## Summary
Pass target tier to billing portal API for deep linking to Stripe's
subscription update confirmation screen when user has an active
subscription.
## Changes
- **What**: When a user with an active subscription clicks a tier in
PricingTable, pass the target tier (including billing cycle) to
`accessBillingPortal` which sends it as `target_tier` in the request
body. This enables the backend to create a Stripe billing portal deep
link directly to the subscription update confirmation screen.
- **Dependencies**: Requires comfy-api PR for `POST /customers/billing`
`target_tier` support
## Review Focus
- PricingTable now differentiates between new subscriptions (checkout
flow) and existing subscriptions (billing portal with deep link)
- Type derivation uses `Parameters<typeof
authStore.accessBillingPortal>[0]` to avoid duplicating the tier union
(matches codebase pattern)
- Registry types manually updated to include `target_tier` field (will
be regenerated when API is deployed)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7692-feat-pass-target-tier-to-billing-portal-for-subscription-updates-2d06d73d365081b38fe4c81e95dce58c)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Christian Byrne <cbyrne@comfy.org>
Co-authored-by: GitHub Action <action@github.com>
## Summary
- Adds "Update ComfyUI" menu item to Help Center for portable/localhost
environments
- Wires existing `/v2/manager/queue/update_comfyui` endpoint to the
frontend
- Only visible in non-desktop, non-cloud distributions (where Electron
update mechanism isn't available)
## Changes
1. **Service layer**: Added `updateComfyUI()` method to
`comfyManagerService.ts`
2. **UI**: Added menu item with download icon to
`HelpCenterMenuContent.vue`
3. **i18n**: Added translation key for the new menu item
## Context
The new Manager UI (v4) lost the ability to update ComfyUI core in
non-desktop environments. This restores that functionality by
integrating the existing manager endpoint into the Help Center menu.
## Test plan
- [ ] Verify menu item appears in portable/localhost environments
- [ ] Verify menu item does NOT appear in desktop (Electron)
environments
- [ ] Verify menu item does NOT appear in cloud environments
- [ ] Test clicking the menu item triggers update and reboot
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7578-feat-add-Update-ComfyUI-option-to-Help-Center-for-non-desktop-environments-2cc6d73d3650811e9e4fe55515f50333)
by [Unito](https://www.unito.io)
Prior to the release of subgraphs, there was a single graph accessed
through `app.graph`. Now that there's multiple graphs, there's a lot of
code that needs to be reviewed and potentially updated depending on if
it cares about nearby nodes, all nodes, or something else requiring
specific attention.
This was done by simply changing the type of `app.graph` to unknown so
the typechecker will complain about every place it's currently used.
References were then updated to `app.rootGraph` if the previous usage
was correct, or actually rewritten.
By not getting rid of `app.graph`, this change already ensures that
there's no loss of functionality for custom nodes, but the prior typing
of `app.graph` can always be restored if future dissuasion of
`app.graph` usage creates issues.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7399-Cleanup-app-graph-usage-2c76d73d365081178743dfdcf07f44d0)
by [Unito](https://www.unito.io)
Re-enables the system notification popup for cloud distribution,
allowing cloud devs to notify cloud users about new features and updates
without requiring a new release.
Cloud now fetches release notes from the "cloud" project (instead of
"comfyui") and uses the `cloud_version` field for version comparison.
Since cloud versions are git hashes rather than semver, a helper handles
both formats gracefully.
The "What's New" popup is enabled for cloud, while the update toast and
red dot indicator remain desktop-only since cloud auto-updates and
doesn't require user action.
You can test this by doing `pnpm dev:cloud` and you will see a
notification I added (for testing):
<img width="1891" height="2077" alt="image"
src="https://github.com/user-attachments/assets/6599a6dc-a3e1-406f-a22d-14262be1f258"
/>
Content is controlled by non-devs at cms.comfy.org.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7277-feat-Enable-system-notifications-on-cloud-2c46d73d365081bcb33cd79ec18faefe)
by [Unito](https://www.unito.io)
There's a warning toast shown if the frontend is considered out-of-date
(relative to the version in the requirements.txt of
comfyanonymous/ComfyUI). As a result, e2e tests run on older release
branches (e.g., when backporting or hotfixing) can sometimes trigger the
warning which obviously causes visual regression tests to fail. This PR
adds a hidden setting to disable the warning and sets it to `true` in
the e2e test fixtures.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7004-test-add-setting-to-ignore-version-compatibility-toast-warnings-in-e2e-tests-2b86d73d3650812d9e07f54a0c86b996)
by [Unito](https://www.unito.io)
## Summary
- Implement pagination for media assets history to handle large datasets
efficiently
- Add infinite scroll support with approach-end event handler
- Support offset parameter in history API for both V1 and V2 endpoints
## Changes
- Add offset parameter support to `api.getHistory()` method
- Update history fetchers (V1/V2) to include offset in API requests
- Implement `loadMoreHistory()` in assetsStore with pagination state
management
- Add `loadMore`, `hasMore`, and `isLoadingMore` to IAssetsProvider
interface
- Add approach-end handler in AssetsSidebarTab for infinite scroll
- Set BATCH_SIZE to 200 for efficient loading
## Implementation Improvements
Simplified offset-based pagination by removing unnecessary
reconciliation logic:
- Remove `reconcileHistory`, `taskItemsMap`, `lastKnownQueueIndex`
(offset is sufficient)
- Replace `assetItemsByPromptId` Map → `loadedIds` Set (store IDs only)
- Replace `findInsertionIndex` binary search → push + sort (faster for
batch operations)
- Replace `loadingPromise` → `isLoadingMore` boolean (simpler state
management)
- Fix memory leak by cleaning up Set together with array slice
## Test Plan
- [x] TypeScript compilation passes
- [x] ESLint and Prettier formatting applied
- [x] Test infinite scroll in media assets tab
- [x] Verify network requests include correct offset parameter
- [x] Confirm no duplicate items when loading more
🤖 Generated with [Claude Code](https://claude.ai/code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Fixes a race causing “No auth header available for session creation”
during sign‑in, by skipping the initial token refresh event, and
wrapping extension auth hooks with async error handling.
Sentry:
https://comfy-org.sentry.io/issues/6990347926/?alert_rule_id=1614600&project=4509681221369857
Context
- Error surfaced as an unhandled rejection when session creation was
triggered without a valid auth header.
- Triggers: both onAuthUserResolved and onAuthTokenRefreshed fired
during initial login.
- Pre‑fix, onIdTokenChanged treated the very first token emission as a
“refresh” as well, so two concurrent createSession() calls ran
back‑to‑back.
- One of those calls could land before a Firebase ID token existed, so
getAuthHeader() returned null → createSession threw “No auth header
available for session creation”.
Exact pre‑fix failure path
- src/extensions/core/cloudSessionCookie.ts
- onAuthUserResolved → useSessionCookie().createSession()
- onAuthTokenRefreshed → useSessionCookie().createSession()
- src/stores/firebaseAuthStore.ts
- onIdTokenChanged increments tokenRefreshTrigger even for the initial
token (treated as a refresh)
- getAuthHeader() → getIdToken() may be undefined briefly during
initialization
- src/platform/auth/session/useSessionCookie.ts
- createSession(): calls authStore.getAuthHeader(); if falsy, throws
Error('No auth header available for session creation')
What this PR changes
1) Skip initial token “refresh”
- Track lastTokenUserId and ignore the first onIdTokenChanged for a
user; only subsequent token changes count as refresh events.
- File: src/stores/firebaseAuthStore.ts
2) Wrap extension auth hooks with async error handling
- Use wrapWithErrorHandlingAsync for
onAuthUserResolved/onAuthTokenRefreshed/onAuthUserLogout callbacks to
avoid unhandled rejections.
- File: src/services/extensionService.ts
Result
- Eliminates the timing window where createSession() runs before
getIdToken() returns a token.
- Ensures any remaining errors are caught and reported instead of
surfacing as unhandled promise rejections.
Notes
- Lint and typecheck run clean (pnpm lint:fix && pnpm typecheck).
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6563-Fix-session-cookie-creation-race-dedupe-calls-skip-initial-token-refresh-wrap-extensio-2a16d73d365081ef8c22c5ac8cb948aa)
by [Unito](https://www.unito.io)
## 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)
## 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 -->
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6400-Cloud-tracking-v2-29c6d73d365081a1ae32e9337f510a9e)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Arjan Singh <arjan@comfy.org>
Resolving an executionId to a locatorId can fail if the current workflow
does not match the queued one. Rather than throwing an error, the
resolution functions have been changed to return undefined.
Of note, all consumers of these functions already had checks to ensure
the returned value is not undefined.
- Even the test for failure state already specified that it should
return instead of throwing an error.
This PR cleans up a frequent sentry error.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6460-On-failed-subgraph-resolution-return-undefined-29c6d73d365081c5860ed7d24a99414c)
by [Unito](https://www.unito.io)
## Summary
Changed history reconciliation to use `promptId` instead of `queueIndex`
to support cloud environments that don't have queue indices.
Refactored into pure functions with expressive naming.
## Changes
- Use `promptId` (unique UUID) for reconciliation instead of
`queueIndex` (not available in cloud)
- Extract `reconcileHistoryWithServer` with helpers: `extractPromptIds`,
`isAddedAfter`, `sortNewestFirst`
- Added 34 tests covering reconciliation edge cases, sorting, and limits
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6327-use-prompt-id-rather-than-queue-index-for-history-reconciliation-2996d73d3650818e9471dbeb53cc23bb)
by [Unito](https://www.unito.io)
## Summary
Implements server-side remote configuration to decouple runtime behavior
from build artifacts, enabling dynamic configuration updates without
redeployment.
## Technical Changes
- **Replaced** build-time constants (`__MIXPANEL_TOKEN__`,
`__BUILD_FLAGS__`) with runtime configuration loaded from
`/api/features`
- Configuration now sourced from `window.__CONFIG__` (hydrated from
`/api/features` endpoint)
- **Added** `src/platform/remoteConfig/` service that polls server
configuration every 30 seconds
- **Modified** application bootstrap sequence in `main.ts` to load
remote config before module initialization (required for cloud builds)
- **Removed** global constants: `__BUILD_FLAGS__`, `__MIXPANEL_TOKEN__`.
Runtime subscription enforcement toggle via `subscription_required` flag
- Server health alerts with variant-based severity rendering
(info/warning/error) via topbar badges
## Rationale
- **Build-once-deploy-anywhere**: Single immutable artifact promoted
through environments (staging → production)
- **Zero-downtime configuration**: Update behavior without rebuilding or
redeploying the application
- **Incident response**: Disable features or display alerts dynamically
in response to outages or degraded service
- **Instant rollback**: Revert configuration changes server-side without
artifact redeployment
- **Progressive delivery**: Enable A/B testing, canary releases, and
user/region-based configuration
- **Environment parity**: Eliminate configuration drift between staging
and production builds
- Decouples deployment cadence from configuration changes
- Enables GitOps workflows for configuration management separate from
code deployments
- Supports real-time operational control of client behavior
- Reduces build matrix complexity (no per-environment builds)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6246-change-cloud-feature-flags-to-be-loaded-dynamically-at-runtime-rather-than-set-in-build-2966d73d3650811cbb41c9093961037a)
by [Unito](https://www.unito.io)
## Summary
Currently, the "What's Changed" popup toast in bottom left appears after
updating if three conditions are true:
1. Using Desktop app
2. Don't have notifications disabled in settings
3. Have not seen/dismissed the notification before
Then the fourth condition is
4. At least 1 of the last 2 notifications is medium or high priority
However, we only ever show the most recent notification, so this logic
is flawed. In addition, it presents issues:
- When the changelog is first generated by AI, it is marked as "low"
priority until human review. But if the changelog _prior_ to that is
"medium" or "high", the AI-generated one might get shown anyway - which
frustrates the intended process.
There's also a bug fixed here concidentally where if the server only
returns a single entry, it is never shown (due to `slice(0, -1)` syntax
when checking priorities).
## Changes
- **What**: Updated Pinia release store to read `attention` from the
newest release only and expanded unit coverage for toast visibility
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5959-fix-what-s-changed-release-toast-attention-level-logic-2856d73d36508141b9b2d8d3b11153b2)
by [Unito](https://www.unito.io)
## Summary
Simplify default scripts. Filtering is still available to users, we can
revisit tagging or grouping later.
This fixes the issue where we had tests that were in the codebase but
never run because they weren't under `/src/components`
Also deletes the duplicate litegraph tests and their associated vitest
config file.
## Changes
- **What**: Test cleanup
## Review Focus
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5888-Tests-Vitest-configuration-cleanup-2806d73d36508197b800f68f0b028279)
by [Unito](https://www.unito.io)
## Summary
This reverts PR #5614 which moved VueFire persistence configuration to
initialization.
## Reason for Revert
It breaks Google SSO login with error:
```
useErrorHandling.ts:12 FirebaseError: Firebase: Error (auth/argument-error).
at createErrorInternal (index-c92d61ad.js:506:41)
at _assert (index-c92d61ad.js:512:15)
at _withDefaultResolver (index-c92d61ad.js:9237:5)
at signInWithPopup (index-c92d61ad.js:9457:30)
at executeAuthAction.createCustomer (firebaseAuthStore.ts:263:25)
at executeAuthAction (firebaseAuthStore.ts:223:28)
at Proxy.loginWithGoogle (firebaseAuthStore.ts:262:5)
at Proxy.wrappedAction (pinia.mjs:1405:26)
at useFirebaseAuthActions.ts:104:28
at Object.signInWithGoogle (useErrorHandling.ts:39:22)
```
## Changes
- Reverts commit ea4e57b60 "Move VueFire persistence configuration to
initialization (#5614)"
- Restores previous Firebase auth persistence behavior
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5729-Revert-Move-VueFire-persistence-configuration-to-initialization-5614-2776d73d3650814c9b80d9c67c852874)
by [Unito](https://www.unito.io)
## Summary
Add asset browser dialog integration for combo widgets with full
animation support and proper state management.
(Thank you Claude from saving me me from merge conflict hell on this
one.)
## Changes
- Widget integration: combo widgets now use AssetBrowserModal for
eligible asset types
- Dialog animations: added animateHide() for smooth close transitions
- Async operations: proper sequencing of widget updates and dialog
animations
- Service layer: added getAssetsForNodeType() and getAssetDetails()
methods
- Type safety: comprehensive TypeScript types and error handling
- Test coverage: unit tests for all new functionality
- Bonus: fixed the hardcoded labels in AssetFilterBar
Widget behavior:
- Shows asset browser button for eligible widgets when asset API enabled
- Handles asset selection with proper callback sequencing
- Maintains widget value updates and litegraph notification
## Review Focus
I will call out some stuff inline.
## Screenshots
https://github.com/user-attachments/assets/9d3a72cf-d2b0-445f-8022-4c49daa04637
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5629-feat-integrate-asset-browser-with-widget-system-2726d73d365081a9a98be9a2307aee0b)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
## Summary
- Replace custom `compareVersions()` with `semver.compare()`
- Replace custom `isSemVer()` with `semver.valid()`
- Remove deprecated version comparison functions from `formatUtil.ts`
- Update all version comparison logic across components and stores
- Fix tests to use semver mocking instead of formatUtil mocking
## Benefits
- **Industry standard**: Uses well-maintained, battle-tested `semver`
package
- **Better reliability**: Handles edge cases more robustly than custom
implementation
- **Consistent behavior**: All version comparisons now use the same
underlying logic
- **Type safety**: Better TypeScript support with proper semver types
Fixes#4787
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5653-refactor-Replace-manual-semantic-version-utilities-functions-with-semver-package-2736d73d365081fb8498ee11cbcc10e2)
by [Unito](https://www.unito.io)
---------
Co-authored-by: DrJKL <DrJKL@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
Reorganized custom nodes manager functionality from scattered technical
layers into a cohesive domain-focused module following [domain-driven
design](https://en.wikipedia.org/wiki/Domain-driven_design) principles.
## Changes
- **What**: Migrated all manager code from technical layers
(`src/components/`, `src/stores/`, etc.) to unified domain structure at
`src/workbench/extensions/manager/`
- **Breaking**: Import paths changed for all manager-related modules
(40+ files updated)
## Review Focus
Verify all import path updates are correct and no circular dependencies
introduced. Check that [Vue 3 composition
API](https://vuejs.org/guide/reusability/composables.html) patterns
remain consistent across relocated composables.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5662-refactor-Migrate-manager-code-to-DDD-structure-2736d73d3650812c87faf6ed0fffb196)
by [Unito](https://www.unito.io)
## Summary
- Enable `verbatimModuleSyntax` compiler option in TypeScript
configuration
- Update all type imports to use explicit `import type` syntax
- This change will Improve tree-shaking and bundler compatibility
## Motivation
The `verbatimModuleSyntax` option ensures that type-only imports are
explicitly marked with the `type` keyword. This:
- Makes import/export intentions clearer
- Improves tree-shaking by helping bundlers identify what can be safely
removed
- Ensures better compatibility with modern bundlers
- Follows TypeScript best practices for module syntax
## Changes
- Added `"verbatimModuleSyntax": true` to `tsconfig.json`
- Updated another 48+ files to use explicit `import type` syntax for
type-only imports
- No functional changes, only import/export syntax improvements
## Test Plan
- [x] TypeScript compilation passes
- [x] Build completes successfully
- [x] Tests pass
- [ ] No runtime behavior changes
🤖 Generated with [Claude Code](https://claude.ai/code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5533-feat-enable-verbatimModuleSyntax-in-TypeScript-config-26d6d73d36508190b424ef9b379b5130)
by [Unito](https://www.unito.io)
## Summary
Some users were authenticating successfully but their email addresses
weren't being extracted from the Firebase token. This happened because
we weren't explicitly requesting the email scope during OAuth
authentication.
While Firebase's default configuration includes basic profile info, it
doesn't guarantee email access for all account types - particularly
Google Workspace accounts with restrictive policies or users with
privacy-conscious settings.
[Github
Scopes](https://docs.github.com/en/apps/oauth-apps/building-oauth-apps/scopes-for-oauth-apps)
## Changes
Adding email scope for Google + Github social OAuth.
## Review Focus
N/A
## Screenshots (if applicable)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5638-Explicitly-add-email-scope-for-social-auth-login-2726d73d3650817ab356fc9c04f8641b)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Alexander Brown <drjkl@comfy.org>
* [refactor] create src/platform/assets
Per @christian-byrne's feedback. Just bringing this into the repo sooner to clean up from my feature branch
* [fix] code review feedback
* [refactor] Move update-related functionality to platform/updates domain
Reorganizes release management, version compatibility, and notification functionality
following Domain-Driven Design principles, mirroring VSCode's architecture pattern.
- Move releaseService.ts to platform/updates/common/
- Move releaseStore.ts to platform/updates/common/
- Move versionCompatibilityStore.ts to platform/updates/common/
- Move useFrontendVersionMismatchWarning.ts to platform/updates/common/
- Move toastStore.ts to platform/updates/common/
- Move ReleaseNotificationToast.vue to platform/updates/components/
- Move WhatsNewPopup.vue to platform/updates/components/
- Update 25+ import paths across codebase and tests
This creates a cohesive "updates" domain containing all functionality related to
software updates, version checking, release notifications, and user communication
about application state changes.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix imports
---------
Co-authored-by: Claude <noreply@anthropic.com>
* refactor: move settingStore to platform/settings
Move src/stores/settingStore.ts to src/platform/settings/settingStore.ts
to separate platform infrastructure from domain logic following DDD principles.
Updates all import references across ~70 files to maintain compatibility.
* fix: update remaining settingStore imports after rebase
* fix: complete remaining settingStore import updates
* fix: update vi.mock paths for settingStore in tests
Update all test files to mock the new settingStore location at
@/platform/settings/settingStore instead of @/stores/settingStore
* fix: resolve remaining settingStore imports and unused imports after rebase
* fix: update settingStore mock path in SelectionToolbox test
Fix vi.mock path from @/stores/settingStore to @/platform/settings/settingStore
to resolve failing Load3D viewer button test.
* refactor: complete comprehensive settings migration to platform layer
This commit completes the migration of all settings-related code to the platform layer
as part of the Domain-Driven Design (DDD) architecture refactoring.
- constants/coreSettings.ts → platform/settings/constants/coreSettings.ts
- types/settingTypes.ts → platform/settings/types.ts
- stores/settingStore.ts → platform/settings/settingStore.ts (already moved)
- composables/setting/useSettingUI.ts → platform/settings/composables/useSettingUI.ts
- composables/setting/useSettingSearch.ts → platform/settings/composables/useSettingSearch.ts
- composables/useLitegraphSettings.ts → platform/settings/composables/useLitegraphSettings.ts
- components/dialog/content/SettingDialogContent.vue → platform/settings/components/SettingDialogContent.vue
- components/dialog/content/setting/SettingItem.vue → platform/settings/components/SettingItem.vue
- components/dialog/content/setting/SettingGroup.vue → platform/settings/components/SettingGroup.vue
- components/dialog/content/setting/SettingsPanel.vue → platform/settings/components/SettingsPanel.vue
- components/dialog/content/setting/ColorPaletteMessage.vue → platform/settings/components/ColorPaletteMessage.vue
- components/dialog/content/setting/ExtensionPanel.vue → platform/settings/components/ExtensionPanel.vue
- components/dialog/content/setting/ServerConfigPanel.vue → platform/settings/components/ServerConfigPanel.vue
- ~100+ import statements updated across the codebase
- Test file imports corrected
- Component imports fixed in dialog service and command menubar
- Composable imports updated in GraphCanvas.vue
```
src/platform/settings/
├── components/ # All settings UI components
├── composables/ # Settings-related composables
├── constants/ # Core settings definitions
├── types.ts # Settings type definitions
└── settingStore.ts # Central settings state management
```
✅ TypeScript compilation successful
✅ All tests passing (settings store, search functionality, UI components)
✅ Production build successful
✅ Domain boundaries properly established
This migration consolidates all settings functionality into a cohesive platform domain,
improving maintainability and following DDD principles for better code organization.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: format and lint after rebase conflict resolution
* fix: update remaining import paths to platform settings
- Fix browser test import: extensionAPI.spec.ts
- Fix script import: collect-i18n-general.ts
- Complete settings migration import path updates
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
* [refactor] move workflow domain to its own folder
* [refactor] Fix workflow platform architecture organization
- Move workflow rendering functionality to renderer/thumbnail domain
- Rename ui folder to management for better semantic clarity
- Update all import paths to reflect proper domain boundaries
- Fix test imports to use new structure
Architecture improvements:
- rendering → renderer/thumbnail (belongs with other rendering logic)
- ui → management (better name for state management and UI integration)
This ensures proper separation of concerns and domain boundaries.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* [fix] Resolve circular dependency between nodeDefStore and subgraphStore
* [fix] Update browser test imports to use new workflow platform paths
---------
Co-authored-by: Claude <noreply@anthropic.com>
* [refactor] Improve renderer architecture organization
Building on PR #5388, this refines the renderer domain structure:
**Key improvements:**
- Group all transform utilities in `transform/` subdirectory for better cohesion
- Move canvas state to dedicated `renderer/core/canvas/` domain
- Consolidate coordinate system logic (TransformPane, useTransformState, sync utilities)
**File organization:**
- `renderer/core/canvas/canvasStore.ts` (was `stores/graphStore.ts`)
- `renderer/core/layout/transform/` contains all coordinate system utilities
- Transform sync utilities co-located with core transform logic
This creates clearer domain boundaries and groups related functionality
while building on the foundation established in PR #5388.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: Clean up linter-modified files
* Fix import paths and clean up unused imports after rebase
- Update all remaining @/stores/graphStore references to @/renderer/core/canvas/canvasStore
- Remove unused imports from selection toolbox components
- All tests pass, only reka-ui upstream issue remains in typecheck
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* [auto-fix] Apply ESLint and Prettier fixes
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
* [ci] ignore local browser tests files
this is where i have claude put its one off playwright scripts
* [feat] carve out path to call asset browser in combo widget
* [feat] use buttons on Model Loaders when Asset API setting is on
* [feat] add Comfy.Assets.UseAssetAPI to CORE_SETTINGS
* [feat] create AssetService
1. Add service for accessing new Asset API
2. Add fallback model paths logic so empty model directories appear for
the user.
3. Copious tests for them all.
Co-Authored-By: Claude <noreply@anthropic.com>
* [feat] switch between assets and file paths for model data
* [feat] ignore assets with "missing" tag
* [fix] formatting and style
* [fix] call assets API with the correct filters
* [feat] elminate unused modelPath code
* [fix] remove stray comment
* [fix] model manager api was not parsed correctly
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix: normalize pack IDs to fix version detection for disabled packs
When a pack is disabled, ComfyUI-Manager returns it with a version suffix
(e.g., "ComfyUI-GGUF@1_1_4") while enabled packs don't have this suffix.
This inconsistency caused disabled packs to incorrectly show as having
updates available even when they were on the latest version.
Changes:
- Add normalizePackId utility to consistently remove version suffixes
- Apply normalization in refreshInstalledList and WebSocket updates
- Use the utility across conflict detection and node help modules
- Ensure pack version info is preserved in the object's ver field
This fixes the "Update Available" indicator incorrectly showing for
disabled packs that are already on the latest version.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* feature: test code added
* test: packUtils test code added
* test: address PR review feedback for test
improvements
- Remove unnecessary .not.toThrow() assertion
in useManagerQueue test
- Add clarifying comments for version
normalization test logic
- Replace 'as any' with vi.mocked() for better
type safety
---------
Co-authored-by: Claude <noreply@anthropic.com>
* Implement subgraph publishing
* Add missing null check
* Fix subgraph blueprint display in workflows tab
* Fix demotion of subgraph blueprints on reload
* Update locales [skip ci]
* Update blueprint def on save, cleanup
* Fix skipped tracking on subgraph publish
When a subgraph is first published, it previously was not added to the
subgraphCache. This would cause deletion to fail until a reload occurred.
* Fix failing vite tests
A couple of tests that were mocking classes broke SubgraphBlueprint
inheritance. Since they aren't testing anythign related to subgraph
blueprints, the subgraph store is mocked as well.
* Make blueprint breadcrumb badge clickable
* Add confirmation for overwrite on publish
* Simplify blueprint badge naming
* Swap to promise.allSettled when fetching subgraphs
* Navigate into subgraph on blueprint edit
* Revert mission of value in blueprint breadcrumb
This was causing the blueprint badge to always display
* Misc code quality fixes
* Set subgraphNode title on blueprint add.
When a subgraph blueprint is added to the graph, the title of the
subgraphNode is now set to be the title of the blueprint.
NOTE: The name of the subgraph node when a blueprint is edited is left
unchanged. This may cause minor user confusion.
* Add "Delete Blueprint" option to breadcrumb
When editing a blueprint, the options provided for the root graph of the
breadcrumb included a Delete Workflow option. This still functioned for
deleting the current blueprint when selected, but didn't make sense. It
has been updated to instead describe that it deletes the current
blueprint
* Extract subgraph load code as function
* Fix subgraphs appearing in library after refresh
Subgraph nodes were hidden from the node library and context menu by
setting skip_list to true. Unfortunately, this causes them to be
mistakenly be caught and registered as vue nodes when a refresh is
performed. This is fixed by adding a check for skip_list.
* Add delete button and confirmation for deletion
* Use more specific warning for blueprint deletion
* At success toast on subgraph publish
Will return later to potentially add a node library link to the toast
* Don't apply subgraph context menu to normal nodes
Subgraph blueprints have a right click -> delete option in the node
library. This was incorrectly being dislplayed on non blueprint nodes.
* Remove hardcoded subgraphs path
Rather happy with this change. Rather than trying to introduce a
recursive import to pass a magic string, this solution is both
sufficient AND allows potential future extensions with less breakage.
* Fix nodeDef update on save
Wait to update the node def cache until after a blueprint has been
saved. Before, changes to links weren't actually being made visisble.
* Fix SaveAs with subgraph blueprints
* Remove ugly serialize/deserialize
Thought I had already tested this, and found that the mere existence of
proxies was causing issues, but simply adding a correct annotation is
sufficient now.
* Improve error specificity
* Framework for user defined blueprint descriptions
BlueprintDescription can be added to a workflows extra field to provide
more useful information about a blueprint's purpose
Actually hooking this up in a way that is user accessible is out of
scope for right now, but this will simplify future implementation.
* Cleanup breadcrumb dropdown options
Removes Dupliate for blueprints, adds a publish subgraph option.
The publish subgraph button currently routes through the save as logic.
Unforunately, this results in the prompt for name referencing workflows.
The cleanest way to resolve this is still being considered
* Move blueprint renaming into blueprint load
Blueprints should automatically set the name of the added node to the
filename when added. This mostly worked, but created uglier edgecases:
The subgraph itself wasn't renamed, and it would need to be
reimplemented to apply when editing a blueprint.
Instead, this is now applied when a subgraphBlueprint is first loaded.
This keeps all the logic routed through a single point
* Move saveAs prompt into workflow class
Ensures that the correct publish text is displayed when editing
blueprints without making an awful mess of imports
* Fix tests by making subgraphBlueprint internal
This has the added benefit of forcing better organization.
Reverts the useWorkflowThumbnail patch as it is no longer required.
* Add tests for subgraph blueprints
* Rewrite confirmation dialog
* Fix overwrite on publish new subgraph
1 is used as a placeholder size as -1 indicates the baking userFile is
temporary, not persisted, and therefore, not able to overwrite when
saved.
* When editing blueprint, tint background blue
* Fix blueprint tint at low LOD
* Set node source for blueprints to Blueprint
* Fix publish test
Making subgraph blueprints non temporary on publish made it so the
following load actually occurs. A mock has been added for this load.
* Fix multiple nits
* Further cleanup: error handling, and comments
* Fixing failing test cases
This also moves the bg tinting to a property of the workflow,
which makes things more extensible in the future.
* Fix temporary marking on publish.
The prior fix to allow overwrite of an existing blueprint on publish was
misguided. By marking a not-yet-loaded file as non-temporary, the load
performed prior to saving was actually fetching the file off disk and
discarding the existing changes. This additionally entirely prevented
publishing when a blueprint did not already exist with the current name.
To fix this, the blueprint is not marked as non-temporary until after
the load occurs. Note that this load is still required as it initializes
the change tracker state required for saving.
* Block unloading subgraph blueprints
Will need to be revisited if lazy loading is implemented, but this
requires solving some ugly sync/async issues.
---------
Co-authored-by: github-actions <github-actions@github.com>
* fix: Replace reactive feature flags with non-reactive approach
- Changed managerUIState from computed to getManagerUIState() function
- Ensures fresh computation on each call to avoid timing issues
- Updates all consumers to use the new function-based approach
- Fixes manager UI state determination issues
This change addresses the reactivity issues where feature flags were not
updating properly due to Vue's reactive system limitations with external
API values.
* fix: Add HelpCenter manager state handling and API version switching
- Fixed HelpCenter manager extension to check manager state
- Fixed 'Manager' command to respect manager state
- Added dynamic API prefix switching based on manager state
- Added debug logging for manager state determination
This ensures legacy manager uses /api/ prefix and new manager uses /api/v2/ prefix
* fix: Simplify manager state determination and fix API timing issues
- Remove unnecessary extension check from manager state store
- Use only feature flags (client and server) for state determination
- Default to NEW_UI when server flags not loaded (safer default)
- Fix ImportFailInfoBulk to not send empty requests
- Resolves initial 404 errors on installed API calls
* fix: Correct manager state determination for non-v4 servers
- Fix serverSupportsV4=false returning DISABLED instead of LEGACY_UI
- Server without v4 support should use legacy manager, not disable it
- Clarify condition for server v4 + client non-v4 case
* chore: Remove debug console.log statements
- Remove all debug logging from manager state store
- Remove logging from comfy manager service
- Clean up code for production
* test: Update manager state store tests to match new logic
- Update test expectations for server feature flags undefined case (returns NEW_UI)
- Update test expectations for server not supporting v4 case (returns LEGACY_UI)
- Tests now correctly reflect the actual behavior of the manager state logic
* fix: Remove dynamic API version handling in manager service
- Remove getApiBaseURL() function and axios interceptor
- Always use /api/v2/ for New Manager (hardcoded)
- Add isManagerServiceAvailable() to block service calls when not in NEW_UI state
- Simplify API handling as manager packages are now completely separated
* refactor: Add helper functions to managerStateStore for better code reuse
- Add isManagerEnabled(), isNewManagerUI(), isLegacyManagerUI() helpers
- Add shouldShowInstallButton(), shouldShowManagerButtons() for UI logic
- Update components to use helper functions where applicable
- Add comprehensive tests for new helper functions
- Centralize state checking logic to reduce duplication
* fix: Ensure SystemStats is loaded before conflict detection
- Move conflict detection from App.vue to GraphCanvas.vue
- Check manager state before running conflict detection
- Ensures SystemStats and feature flags are loaded first
- Prevents unnecessary API calls when manager is disabled
* docs: Clarify feature flag default behavior in manager state
- Add detailed comments explaining why NEW_UI is the default
- Clarify that undefined state is temporary during WebSocket connection
- Document graceful error handling when server doesn't support v2 API
* fix: Ensure consistent manager state handling for legacy commands
- Legacy commands now show error toast in NEW_UI mode
- Settings fallback for DISABLED mode
- Consistent error handling across all manager entry points
- Legacy commands only work in LEGACY_UI mode as expected
* refactor: centralize manager opening logic into managerStateStore
- Create openManager() function in managerStateStore to eliminate duplicate code
- Replace 8+ repeated switch statements across different files with single function
- Fix inconsistency where legacy command failure in LEGACY_UI mode incorrectly opened new manager
- Add support for legacy-only commands that should show error in NEW_UI mode
- Ensure all manager entry points behave consistently according to feature flags
- Clean up unused imports and fix ESLint errors
This addresses Christian's code review feedback about duplicate switch statements
and improves maintainability by providing a single source of truth for manager
opening logic.
* fix: use correct i18n import in managerStateStore
- Replace useI18n with direct t import from @/i18n
- Fixes issue where error messages showed as numbers (e.g. '26') instead of text
- Ensures toast messages display correctly in NEW_UI mode when legacy commands are invoked
* feature: initial tab fix
* test: Fix managerStateStore test failures by adding missing mocks
The test was failing because managerStateStore imports dialogService,
which imports ErrorDialogContent.vue that instantiates the app object.
This caused api.addEventListener errors in tests.
Added proper mocks for:
- dialogService
- commandStore
- toastStore
This prevents the problematic import chain and fixes the test failures.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor: convert managerStateStore to composable
- Move managerStateStore from store to composable pattern
- All functions are non-reactive utilities that don't need state management
- Follows Pinia guideline: "If it's not reactive, it shouldn't be in a store"
- Update all import paths across 8 files
- Move and update test file accordingly
This change improves architecture consistency as other utility functions
in the codebase also use composables rather than stores when reactivity
is not required.
* refactor: use readonly computed properties instead of getter methods
- Convert all getter methods to readonly computed properties
- Follows Vue conventions for better performance through caching
- Change access pattern from function calls to .value properties
- Update all usages across 6 files
- Thanks to @DrJKL for the suggestion
This improves performance by caching computed values and aligns
with Vue's reactive system patterns.
* fix: check isManagerEnabled check to GraphCanvas.vue to avoid the side-effects of calling useConflictDetection which include calling useComfyManagerStore
* chore: console.log to console.debug
* chore: useConflictDetection().initializeConflictDetection()
* test: add mockManagerDisabled option to disable manager in Playwright tests
- Add mockManagerDisabled parameter to ComfyPage.setup() (defaults to true)
- Override api.getServerFeature() to return false for manager feature flag
- Prevents manager initialization from interfering with subgraph tests
- Individual tests can still enable manager when needed by passing mockManagerDisabled: false
* chore: text modified
* fix: resolve CI/CD failures by fixing manager initialization timing
## Problem
GraphCanvas.vue was initializing conflict detection during component setup,
causing side effects in test environment where manager is disabled. This led
to 4 Playwright test failures in PR #5317.
## Root Cause
- GraphCanvas.vue called useConflictDetection() in setup phase
- This triggered store side effects even when manager was disabled
- systemStats wasn't ready when checking manager state
## Solution
1. Removed conflict detection initialization from GraphCanvas.vue entirely
2. Refactored systemStatsStore to use VueUse's useAsyncState pattern
3. Added isInitialized check in useManagerState to wait for systemStats
4. Updated useConflictDetection to check manager state internally
## Changes
- **GraphCanvas.vue**: Removed all conflict detection code
- **systemStatsStore**: Implemented useAsyncState for proper async handling
- **useManagerState**: Added isInitialized check before checking manager state
- **useConflictDetection**: Added internal manager state validation
- **App.vue**: Removed unnecessary fetchSystemStats calls
- **Tests**: Updated unit tests for new async behavior
## Test Results
All 4 previously failing Playwright tests now pass:
- featureFlags.spec.ts (feature flag handling)
- subgraph.spec.ts (breadcrumb updates, DOM cleanup)
- widget.spec.ts (image changes)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* chore: modified the note
* fix: test code modified
* fix: when manager is new manager ui, conflict detectetion should work
* fix: ensure fetch system stats before determine manager stats & when new ui & call legacy ui, open new manger dialog by default
* chore: unnecessary .value deleted & fetch name modified to refetch
* fix: ref type .value needed
* chore: vue use until pattern for waiting initializing
* fix: .value added
* fix: useManagerState test to properly mock reactive refs
The test was failing because it was mocking systemStats and isInitialized as plain values instead of reactive refs. The actual composable uses storeToRefs which returns refs with .value properties.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: when system stats initialized, use until(systemStatsStore.isInitialized)
* fix: test
---------
Co-authored-by: Claude <noreply@anthropic.com>
