf1a7e6377578b78aa32be2fcR1-R284)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5627-Vuenodes-audio-widgets-2716d73d365081fbbc06c1e6cf4ebf4d)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Arjan Singh <1598641+arjansingh@users.noreply.github.com>
Co-authored-by: filtered <176114999+webfiltered@users.noreply.github.com>
Co-authored-by: Christian Byrne <cbyrne@comfy.org>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
Co-authored-by: Jin Yi <jin12cc@gmail.com>
Co-authored-by: DrJKL <DrJKL@users.noreply.github.com>
Co-authored-by: Robin Huang <robin.j.huang@gmail.com>
Co-authored-by: github-actions <github-actions@github.com>
Implements droponcanvas functionality and a linkconnectoradapter
refactor.
- Drop on canvas (Shift and default) integrated via LinkConnector
‘dropped-on-canvas’ with proper CanvasPointerEvent.
- LinkConnector adapter: now wraps the live canvas linkConnector (no
duplicate state); added dropOnCanvas() helper.
- Tests: Playwright scenarios for Shift-drop context menu/searchbox,
pinned endpoint, type prefilter, and post-selection auto-connect
(browser_tests/tests/vueNodes/interactions/links/linkInteraction.spec.ts).
There are some followup PRs that will fix/refactor some more noncritical
things, like the terrible slotid, the number/string nodeid confusion,
etc.
https://github.com/Comfy-Org/ComfyUI_frontend/pull/5780 (snapping) <--
https://github.com/Comfy-Org/ComfyUI_frontend/pull/5898 (drop on canvas
+ linkconnectoradapter refactor) <--
https://github.com/Comfy-Org/ComfyUI_frontend/pull/5903 (fix reroute
snapping)
---------
Co-authored-by: github-actions <github-actions@github.com>
Co-authored-by: Christian Byrne <cbyrne@comfy.org>
Co-authored-by: GitHub Action <action@github.com>
