Repository navigation
Conversation
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughAdds a Changesmodel_type Tag Feature Flag and Parsing
assetsStore Unconditional Model Cache
Sequence Diagram(s)sequenceDiagram
participant Component as AssetBrowserModal / useAssetBrowser
participant useFeatureFlags
participant assetMetadataUtils
participant filterByCategory
participant assetsStore
Component->>useFeatureFlags: flags.supportsModelTypeTags
useFeatureFlags-->>Component: boolean (server feature)
Component->>assetMetadataUtils: getAssetCategories(asset, modelTypeMode)
assetMetadataUtils-->>Component: stripped model_type values OR legacy tags
Component->>filterByCategory: filterByCategory(category, modelTypeMode)
filterByCategory->>assetMetadataUtils: getAssetCategories(asset, modelTypeMode).includes(category)
assetMetadataUtils-->>filterByCategory: boolean match
filterByCategory-->>Component: filtered assets
Component->>assetMetadataUtils: getAssetTypeBadge(asset, modelTypeMode)
assetMetadataUtils-->>Component: badge label (prefix-stripped when modelTypeMode=true)
Component->>assetMetadataUtils: stripModelTypePrefix(tagFromAssets)
assetMetadataUtils-->>Component: display title
Component->>assetsStore: updateModelsForTag / updateModelsForNodeType
assetsStore-->>Component: getAssets(category)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Suggested labels
Suggested reviewers
Poem
Caution Pre-merge checks failedPlease resolve all errors before merging. Addressing warnings is optional.
❌ Failed checks (1 error)
✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
🎭 Playwright: ✅ 1740 passed, 0 failed📊 Browser Reports
🎨 Storybook: ✅ Built — View Storybook📦 Bundle: 8.13 MB gzip 🔴 +4.88 kBDetailsSummary
Category Glance App Entry Points — 3.64 kB (baseline 3.64 kB) • ⚪ 0 BMain entry bundles and manifests
Status: 1 added / 1 removed Graph Workspace — 1.28 MB (baseline 1.28 MB) • ⚪ 0 BGraph editor runtime, canvas, workflow orchestration
Status: 1 added / 1 removed / 1 unchanged Views & Navigation — 112 kB (baseline 112 kB) • ⚪ 0 BTop-level views, pages, and routed surfaces
Status: 12 added / 12 removed / 4 unchanged Panels & Settings — 551 kB (baseline 551 kB) • ⚪ 0 BConfiguration panels, inspectors, and settings screens
Status: 11 added / 11 removed / 15 unchanged User & Accounts — 29.1 kB (baseline 29.1 kB) • ⚪ 0 BAuthentication, profile, and account management bundles
Status: 7 added / 7 removed / 3 unchanged Editors & Dialogs — 121 kB (baseline 121 kB) • ⚪ 0 BModals, dialogs, drawers, and in-app editors
Status: 5 added / 5 removed / 1 unchanged UI Components — 64.7 kB (baseline 64.7 kB) • ⚪ 0 BReusable component library chunks
Status: 6 added / 6 removed / 8 unchanged Data & Services — 3.39 MB (baseline 3.37 MB) • 🔴 +13.4 kBStores, services, APIs, and repositories
Status: 14 added / 14 removed / 3 unchanged Utilities & Hooks — 363 kB (baseline 357 kB) • 🔴 +5.58 kBHelpers, composables, and utility bundles
Status: 19 added / 19 removed / 16 unchanged Vendor & Third-Party — 15.7 MB (baseline 15.7 MB) • ⚪ 0 BExternal libraries and shared vendor chunks Status: 16 unchanged Other — 12.5 MB (baseline 12.5 MB) • 🔴 +34 BBundles that do not match a named category
Status: 79 added / 79 removed / 198 unchanged ⚡ Performance Report
✅ No regressions detected. All metrics
Historical variance (last 15 runs)
Trend (last 15 commits on main)
Raw data{
"timestamp": "2026-07-28T01:07:18.522Z",
"gitSha": "8791d3462ddb49da84a0f05d63dcdf2e49763e21",
"branch": "synap5e/fe-1076-fe-group-the-model-library-by-model_type-tags",
"measurements": [
{
"name": "canvas-idle",
"durationMs": 2004.4399999999882,
"styleRecalcs": 11,
"styleRecalcDurationMs": 5.635,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 400.086,
"heapDeltaBytes": -21734792,
"heapUsedBytes": 45694676,
"domNodes": 22,
"jsHeapTotalBytes": 19922944,
"scriptDurationMs": 12.323000000000002,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-idle",
"durationMs": 1997.829999999965,
"styleRecalcs": 10,
"styleRecalcDurationMs": 5.495,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 367.34299999999996,
"heapDeltaBytes": 3495132,
"heapUsedBytes": 71132452,
"domNodes": 20,
"jsHeapTotalBytes": 20582400,
"scriptDurationMs": 12.474,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66999999999998,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1775.322999999986,
"styleRecalcs": 72,
"styleRecalcDurationMs": 27.081,
"layouts": 12,
"layoutDurationMs": 3.2680000000000002,
"taskDurationMs": 752.918,
"heapDeltaBytes": -18241940,
"heapUsedBytes": 49426644,
"domNodes": -268,
"jsHeapTotalBytes": 20709376,
"scriptDurationMs": 87.027,
"eventListeners": -133,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1705.4529999999772,
"styleRecalcs": 73,
"styleRecalcDurationMs": 22.697,
"layouts": 12,
"layoutDurationMs": 2.313,
"taskDurationMs": 623.938,
"heapDeltaBytes": -19204412,
"heapUsedBytes": 48492944,
"domNodes": -263,
"jsHeapTotalBytes": 21233664,
"scriptDurationMs": 75.782,
"eventListeners": -131,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1670.9510000000023,
"styleRecalcs": 30,
"styleRecalcDurationMs": 12.416,
"layouts": 6,
"layoutDurationMs": 0.56,
"taskDurationMs": 291.499,
"heapDeltaBytes": 7219712,
"heapUsedBytes": 74783964,
"domNodes": 79,
"jsHeapTotalBytes": 20320256,
"scriptDurationMs": 13.870000000000001,
"eventListeners": 19,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1670.635999999945,
"styleRecalcs": 32,
"styleRecalcDurationMs": 11.639000000000001,
"layouts": 6,
"layoutDurationMs": 0.49699999999999994,
"taskDurationMs": 296.116,
"heapDeltaBytes": 7148568,
"heapUsedBytes": 74732600,
"domNodes": 79,
"jsHeapTotalBytes": 20058112,
"scriptDurationMs": 13.802000000000001,
"eventListeners": 19,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "dom-widget-clipping",
"durationMs": 472.92300000000864,
"styleRecalcs": 12,
"styleRecalcDurationMs": 5.1579999999999995,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 272.549,
"heapDeltaBytes": -12203188,
"heapUsedBytes": 55302696,
"domNodes": 20,
"jsHeapTotalBytes": 21106688,
"scriptDurationMs": 37.472,
"eventListeners": 2,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999727
},
{
"name": "dom-widget-clipping",
"durationMs": 473.47100000001774,
"styleRecalcs": 13,
"styleRecalcDurationMs": 5.457999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 265.274,
"heapDeltaBytes": -12261668,
"heapUsedBytes": 55457124,
"domNodes": 22,
"jsHeapTotalBytes": 21368832,
"scriptDurationMs": 35.511,
"eventListeners": 2,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-idle",
"durationMs": 2005.2109999999743,
"styleRecalcs": 10,
"styleRecalcDurationMs": 5.975000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 527.5540000000001,
"heapDeltaBytes": 7226204,
"heapUsedBytes": 66181096,
"domNodes": -267,
"jsHeapTotalBytes": 4788224,
"scriptDurationMs": 66.489,
"eventListeners": -129,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-idle",
"durationMs": 1998.3950000000164,
"styleRecalcs": 10,
"styleRecalcDurationMs": 5.878999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 525.081,
"heapDeltaBytes": 6564128,
"heapUsedBytes": 66121844,
"domNodes": -268,
"jsHeapTotalBytes": 5050368,
"scriptDurationMs": 63.584,
"eventListeners": -127,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-pan",
"durationMs": 2087.713000000008,
"styleRecalcs": 69,
"styleRecalcDurationMs": 11.583,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 955.0920000000001,
"heapDeltaBytes": 4100984,
"heapUsedBytes": 64234280,
"domNodes": -268,
"jsHeapTotalBytes": 4730880,
"scriptDurationMs": 294.161,
"eventListeners": -129,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-pan",
"durationMs": 2076.618999999994,
"styleRecalcs": 68,
"styleRecalcDurationMs": 9.297,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 886.7530000000002,
"heapDeltaBytes": 18266388,
"heapUsedBytes": 80636240,
"domNodes": -268,
"jsHeapTotalBytes": 4468736,
"scriptDurationMs": 265.35400000000004,
"eventListeners": -129,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "large-graph-zoom",
"durationMs": 3076.607000000024,
"styleRecalcs": 66,
"styleRecalcDurationMs": 13.185999999999998,
"layouts": 60,
"layoutDurationMs": 7.034999999999999,
"taskDurationMs": 1105.9170000000001,
"heapDeltaBytes": 19556172,
"heapUsedBytes": 80830536,
"domNodes": 16,
"jsHeapTotalBytes": 8126464,
"scriptDurationMs": 353.67199999999997,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "large-graph-zoom",
"durationMs": 3058.132999999998,
"styleRecalcs": 67,
"styleRecalcDurationMs": 14.589999999999998,
"layouts": 60,
"layoutDurationMs": 7.226,
"taskDurationMs": 1123.0500000000002,
"heapDeltaBytes": 20741132,
"heapUsedBytes": 84165032,
"domNodes": 20,
"jsHeapTotalBytes": 8388608,
"scriptDurationMs": 354.406,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66999999999998,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "minimap-idle",
"durationMs": 2017.5449999999842,
"styleRecalcs": 9,
"styleRecalcDurationMs": 5.44,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 522.944,
"heapDeltaBytes": 7194304,
"heapUsedBytes": 69438836,
"domNodes": -268,
"jsHeapTotalBytes": 4526080,
"scriptDurationMs": 64.72,
"eventListeners": -129,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "minimap-idle",
"durationMs": 2000.0210000000038,
"styleRecalcs": 9,
"styleRecalcDurationMs": 5.382999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 522.0409999999999,
"heapDeltaBytes": 6817608,
"heapUsedBytes": 69145556,
"domNodes": -269,
"jsHeapTotalBytes": 4526080,
"scriptDurationMs": 64.09300000000002,
"eventListeners": -127,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 487.9270000000133,
"styleRecalcs": 48,
"styleRecalcDurationMs": 7.0120000000000005,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 283.188,
"heapDeltaBytes": -11876376,
"heapUsedBytes": 55820256,
"domNodes": 22,
"jsHeapTotalBytes": 21630976,
"scriptDurationMs": 76.92900000000002,
"eventListeners": 8,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.663333333333338,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 455.882000000031,
"styleRecalcs": 48,
"styleRecalcDurationMs": 7.249999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 285.86899999999997,
"heapDeltaBytes": -11935076,
"heapUsedBytes": 55753596,
"domNodes": 22,
"jsHeapTotalBytes": 21893120,
"scriptDurationMs": 76.281,
"eventListeners": 8,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.663333333333338,
"p95FrameDurationMs": 16.700000000000273
},
{
"name": "subgraph-idle",
"durationMs": 2005.3340000000048,
"styleRecalcs": 11,
"styleRecalcDurationMs": 6.162000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 356.789,
"heapDeltaBytes": 3570876,
"heapUsedBytes": 71368408,
"domNodes": 22,
"jsHeapTotalBytes": 20582400,
"scriptDurationMs": 10.983999999999998,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-idle",
"durationMs": 2012.0430000000056,
"styleRecalcs": 10,
"styleRecalcDurationMs": 5.376,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 357.34299999999996,
"heapDeltaBytes": 3572580,
"heapUsedBytes": 71352712,
"domNodes": 20,
"jsHeapTotalBytes": 20844544,
"scriptDurationMs": 11.515,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1677.6370000000043,
"styleRecalcs": 75,
"styleRecalcDurationMs": 24.910999999999994,
"layouts": 16,
"layoutDurationMs": 3.236,
"taskDurationMs": 568.237,
"heapDeltaBytes": -5549372,
"heapUsedBytes": 62149132,
"domNodes": 61,
"jsHeapTotalBytes": 21106688,
"scriptDurationMs": 61.427,
"eventListeners": 4,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1671.6709999999466,
"styleRecalcs": 76,
"styleRecalcDurationMs": 23.241999999999997,
"layouts": 16,
"layoutDurationMs": 2.875,
"taskDurationMs": 553.884,
"heapDeltaBytes": -5706420,
"heapUsedBytes": 61850340,
"domNodes": 64,
"jsHeapTotalBytes": 21106688,
"scriptDurationMs": 57.63,
"eventListeners": 4,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-transition-enter",
"durationMs": 882.7709999999911,
"styleRecalcs": 18,
"styleRecalcDurationMs": 22.463000000000005,
"layouts": 14,
"layoutDurationMs": 8.253,
"taskDurationMs": 683.991,
"heapDeltaBytes": 6883428,
"heapUsedBytes": 98417436,
"domNodes": 13673,
"jsHeapTotalBytes": 12058624,
"scriptDurationMs": 19.037999999999993,
"eventListeners": 2369,
"totalBlockingTimeMs": 94,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "viewport-pan-sweep",
"durationMs": 8087.9180000000015,
"styleRecalcs": 249,
"styleRecalcDurationMs": 29.367,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 3244.421,
"heapDeltaBytes": 9245940,
"heapUsedBytes": 68578016,
"domNodes": -267,
"jsHeapTotalBytes": 6828032,
"scriptDurationMs": 917.5569999999999,
"eventListeners": -111,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "viewport-pan-sweep",
"durationMs": 8083.570000000009,
"styleRecalcs": 249,
"styleRecalcDurationMs": 29.386999999999997,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 3183.0830000000005,
"heapDeltaBytes": 9566896,
"heapUsedBytes": 68536480,
"domNodes": -265,
"jsHeapTotalBytes": 6828032,
"scriptDurationMs": 916.015,
"eventListeners": -111,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.670000000000012,
"p95FrameDurationMs": 16.80000000000109
},
{
"name": "vue-large-graph-idle",
"durationMs": 11111.473999999987,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 11081.486,
"heapDeltaBytes": -35264000,
"heapUsedBytes": 173684376,
"domNodes": -8311,
"jsHeapTotalBytes": -9113600,
"scriptDurationMs": 424.196,
"eventListeners": -16388,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333326,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "vue-large-graph-idle",
"durationMs": 11146.890999999981,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 11134.794,
"heapDeltaBytes": -61318172,
"heapUsedBytes": 166842476,
"domNodes": -8311,
"jsHeapTotalBytes": -13279232,
"scriptDurationMs": 416.45899999999995,
"eventListeners": -16388,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "vue-large-graph-pan",
"durationMs": 14379.713000000038,
"styleRecalcs": 70,
"styleRecalcDurationMs": 12.925999999999993,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 14361.208999999999,
"heapDeltaBytes": -51505764,
"heapUsedBytes": 156639916,
"domNodes": -8315,
"jsHeapTotalBytes": -15142912,
"scriptDurationMs": 707.3929999999999,
"eventListeners": -16382,
"totalBlockingTimeMs": 8,
"frameDurationMs": 17.223333333333358,
"p95FrameDurationMs": 16.80000000000291
},
{
"name": "vue-large-graph-pan",
"durationMs": 14266.965000000027,
"styleRecalcs": 68,
"styleRecalcDurationMs": 12.625999999999998,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 14248.301000000001,
"heapDeltaBytes": -55097416,
"heapUsedBytes": 154204772,
"domNodes": -8311,
"jsHeapTotalBytes": -17764352,
"scriptDurationMs": 734.637,
"eventListeners": -16382,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.223333333333358,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "workflow-execution",
"durationMs": 436.15100000005214,
"styleRecalcs": 15,
"styleRecalcDurationMs": 13.031,
"layouts": 3,
"layoutDurationMs": 0.54,
"taskDurationMs": 92.138,
"heapDeltaBytes": -15777684,
"heapUsedBytes": 51167312,
"domNodes": 134,
"jsHeapTotalBytes": 7999488,
"scriptDurationMs": 6.981000000000001,
"eventListeners": 65,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666682,
"p95FrameDurationMs": 16.700000000000273
},
{
"name": "workflow-execution",
"durationMs": 452.4269999999433,
"styleRecalcs": 16,
"styleRecalcDurationMs": 18.265000000000004,
"layouts": 4,
"layoutDurationMs": 1.3940000000000001,
"taskDurationMs": 103.313,
"heapDeltaBytes": -16224680,
"heapUsedBytes": 50600384,
"domNodes": 121,
"jsHeapTotalBytes": 7475200,
"scriptDurationMs": 6.8690000000000015,
"eventListeners": 67,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.663333333333338,
"p95FrameDurationMs": 16.700000000000273
}
]
} |
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
b9bef2a to
921514f
Compare
🔗 Website PreviewWebsite Preview: https://comfy-website-preview-pr-12948.vercel.app This commit: https://website-frontend-r5irhmo04-comfyui.vercel.app Last updated: 2026-06-19T09:57:09Z for |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 921514f806
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
24ed6a7 to
4d20b2b
Compare
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/platform/assets/utils/resolveModelNodeFromAsset.test.ts`:
- Around line 60-64: The beforeEach hook mocks console.error but never restores
it after each test, causing the mock to leak across subsequent tests. Add an
afterEach hook that calls vi.restoreAllMocks() to ensure all spies and mocks are
properly restored to their original implementations after each test completes,
preventing global state contamination.
In `@src/platform/assets/utils/resolveModelNodeFromAsset.ts`:
- Around line 50-113: The error.message fields returned in the error objects
throughout the resolveModelNodeFromAsset function are hardcoded English strings,
but they should be internationalized using vue-i18n as per coding guidelines.
Add all user-facing error message strings to src/locales/en/main.json and update
each error return statement (the ones returning INVALID_ASSET, NO_PROVIDER, and
other error codes) to use i18n translation keys instead of hardcoded text
strings, following the same pattern used elsewhere in the codebase for UI-facing
messages.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro Plus
Run ID: 869fd5d3-3eab-4cbc-b742-678998394f9b
📒 Files selected for processing (13)
src/composables/useFeatureFlags.tssrc/platform/assets/components/AssetBrowserModal.test.tssrc/platform/assets/components/AssetBrowserModal.vuesrc/platform/assets/composables/useAssetBrowser.test.tssrc/platform/assets/composables/useAssetBrowser.tssrc/platform/assets/utils/assetFilterUtils.test.tssrc/platform/assets/utils/assetFilterUtils.tssrc/platform/assets/utils/assetMetadataUtils.test.tssrc/platform/assets/utils/assetMetadataUtils.tssrc/platform/assets/utils/resolveModelNodeFromAsset.test.tssrc/platform/assets/utils/resolveModelNodeFromAsset.tssrc/stores/assetsStore.test.tssrc/stores/assetsStore.ts
🚧 Files skipped from review as they are similar to previous changes (11)
- src/platform/assets/utils/assetFilterUtils.ts
- src/composables/useFeatureFlags.ts
- src/platform/assets/utils/assetFilterUtils.test.ts
- src/stores/assetsStore.test.ts
- src/platform/assets/composables/useAssetBrowser.test.ts
- src/platform/assets/composables/useAssetBrowser.ts
- src/platform/assets/components/AssetBrowserModal.vue
- src/platform/assets/utils/assetMetadataUtils.test.ts
- src/platform/assets/components/AssetBrowserModal.test.ts
- src/platform/assets/utils/assetMetadataUtils.ts
- src/stores/assetsStore.ts
There was a problem hiding this comment.
Caution
Inline review comments failed to post. This is likely due to GitHub's internal server error or limits when posting large numbers of comments. If you are seeing this consistently it is likely a permissions issue. Please check "Moderation" -> "Code review limits" under your organization settings.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/platform/assets/utils/resolveModelNodeFromAsset.test.ts`:
- Around line 60-64: The beforeEach hook mocks console.error but never restores
it after each test, causing the mock to leak across subsequent tests. Add an
afterEach hook that calls vi.restoreAllMocks() to ensure all spies and mocks are
properly restored to their original implementations after each test completes,
preventing global state contamination.
In `@src/platform/assets/utils/resolveModelNodeFromAsset.ts`:
- Around line 50-113: The error.message fields returned in the error objects
throughout the resolveModelNodeFromAsset function are hardcoded English strings,
but they should be internationalized using vue-i18n as per coding guidelines.
Add all user-facing error message strings to src/locales/en/main.json and update
each error return statement (the ones returning INVALID_ASSET, NO_PROVIDER, and
other error codes) to use i18n translation keys instead of hardcoded text
strings, following the same pattern used elsewhere in the codebase for UI-facing
messages.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro Plus
Run ID: 869fd5d3-3eab-4cbc-b742-678998394f9b
📒 Files selected for processing (13)
src/composables/useFeatureFlags.tssrc/platform/assets/components/AssetBrowserModal.test.tssrc/platform/assets/components/AssetBrowserModal.vuesrc/platform/assets/composables/useAssetBrowser.test.tssrc/platform/assets/composables/useAssetBrowser.tssrc/platform/assets/utils/assetFilterUtils.test.tssrc/platform/assets/utils/assetFilterUtils.tssrc/platform/assets/utils/assetMetadataUtils.test.tssrc/platform/assets/utils/assetMetadataUtils.tssrc/platform/assets/utils/resolveModelNodeFromAsset.test.tssrc/platform/assets/utils/resolveModelNodeFromAsset.tssrc/stores/assetsStore.test.tssrc/stores/assetsStore.ts
🚧 Files skipped from review as they are similar to previous changes (11)
- src/platform/assets/utils/assetFilterUtils.ts
- src/composables/useFeatureFlags.ts
- src/platform/assets/utils/assetFilterUtils.test.ts
- src/stores/assetsStore.test.ts
- src/platform/assets/composables/useAssetBrowser.test.ts
- src/platform/assets/composables/useAssetBrowser.ts
- src/platform/assets/components/AssetBrowserModal.vue
- src/platform/assets/utils/assetMetadataUtils.test.ts
- src/platform/assets/components/AssetBrowserModal.test.ts
- src/platform/assets/utils/assetMetadataUtils.ts
- src/stores/assetsStore.ts
🛑 Comments failed to post (2)
src/platform/assets/utils/resolveModelNodeFromAsset.test.ts (1)
60-64: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Restore the
console.errorspy to prevent global mock leakage across tests.
vi.clearAllMocks()only clears call history; it does not restore original
implementations. AddafterEach(() => vi.restoreAllMocks())(or explicitly
restore this spy) so subsequent tests don’t inherit a mocked global console.Suggested fix
import { beforeEach, describe, expect, it, vi } from 'vitest' +import { afterEach } from 'vitest' @@ describe('resolveModelNodeFromAsset', () => { beforeEach(() => { vi.clearAllMocks() vi.spyOn(console, 'error').mockImplementation(() => {}) mockSupportsModelTypeTags.value = false }) + + afterEach(() => { + vi.restoreAllMocks() + })As per coding guidelines for
**/*.test.ts, keep test mocks contained and avoid leaking mutable global state across tests.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/platform/assets/utils/resolveModelNodeFromAsset.test.ts` around lines 60 - 64, The beforeEach hook mocks console.error but never restores it after each test, causing the mock to leak across subsequent tests. Add an afterEach hook that calls vi.restoreAllMocks() to ensure all spies and mocks are properly restored to their original implementations after each test completes, preventing global state contamination.Source: Coding guidelines
src/platform/assets/utils/resolveModelNodeFromAsset.ts (1)
50-113: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Localize resolver error messages instead of hardcoding English text.
The returned
error.messagefields are part of the UI-facing result contract, but
the new messages are hardcoded strings. Please route these through i18n (or
return message keys and translate at the presentation layer).As per coding guidelines,
src/**/*.{ts,tsx,vue}must “Usevue-i18nfor ALL user-facing strings, configured insrc/locales/en/main.json.”🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/platform/assets/utils/resolveModelNodeFromAsset.ts` around lines 50 - 113, The error.message fields returned in the error objects throughout the resolveModelNodeFromAsset function are hardcoded English strings, but they should be internationalized using vue-i18n as per coding guidelines. Add all user-facing error message strings to src/locales/en/main.json and update each error return statement (the ones returning INVALID_ASSET, NO_PROVIDER, and other error codes) to use i18n translation keys instead of hardcoded text strings, following the same pattern used elsewhere in the codebase for UI-facing messages.Source: Coding guidelines
Codecov Report❌ Patch coverage is
@@ Coverage Diff @@
## main #12948 +/- ##
==========================================
- Coverage 79.11% 78.86% -0.25%
==========================================
Files 1697 1742 +45
Lines 95330 97996 +2666
Branches 29736 31279 +1543
==========================================
+ Hits 75416 77281 +1865
- Misses 19541 20320 +779
- Partials 373 395 +22
Flags with carried forward coverage won't be shown. Click here to find out more.
... and 106 files with indirect coverage changes 🚀 New features to boost your workflow:
|
|
Thanks @coderabbitai — responses to the two items:
(Created by Claude Code) |
|
Tip For best results, initiate chat on the files or code changes.
|
eec9abb to
f7d672e
Compare
There was a problem hiding this comment.
🔍 Cursor Review — Consolidated panel
Triggered by @mattmillerai.
Found 7 finding(s).
| Severity | Count |
|---|---|
| 🟠 High | 2 |
| 🟡 Medium | 3 |
| 🟢 Low | 2 |
Panel: 8/8 reviewers contributed findings.
An asset tagged with both `model_type:LLM` and a bare `LLM` yielded the same category twice in the candidate list; deduplicate the tier-1 values so a provider is only ever tried once per distinct category. Flat output pagination stopped on the first all-duplicate page, so a pagination window shifted by concurrent writes could hide later unseen assets. Reuse the model-category walk's no-progress tolerance for both walks instead of duplicating the counter, letting a bounded run of all-duplicate pages be traversed while still guaranteeing termination.
99a4398 to
8c3b946
Compare
|
On the twin-sync question (consolidating here as you suggested), I checked both backends. Cloud (BE-1480): synchronous, same transaction. The bare directory tag and its Core (ComfyUI#14511, That divergence does not break the core sidebar, because the sidebar never fetches per-folder by the bare tag. So the invariant the sidebar leans on is not "the bare twin exists", it is "core advertises The one thing this makes load-bearing is that the flag read stays reliable, which is exactly the bucket-cache memoization point you raised on #13464 (a stale Posted by Claude Code on Simon's behalf. |
|
Force-pushed this branch (old tip What changed in the reviewed code:
What did not change: every feature change in this PR is preserved verbatim, the non-cloud model store, Posted by Claude Code on Simon's behalf. |
christian-byrne
left a comment
There was a problem hiding this comment.
Verified at head 8c3b946. Leaving these as COMMENT-level notes — not approving (Christian holds that).
issue (contract) — assetFilterUtils.ts:36-37, 48. filterByOwnership gates my-models on asset.is_immutable === false and public-models on === true; filterItemByOwnership does item.is_immutable === isPublic. But is_immutable is z.boolean().optional() — a backend that omits the field (core's exclude_none, or any response that doesn't set it) makes the asset satisfy neither === false nor === true, so it drops out of both ownership views while still appearing under all. Dormant while the private-models UI is hidden, but the filter logic itself is unconditional. Suggest a presence-aware check — e.g. branch explicitly on is_immutable == null, or treat missing as owned (is_immutable !== true for my-models, is_immutable === true for public).
note — getAssetModelFolders (assetService.ts:~354). The folder-tag filter is tag !== MODELS_TAG && !blacklistedDirectories.has(tag); it doesn't strip model_type: tags. Under a namespaced backend (supports_model_type_tags), every model_type:<x> tag would surface as a spurious folder in the legacy node-graph model dropdown. A one-line && !tag.startsWith('model_type:') (or reuse of MODEL_TYPE_TAG_PREFIX) closes it. Flagging as a note since this path isn't edited by the PR but becomes newly reachable once the backend flag ships.
pagination. The batch-walk rework looks sound — the walk matches the merged #12827 cursor walk and the termination conditions (short page + no-progress backstop) are covered by the store tests. No concerns there.
(Skipping the MODEL_TYPE_TAG_PREFIX / 'model_type:' duplication nit — already covered by the constants-module threads on assetMetadataUtils.ts:7 and assetsStore.ts:415.)
| * still routes by its bare tags. When false (the default) categories come from | ||
| * the legacy bare-tag top-level grouping and `model_type:` is ignored. | ||
| */ | ||
| export function getAssetCategories( |
There was a problem hiding this comment.
suggestion (non-blocking, naming-consistency): one grouping-key value goes by four+ nouns across the stack.
The stripped model_type:* value (the folder a model groups under) is referred to as:
- category —
getAssetCategories,getPrimaryCategoryTag,getBareTagCategories,selectedCategory/typeCategories(browser layer) - model type —
getEditableModelType,getPrimaryModelType,modelTypeMode(this file / edit dropdown) - type badge —
getAssetTypeBadges(cards) - folder / bucket key —
modelFolderFromTag,buildModelBuckets/modelBuckets(service layer, added in feat(assets): Support the models sidebar under assets mode, and make this default for non-cloud #13464)
They're the same post-strip value: getAssetTypeBadges' own docstring says it returns "the same keys it groups under (getAssetCategories)", and getEditableModelType is documented as "the folder_name shown as the asset's current model type" — collapsing folder/model-type/category in one sentence.
The UI-vs-data-vs-domain layer split (category/badge vs folder/bucket vs model type) is defensible, so this isn't a blocker. The reason I'm flagging it for the cutover: when supports_model_type_tags wins and the legacy bare-tag branching is deleted, a cleanup that greps modelType/ModelType will silently miss every category/folder/bucket spelling of the identical value. Worth either a one-line alias/comment pinning "these are one value" or standardizing on a single noun for the grouping key, so the eventual removal stays a mechanical find-replace.
For contrast, the transitional flag itself is already clean: the modelTypeMode parameter is spelled identically everywhere and always sourced from flags.supportsModelTypeTags — that one will find-replace cleanly. This is just about the surrounding key vocabulary.
There was a problem hiding this comment.
Agreed the grouping-key noun sprawls (category / model type / badge / folder for one post-strip value). Tracked as FE-1400 so the post-cutover cleanup stays a mechanical find-replace rather than riding this PR.
Created by Claude Code.
christian-byrne
left a comment
There was a problem hiding this comment.
Approving after a manual read of the full diff. The pagination rework is sound: I diffed the loadBatches cursor walk against origin/main and the termination machinery (seenCursors / MAX_PAGINATION_BATCHES / walkCompleted) matches #12827's already-merged walk; the change here is the isCloud-gate removal plus the round-3 error hardening. Termination holds, nothing is dropped, AC3 is genuinely fixed by client-side bucketing, and the ModelInfoPanel write path works.
To be explicit: this approval does not waive the non-blocking review comments still open on this PR (upload-button a11y, is_immutable strict-equality dropping field-omitting assets, the getAssetModelFolders model_type: leak into the legacy dropdown, and the grouping-key naming sprawl in assetMetadataUtils). Pick those up in your revisit pass, they just aren't merge blockers. Green from me.
Namespaced model_type: tags surfaced as spurious folders (and duplicate categories) in the classic Model Library dropdown once the backend flag ships. Share MODEL_TYPE_TAG_PREFIX from assetMetadataUtils so the folder scan and category grouping filter on one constant.
is_immutable is optional in the asset schema; strict === checks dropped field-omitting assets from both my-models and public-models while they still appeared under all. Treat a missing flag as owned.
The icon-mode upload button CSS-hides its label and SearchInput exposed no accessible name, failing WCAG 2.1 SC 4.1.2. Forward an aria-label prop through SearchInput to its inner input.
christian-byrne
left a comment
There was a problem hiding this comment.
Re-approving at the new head. The three commits since my approval resolve three of the review comments I left, and each checks out against the diff:
1ba23b543(D4 a11y): SearchInput now takes anariaLabelprop wired to the input, and the asset browser setsaria-labelon both the search ("Search models") and the upload button ("Upload model"), so the icon-only upload button and the search field now have accessible names.b7c0e2dcd(P10):getAssetModelFoldersnow filters outmodel_type:-prefixed tags via the sharedMODEL_TYPE_TAG_PREFIX, so namespaced tags no longer leak into the legacy folder dropdown, with a test asserting it.61210a20e(P9):filterByOwnershipnow keepsis_immutable-omitting assets, covered by both the unit table (is_immutable: undefined -> kept) and the property test (fc.option(..., { nil: undefined })).
No regressions in the delta. Still green.
…rg#12972) ## Summary The FE writes `model_type:<folder_name>` (from `/api/experiment/models` discovery) on model import and on model-type edits, gated on the `supports_model_type_tags` backend capability. Default off = today's bare directory-path write, so this is a no-op until the backend declares the cutover. Note: Comfy-Org#13302 (stacked on top of this) disables model type editing on Core for now, so the edit-path write is cloud-only in practice. The import write path stays live on core once a core backend advertises the flag, which is safe: an import creates the file in the folder its tag names, so no move-on-retag divergence can arise. ## Changes - **What**: - `assetMetadataUtils.ts`: export `MODEL_TYPE_TAG_PREFIX`; add pure helpers `toModelTypeTag`, `getEditableModelType`, `buildModelTypeTagUpdate` (flag-param, mirroring FE-1076's `modelTypeMode`). - Import (`useUploadModelWizard.ts`) and edit-type (`ModelInfoPanel.vue`) write paths prefix the selected folder_name when the flag is on; the edit dropdown read is prefix-aware so it round-trips. `user_metadata.model_type` stays bare. - Gated on the same `supports_model_type_tags` signal FE-1076 grouping reads (`useFeatureFlags().flags.supportsModelTypeTags`). - **Breaking**: none. Flag default false preserves the legacy bare-tag write/read exactly. ## Review Focus - The FE writes only the single selected folder_name's `model_type:` form; the backend keeps the bare directory-path twin in sync and forward-expands shared paths (spec-drift §1), so the edit path manages just the `model_type:` tags and the resolver read path is untouched. - An uncovered asset (no `model_type:` tag) still routes by its bare tag on both read and write, mirroring FE-1076. - Stacked on FE-1076 (Comfy-Org#12948, `synap5e/fe-1076-...`), which is self-contained on `main`. Chain: `main` → Comfy-Org#12948 (FE-1076) → Comfy-Org#12972 (this PR); the diff carries FE-1076's commits until it merges, then GitHub auto-retargets this PR to `main` (or this PR is merged to Comfy-Org#12948) Part of the cross-BE `model_type:<folder_name>` tag migration. Flipping the cutover dynconfig is a separate coordinated action, not in this PR. Treats FE-1069 (migrate model-upload import-by-URL off hard-coded `['models', <type>]` tags): the import-by-URL write here emits `models` + `model_type:<folder_name>` (flag-gated) per FE-1069's AC, with `user_metadata.model_type` kept bare.
Summary
Group the asset-mode Model Library by namespaced
model_type:*tags, gated on a backend-reported capability, and make the asset-mode model store run on non-cloud builds (FE-1076).Changes
supports_model_type_tagsserver feature flag (getServerFeature, viauseFeatureFlags).model_type:ignored. Unchanged for backends that don't declare the capability.model_type:*values; an uncovered asset (nomodel_type:tag) still routes by its bare tags. Card badges and the browser title key off the samemodel_type:values the asset groups under, so they cannot diverge from the grouping (a shared multi-type asset badges every value; a node-typed picker titles off the requested node's category). Node-provider lookup is tiered:model_type:values plus related bare twins (equal to or path-extending one, soLLM/Qwen-VLstill beats flatmodel_type:LLM) ordered deepest-first, then unrelated bare tags only as trailing fallbacks, so an incidentalfoo/barcan never pre-empt a resolvablemodel_type:vaebut still resolves when nothing authoritative does.modelTypeModearg on the pure helpers (getAssetCategories/getAssetTypeBadges/getAssetNodeCategoryCandidates/getPrimaryCategoryTag/filterByCategory); only call sites read the flag, and new call sites must choose a mode explicitly.if (isCloud)gate inassetsStore.getModelStateso the asset-mode Model Library populates on non-cloud distributions (it previously returned an empty no-op store off-cloud). Scoped to the model-asset store only.getAssetModelTypecan never return a rawmodel_type:literal (kills the prefixed round-trip through edit widgets).Comfy.Assets.UseAssetAPIsetting (default off on core, on for cloud)).Review Focus
getModelStateunwrap is the minimal scoped change — only the model-asset store; otherisCloudusages in the file (e.g. input-file fetch) are untouched.supports_model_type_tags— default-false keeps current behavior everywhere until then.model_type:LLM/model_type:ultralyticscreate no loader node (their loaders are registered hierarchically with no bare key); needs a subtype-carrying backend tag.model_type:leaking into labels, correct loader-node creation.https://pr-4394.testenvs.comfy.org/ is a live environment with new tags behaviour +
supports_model_type_tagsset