Repository navigation
test: flatten the FE-258 fixtures and trim the realign commentary - #18682
christian-byrne wants to merge 13 commits into
Conversation
The three ComfyNodeDefV1 literals nested up to 13 levels. Build them from dynamicCombo/autogrow/testNodeDef instead, which caps the region at 5.
40 lines of added commentary down to 12. Rationale that belongs in the PR description, and text restating an adjacent guard clause, are gone.
dynamicCombo took a record, so option order rode Object.entries and an integer-like key would silently displace the default option. Take the default option as its own parameter instead. Also stop aliasing output_name to the output array, and restore the two clauses the doc trim cut too deep: why a saved link lands on the wrong slot, and why unselected-option children are safe to realign.
The defaultOption parameter promised something Object.keys does not honour: an array-index key is hoisted ahead of insertion order, so '0' would displace it. The sibling addDynamicCombo helper keys its options '0'/'1'/'2', so the next fixture author would hit this. Reject array-index keys instead. Also restore the fixture display names and resize's 'resized' output label, which testNodeDef had collapsed.
The regex rejected any numeric key, but JS only hoists array indices below 2^32-1, and a numeric default option lands first anyway. Build the record production builds and compare its first key instead, which is exact in both directions and also catches duplicate keys.
…tion The guard had no test holding it in place, so a future cleanup could drop it as a redundant expression and silently change which option a fixture defaults to. Deleting it now fails a named test. Also restore 'before the node's group widget values are applied' to realignGroupWidgetChildLinks, dropped by accident in c751096 while restoring a different clause. The realign runs ahead of the widget value restoration in LGraphNode.configure; reordering them would reintroduce FE-258.
A tail reorder like ['0','2','1'] is exactly a key set that does not survive the round trip, but the head-only check accepted it, so the test pinning the guard claimed more than the guard delivered.
Nothing asserted that the definition's default option lays out width, so if that flipped the realign would become a no-op and the test would stay green while testing nothing. Also report only the actual duplicate keys, and name the guard test for both directions it asserts.
🎨 Storybook: ✅ Built — View Storybook🎭 Playwright: ✅ 2128 passed, 0 failed · 2 flaky📊 Browser Reports
🌾 Fallow: ✅ PassedFallow audits the diff for duplication, complexity and dead code. It gates on new findings only, so when the audit completes, a red is something this PR introduced or moved — never inherited backlog. A red can also mean the audit itself did not finish; the details below say which happened. No new findings in the changed files. Reproduce locally# the base the gate compared against
npx fallow audit --root . --quiet --format json \
--changed-since 4c04e82cb7e182e45b355bc268213632e8e94573
Full docs: ADR-DEVEX-LINT-0015 📦 Bundle: 10 MB gzip 🔴 +304 kBDetailsSummary
Category Glance App Entry Points — 3.77 kB (baseline 3.73 kB) • 🔴 +38 BMain entry bundles and manifests
Status: 1 added / 1 removed Graph Workspace — 1.45 MB (baseline 1.43 MB) • 🔴 +22.4 kBGraph editor runtime, canvas, workflow orchestration
Status: 2 added / 2 removed / 1 unchanged Views & Navigation — 134 kB (baseline 124 kB) • 🔴 +9.77 kBTop-level views, pages, and routed surfaces
Status: 17 added / 16 removed / 1 unchanged Panels & Settings — 639 kB (baseline 621 kB) • 🔴 +17.5 kBConfiguration panels, inspectors, and settings screens
Status: 28 added / 27 removed User & Accounts — 39.4 kB (baseline 27.8 kB) • 🔴 +11.6 kBAuthentication, profile, and account management bundles
Status: 9 added / 9 removed / 2 unchanged Editors & Dialogs — 126 kB (baseline 126 kB) • 🔴 +690 BModals, dialogs, drawers, and in-app editors
Status: 8 added / 8 removed UI Components — 125 kB (baseline 117 kB) • 🔴 +8.01 kBReusable component library chunks
Status: 16 added / 15 removed Data & Services — 3.98 MB (baseline 3.6 MB) • 🔴 +378 kBStores, services, APIs, and repositories
Status: 15 added / 16 removed Utilities & Hooks — 576 kB (baseline 559 kB) • 🔴 +17.6 kBHelpers, composables, and utility bundles
Status: 30 added / 30 removed / 6 unchanged Vendor & Third-Party — 17.4 MB (baseline 18.1 MB) • 🟢 -686 kBExternal libraries and shared vendor chunks
Status: 12 added / 12 removed / 6 unchanged Other — 17.9 MB (baseline 16.4 MB) • 🔴 +1.49 MBBundles that do not match a named category
Status: 179 added / 174 removed / 122 unchanged ⚡ Performance Report
Show regressions
All metrics
Historical variance (last 15 runs)
Trend (last 15 commits on main)
Raw data{
"timestamp": "2026-09-25T11:17:55.469Z",
"gitSha": "c92d6ee2aeb81bfe992da1454ad9f0ce3115af88",
"branch": "glary/fe-2875-flatten-dynamic-combo-fixtures",
"measurements": [
{
"name": "canvas-idle",
"durationMs": 2016.3049999999885,
"styleRecalcs": 10,
"styleRecalcDurationMs": 3.8659999999999997,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 496.56999999999994,
"heapDeltaBytes": -1131208,
"heapUsedBytes": 61284808,
"domNodes": 0,
"jsHeapTotalBytes": 5767168,
"scriptDurationMs": 11.83,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "canvas-idle",
"durationMs": 1997.7680000000646,
"styleRecalcs": 8,
"styleRecalcDurationMs": 3.2929999999999993,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 529.0600000000001,
"heapDeltaBytes": 4485868,
"heapUsedBytes": 66978264,
"domNodes": -228,
"jsHeapTotalBytes": 4972544,
"scriptDurationMs": 11.354999999999997,
"eventListeners": -142,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1843.5400000000186,
"styleRecalcs": 72,
"styleRecalcDurationMs": 19.381,
"layouts": 12,
"layoutDurationMs": 4.757,
"taskDurationMs": 905.2260000000001,
"heapDeltaBytes": -9146200,
"heapUsedBytes": 58344580,
"domNodes": -228,
"jsHeapTotalBytes": 5758976,
"scriptDurationMs": 166.981,
"eventListeners": -164,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1918.608000000063,
"styleRecalcs": 72,
"styleRecalcDurationMs": 18.837,
"layouts": 12,
"layoutDurationMs": 4.978,
"taskDurationMs": 889.4110000000001,
"heapDeltaBytes": 8808960,
"heapUsedBytes": 71223548,
"domNodes": -228,
"jsHeapTotalBytes": 6283264,
"scriptDurationMs": 165.675,
"eventListeners": -142,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1716.9319999999857,
"styleRecalcs": 31,
"styleRecalcDurationMs": 10.989,
"layouts": 6,
"layoutDurationMs": 1.346,
"taskDurationMs": 381.97400000000005,
"heapDeltaBytes": 2591668,
"heapUsedBytes": 70225548,
"domNodes": 53,
"jsHeapTotalBytes": 5505024,
"scriptDurationMs": 14.809000000000003,
"eventListeners": 15,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1723.0780000001005,
"styleRecalcs": 32,
"styleRecalcDurationMs": 11.161,
"layouts": 6,
"layoutDurationMs": 1.3259999999999998,
"taskDurationMs": 365.17199999999997,
"heapDeltaBytes": 2636808,
"heapUsedBytes": 70125468,
"domNodes": 53,
"jsHeapTotalBytes": 4980736,
"scriptDurationMs": 13.571000000000003,
"eventListeners": 15,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "dom-widget-clipping",
"durationMs": 700.4599999999641,
"styleRecalcs": 12,
"styleRecalcDurationMs": 4.433,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 447.32899999999995,
"heapDeltaBytes": 18476680,
"heapUsedBytes": 80833180,
"domNodes": 0,
"jsHeapTotalBytes": 5505024,
"scriptDurationMs": 179.078,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "dom-widget-clipping",
"durationMs": 689.3920000001117,
"styleRecalcs": 11,
"styleRecalcDurationMs": 3.333000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 441.32000000000005,
"heapDeltaBytes": 18502920,
"heapUsedBytes": 81060680,
"domNodes": 0,
"jsHeapTotalBytes": 5767168,
"scriptDurationMs": 161.41,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "dom-widget-order-dense-n200-w200-i120",
"durationMs": 1191.677000000027,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 893.7439999999999,
"heapDeltaBytes": 31692788,
"heapUsedBytes": 97109316,
"domNodes": -229,
"jsHeapTotalBytes": 20344832,
"scriptDurationMs": 67.34000000000002,
"eventListeners": -185,
"totalBlockingTimeMs": 568,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "dom-widget-order-dense-n200-w200-i120",
"durationMs": 1235.8849999999393,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 941.248,
"heapDeltaBytes": -13795312,
"heapUsedBytes": 72655276,
"domNodes": -259,
"jsHeapTotalBytes": 19791872,
"scriptDurationMs": 66.668,
"eventListeners": -185,
"totalBlockingTimeMs": 555,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "dom-widget-order-sparse-n1000-w20-i120",
"durationMs": 861.1789999999928,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 501.55599999999987,
"heapDeltaBytes": 15115360,
"heapUsedBytes": 90727812,
"domNodes": 0,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 150.953,
"eventListeners": 0,
"totalBlockingTimeMs": 230,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "dom-widget-order-sparse-n1000-w20-i120",
"durationMs": 844.694000000004,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 471.0409999999996,
"heapDeltaBytes": 14815584,
"heapUsedBytes": 90052544,
"domNodes": 0,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 144.24900000000002,
"eventListeners": 0,
"totalBlockingTimeMs": 224,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "extension-clean-combined-draw",
"durationMs": 9344.288000000006,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 8417.131,
"heapDeltaBytes": -1581016,
"heapUsedBytes": 81349640,
"domNodes": 0,
"jsHeapTotalBytes": 5242880,
"scriptDurationMs": 6972.395,
"eventListeners": 0,
"totalBlockingTimeMs": 561,
"frameDurationMs": 18.330000000000048,
"p95FrameDurationMs": 33.30000000000109
},
{
"name": "extension-clean-combined-draw",
"durationMs": 8896.55700000003,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 8146.1810000000005,
"heapDeltaBytes": 15372356,
"heapUsedBytes": 98403388,
"domNodes": 0,
"jsHeapTotalBytes": 4718592,
"scriptDurationMs": 6714.599999999999,
"eventListeners": 0,
"totalBlockingTimeMs": 344,
"frameDurationMs": 18.886666666666677,
"p95FrameDurationMs": 33.30000000000109
},
{
"name": "extension-rgthree-prototype-draw",
"durationMs": 9155.628000000035,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 8275.145999999999,
"heapDeltaBytes": -11680044,
"heapUsedBytes": 71178220,
"domNodes": 0,
"jsHeapTotalBytes": 4980736,
"scriptDurationMs": 6595.8769999999995,
"eventListeners": 0,
"totalBlockingTimeMs": 310,
"frameDurationMs": 17.776666666666703,
"p95FrameDurationMs": 33.29999999999927
},
{
"name": "extension-rgthree-prototype-draw",
"durationMs": 8765.168000000016,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 8035.8640000000005,
"heapDeltaBytes": -4473168,
"heapUsedBytes": 71501596,
"domNodes": 0,
"jsHeapTotalBytes": 4980736,
"scriptDurationMs": 6651.937,
"eventListeners": 0,
"totalBlockingTimeMs": 336,
"frameDurationMs": 17.779999999999987,
"p95FrameDurationMs": 33.29999999999927
},
{
"name": "large-graph-idle",
"durationMs": 2040.0869999999713,
"styleRecalcs": 8,
"styleRecalcDurationMs": 3.556,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 656.1599999999999,
"heapDeltaBytes": 2800656,
"heapUsedBytes": 83120668,
"domNodes": -228,
"jsHeapTotalBytes": 253952,
"scriptDurationMs": 38.739999999999995,
"eventListeners": -142,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "large-graph-idle",
"durationMs": 2029.0899999999965,
"styleRecalcs": 6,
"styleRecalcDurationMs": 2.5370000000000013,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 634.352,
"heapDeltaBytes": 2658404,
"heapUsedBytes": 83506632,
"domNodes": -228,
"jsHeapTotalBytes": -270336,
"scriptDurationMs": 35.74200000000001,
"eventListeners": -142,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-pan",
"durationMs": 2670.0230000000147,
"styleRecalcs": 64,
"styleRecalcDurationMs": 10.039,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1703.9780000000003,
"heapDeltaBytes": 23856936,
"heapUsedBytes": 88594320,
"domNodes": 0,
"jsHeapTotalBytes": 749568,
"scriptDurationMs": 866.479,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-pan",
"durationMs": 2736.1459999999624,
"styleRecalcs": 67,
"styleRecalcDurationMs": 10.705,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1704.1720000000003,
"heapDeltaBytes": 24141496,
"heapUsedBytes": 88834468,
"domNodes": 0,
"jsHeapTotalBytes": 749568,
"scriptDurationMs": 866.654,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66999999999998,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "large-graph-zoom",
"durationMs": 3820.3040000000215,
"styleRecalcs": 62,
"styleRecalcDurationMs": 12.146,
"layouts": 60,
"layoutDurationMs": 15.982,
"taskDurationMs": 1788.252,
"heapDeltaBytes": 7426268,
"heapUsedBytes": 73674380,
"domNodes": 0,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 822.455,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "large-graph-zoom",
"durationMs": 3815.277999999921,
"styleRecalcs": 63,
"styleRecalcDurationMs": 12.804,
"layouts": 60,
"layoutDurationMs": 15.928,
"taskDurationMs": 1789.456,
"heapDeltaBytes": 9420396,
"heapUsedBytes": 74709412,
"domNodes": 0,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 817.7260000000001,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "legacy-node-drag",
"durationMs": 3126.592999999957,
"styleRecalcs": 41,
"styleRecalcDurationMs": 6.026,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 2016.7440000000001,
"heapDeltaBytes": 2806284,
"heapUsedBytes": 68950120,
"domNodes": 0,
"jsHeapTotalBytes": 524288,
"scriptDurationMs": 1016.1320000000001,
"eventListeners": 180,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "legacy-node-drag",
"durationMs": 3184.174999999982,
"styleRecalcs": 44,
"styleRecalcDurationMs": 7.246000000000002,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 2085.096,
"heapDeltaBytes": 24086976,
"heapUsedBytes": 90104556,
"domNodes": 0,
"jsHeapTotalBytes": 2252800,
"scriptDurationMs": 1050.564,
"eventListeners": 180,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666636,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "minimap-idle",
"durationMs": 1996.6459999999984,
"styleRecalcs": 6,
"styleRecalcDurationMs": 2.0230000000000006,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 502.3810000000001,
"heapDeltaBytes": 19335808,
"heapUsedBytes": 83722676,
"domNodes": -228,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 39.301,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66999999999998,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "minimap-idle",
"durationMs": 2016.6940000000295,
"styleRecalcs": 6,
"styleRecalcDurationMs": 3.3159999999999994,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 584.365,
"heapDeltaBytes": -4073740,
"heapUsedBytes": 84256168,
"domNodes": -228,
"jsHeapTotalBytes": -1019904,
"scriptDurationMs": 35.090999999999994,
"eventListeners": -142,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 595.5859999999689,
"styleRecalcs": 48,
"styleRecalcDurationMs": 7.0870000000000015,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 391.31899999999996,
"heapDeltaBytes": 13451280,
"heapUsedBytes": 76972192,
"domNodes": 0,
"jsHeapTotalBytes": 4980736,
"scriptDurationMs": 156.90099999999998,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 618.8139999999294,
"styleRecalcs": 46,
"styleRecalcDurationMs": 6.156,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 406.20899999999995,
"heapDeltaBytes": 13946976,
"heapUsedBytes": 77332120,
"domNodes": 0,
"jsHeapTotalBytes": 5767168,
"scriptDurationMs": 154.42700000000002,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "subgraph-idle",
"durationMs": 2022.8950000000054,
"styleRecalcs": 10,
"styleRecalcDurationMs": 4.410999999999998,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 538.456,
"heapDeltaBytes": -4883328,
"heapUsedBytes": 58448628,
"domNodes": -244,
"jsHeapTotalBytes": 4972544,
"scriptDurationMs": 11.342999999999998,
"eventListeners": -210,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66999999999998,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "subgraph-idle",
"durationMs": 2012.6080000000002,
"styleRecalcs": 10,
"styleRecalcDurationMs": 3.8589999999999978,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 504.505,
"heapDeltaBytes": 13470204,
"heapUsedBytes": 76089072,
"domNodes": -244,
"jsHeapTotalBytes": 4972544,
"scriptDurationMs": 9.889999999999999,
"eventListeners": -188,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1722.5129999999922,
"styleRecalcs": 73,
"styleRecalcDurationMs": 19.73,
"layouts": 16,
"layoutDurationMs": 6.420999999999999,
"taskDurationMs": 820.3090000000001,
"heapDeltaBytes": 601028,
"heapUsedBytes": 70451424,
"domNodes": -244,
"jsHeapTotalBytes": 6283264,
"scriptDurationMs": 113.83399999999999,
"eventListeners": -188,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1768.9260000000786,
"styleRecalcs": 75,
"styleRecalcDurationMs": 21.138000000000005,
"layouts": 16,
"layoutDurationMs": 6.399000000000001,
"taskDurationMs": 876.607,
"heapDeltaBytes": 6105668,
"heapUsedBytes": 69850144,
"domNodes": -244,
"jsHeapTotalBytes": 5234688,
"scriptDurationMs": 130.327,
"eventListeners": -188,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "subgraph-transition-enter",
"durationMs": 1385.5859999999893,
"styleRecalcs": 18,
"styleRecalcDurationMs": 29.32,
"layouts": 5,
"layoutDurationMs": 12.87,
"taskDurationMs": 976.2469999999998,
"heapDeltaBytes": -6269628,
"heapUsedBytes": 81865544,
"domNodes": 13667,
"jsHeapTotalBytes": 11272192,
"scriptDurationMs": 29.841,
"eventListeners": 2383,
"totalBlockingTimeMs": 133,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "viewport-pan-sweep",
"durationMs": 9188.597000000016,
"styleRecalcs": 250,
"styleRecalcDurationMs": 45.196,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 6083.552,
"heapDeltaBytes": 824196,
"heapUsedBytes": 81329092,
"domNodes": -190,
"jsHeapTotalBytes": 1495040,
"scriptDurationMs": 2805.897,
"eventListeners": -164,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.80000000000109
},
{
"name": "viewport-pan-sweep",
"durationMs": 8627.31000000008,
"styleRecalcs": 250,
"styleRecalcDurationMs": 36.906,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 5845.316000000001,
"heapDeltaBytes": 192328,
"heapUsedBytes": 81568068,
"domNodes": -228,
"jsHeapTotalBytes": 1040384,
"scriptDurationMs": 2742.148,
"eventListeners": -142,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "vue-large-graph-idle",
"durationMs": 17656.966999999953,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 17579.161000000004,
"heapDeltaBytes": -61435224,
"heapUsedBytes": 179220896,
"domNodes": -8049,
"jsHeapTotalBytes": -18026496,
"scriptDurationMs": 289.182,
"eventListeners": -14923,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.776666666666642,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-large-graph-idle",
"durationMs": 17753.510000000006,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 17641.135,
"heapDeltaBytes": -25978272,
"heapUsedBytes": 195538356,
"domNodes": -8049,
"jsHeapTotalBytes": -14950400,
"scriptDurationMs": 284.89200000000005,
"eventListeners": -14923,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.776666666666642,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "vue-large-graph-pan",
"durationMs": 22202.237999999965,
"styleRecalcs": 159,
"styleRecalcDurationMs": 23.809000000000026,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 22104.296,
"heapDeltaBytes": -60981648,
"heapUsedBytes": 180239784,
"domNodes": -8049,
"jsHeapTotalBytes": -15011840,
"scriptDurationMs": 920.335,
"eventListeners": -14923,
"totalBlockingTimeMs": 22,
"frameDurationMs": 19.44333333333319,
"p95FrameDurationMs": 16.700000000004366
},
{
"name": "vue-large-graph-pan",
"durationMs": 21742.835000000014,
"styleRecalcs": 151,
"styleRecalcDurationMs": 19.75200000000002,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 21643.286,
"heapDeltaBytes": -72572220,
"heapUsedBytes": 168139316,
"domNodes": -8049,
"jsHeapTotalBytes": -21696512,
"scriptDurationMs": 884.8290000000001,
"eventListeners": -14923,
"totalBlockingTimeMs": 107,
"frameDurationMs": 17.776666666666763,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-node-resize-workload",
"durationMs": 6564.074000000005,
"styleRecalcs": 20,
"styleRecalcDurationMs": 73.657,
"layouts": 20,
"layoutDurationMs": 26.788999999999994,
"taskDurationMs": 6548.989,
"heapDeltaBytes": -35860568,
"heapUsedBytes": 183439424,
"domNodes": -8049,
"jsHeapTotalBytes": 16244736,
"scriptDurationMs": 119.384,
"eventListeners": -14923,
"totalBlockingTimeMs": 523,
"frameDurationMs": 17.77333333333336,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-node-resize-workload",
"durationMs": 6425.824000000034,
"styleRecalcs": 20,
"styleRecalcDurationMs": 75.821,
"layouts": 20,
"layoutDurationMs": 25.756,
"taskDurationMs": 6390.649,
"heapDeltaBytes": -71287604,
"heapUsedBytes": 165699260,
"domNodes": -8049,
"jsHeapTotalBytes": 11878400,
"scriptDurationMs": 123.86200000000001,
"eventListeners": -14923,
"totalBlockingTimeMs": 470,
"frameDurationMs": 17.776666666666763,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-zoom-out-idle",
"durationMs": 8755.106999999953,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 8709.836,
"heapDeltaBytes": 14998744,
"heapUsedBytes": 189489736,
"domNodes": 0,
"jsHeapTotalBytes": -16777216,
"scriptDurationMs": 166.96099999999993,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.776666666666642,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-zoom-out-idle",
"durationMs": 8867.543999999953,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 8782.019999999999,
"heapDeltaBytes": 29700724,
"heapUsedBytes": 189425820,
"domNodes": 0,
"jsHeapTotalBytes": -13631488,
"scriptDurationMs": 167.83100000000007,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.773333333333238,
"p95FrameDurationMs": 16.80000000000291
},
{
"name": "workflow-execution",
"durationMs": 486.32099999997536,
"styleRecalcs": 18,
"styleRecalcDurationMs": 20.545,
"layouts": 4,
"layoutDurationMs": 2.2009999999999996,
"taskDurationMs": 134.93600000000004,
"heapDeltaBytes": 5134448,
"heapUsedBytes": 67198672,
"domNodes": 129,
"jsHeapTotalBytes": 262144,
"scriptDurationMs": 12.783000000000001,
"eventListeners": 81,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "workflow-execution",
"durationMs": 97.2229999999854,
"styleRecalcs": 8,
"styleRecalcDurationMs": 11.618,
"layouts": 3,
"layoutDurationMs": 1.0479999999999998,
"taskDurationMs": 73.548,
"heapDeltaBytes": 3074572,
"heapUsedBytes": 65165316,
"domNodes": 94,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 7.056999999999999,
"eventListeners": 33,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
}
]
}🔬 E2E Coverage
Top 10 files by uncovered lines
🧩 Custom Nodes Ecosystem Matrix: ✅ PassedExecutes the frontend JS of every registry pack that ships any, against this change, in both renderers. The eight
On a pin-bump PR, a red gate means the ecosystem moved — not that your diff broke something. That is what the bump PR is for. Outside a pin bump or a cold rebuild of the unpinned tail, a red means the diff. Note the corpus size and the executed count are different numbers, and the verdict is against the executed count: roughly 3,100 of ~5,100 packs ship no extension-shaped JS and are skipped. Reproduce locally# build one shard's fixture, then run it
python3 scripts/registry-census/build_matrix.py --shard 1/4
MATRIX_RENDERER=vue MATRIX_VUE=1 \
npx vitest run --config vitest.matrix.config.mts
Full docs: scripts/registry-census/README.md |
|
Answering @christian-byrne's two tracking comments on #17486 here, since that PR is merged and its review threads no longer accept replies. Both findings were still true at #17486 squash-merged as
This PR is those commits cherry-picked onto current On FE-2875. @AustinMroz — this settles the tuple-vs-record question I left open on your thread, with evidence rather than preference. Options are an ordered tuple list because option order decides which option is the default: That turned up something worse than the nesting: On FE-2876. @DrJKL — the trim keeps only what the code cannot state: autogrow renumbering its own slots mid-configure, the realign running before widget values are applied, and why a saved link lands on the wrong slot at all. I did not restore the Zero executable production lines change: compiling both non-test files at |
|
Navigate logical layers of code changes, visualize relationships, and explore their blast radius. No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository: Comfy-Org/ComfyUI_frontend/.coderabbit.yaml Review profile: ASSERTIVE Plan: Team Run ID: 📒 Files selected for processing (1)
💤 Files with no reviewable changes (1)
Included review availability: 5 reviews are currently available. Your included PR review attempts over the past 7 days set your current allowance at 8 reviews per hour. 📝 WalkthroughWalkthroughDynamic-widget tests add shared fixture builders and a test for the default resize child. Comments clarify how group growth affects slots and how link realignment handles child inputs. ChangesDynamic widget behavior
Priority: ⬇️ Low Estimated code review effort: 2 (Simple) | ~10 minutes Change: Other Suggested reviewers: Merge Risk: ⚪ Minimal · up to The changes are ready to merge after normal checks; no actionable behavior or deployment risk remains. 🚥 Pre-merge checks | ✅ 7✅ Passed checks (7 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
|
@coderabbitai — no changes made, because there's nothing to act on yet: that comment is still the "review in progress" placeholder ( Declining both Finishing Touches opt-ins, since each cuts against what this PR is for:
Worth noting your fortune picked the exact principle this PR applies, twice:
A Branch is unchanged at |
|
Thanks @coderabbitai — no changes made, since the completed review reports "No actionable comments were generated" with 7/7 pre-merge checks passing, including Adr Compliance For Entity/Litegraph Changes. Branch stays at One correction to the generated walkthrough, since it's what a human skimming this PR will read first. It says:
That reads as though commentary was added. It was removed — that's half the point of the PR. Also worth noting for anyone reading the walkthrough's scope summary: the two non-test files in this diff contain zero executable changes. Compiling them at Still declining both Finishing Touches opt-ins, for the reasons given earlier: auto-generated docstrings would reintroduce exactly what FE-2876 asked to remove, and the two tests here are mutation-checked — each confirmed to fail against the specific regression it guards — which a generated test passing against current code would not be. |
Codecov Report✅ All modified and coverable lines are covered by tests. @@ Coverage Diff @@
## main #18682 +/- ##
==========================================
+ Coverage 83.74% 84.89% +1.15%
==========================================
Files 3020 2304 -716
Lines 149137 132439 -16698
Branches 38534 34275 -4259
==========================================
- Hits 124890 112438 -12452
+ Misses 22943 19702 -3241
+ Partials 1304 299 -1005
Flags with carried forward coverage won't be shown. Click here to find out more.
... and 886 files with indirect coverage changes 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
Verdict: ready to approve with nits
I found no production behavior regression. The comment-only production changes emit byte-identical JavaScript, the three flattened fixtures serialize identically to the base versions, and the focused mutation checks fail for the intended reasons.
Checked and cleared
- I kept the dynamic-combo ordering guard. Production performs the same
Object.fromEntriesthenObject.keysconversion, so numeric-like keys can change the default and duplicate keys can collapse. - The PR diff contains the stated three files. A two-dot history comparison that showed 383 files included unrelated base history.
- The exact stated test scope passes 78 tests in 6 files.
pnpm lintexits 0 with the stated 195 existing warnings, andpnpm format:checkexits 0.- The fixture-byte comparison and all three named mutation checks reproduced at
72cb115bb38e34df85027356c71b98fab9a6ce48.
Coverage
Standard-depth correctness, maintainability, idiomaticity, tests, and documentation lanes reviewed all three changed files. The tests lane applied unit-test design, mock-boundary, fixture-setup, and proof lenses. Playwright guidance was not applicable because no browser_tests/ file changed. Fallow found no new duplication; its other diagnostics were existing repository findings. Focused tests passed 78/78, typecheck passed, lint passed with 195 warnings, and format checking passed.
| } | ||
| } | ||
|
|
||
| type ChildInputs = Record<string, InputSpec> |
There was a problem hiding this comment.
P2: These helpers claim the broad InputSpec type, but that type accepts custom input options through a passthrough schema. It therefore does not check the COMFY_DYNAMICCOMBO_V3 or COMFY_AUTOGROW_V3 shapes that the helpers hand-declare. Please derive the helper types from the exported zDynamicComboInputSpec, zAutogrowOptions, and zComfyInputsSpec schemas, or export their inferred aliases, so schema drift fails typecheck instead of reaching the runtime parser.
There was a problem hiding this comment.
Fixed in 587187725a. You were right, and it was worse than "not checked": on origin/main the nested combo payload wasn't type-checked at all. zBaseInputOptions is .passthrough(), so ['COMFY_DYNAMICCOMBO_V3', { options: [...] }] matched zCustomInputSpec's [string, BaseInputOptions] and the whole options array landed in the index signature as unknown — inputs: { require: {...} } would have compiled clean.
The helpers now derive from the schemas:
type ChildInputs = NonNullable<ComfyInputsSpec['required']>
type DynamicComboInputSpec = z.infer<typeof zDynamicComboInputSpec>
type AutogrowInputSpec = [
type: 'COMFY_AUTOGROW_V3',
options: z.infer<typeof zAutogrowOptions>
]dynamicCombo returns DynamicComboInputSpec, autogrow returns AutogrowInputSpec, and ChildInputs comes off zComfyInputsSpec rather than being spelled out.
I confirmed drift now fails typecheck rather than reaching the runtime parser, by mutating the builders and running pnpm typecheck:app:
dynamicWidgets.test.ts(709,7): error TS2322: Type '{ keyy: string; ... }[]' is not assignable to
type '{ inputs: {...}; key: string; }[]'
dynamicWidgets.test.ts(763,19): error TS2561: Object literal may only specify known properties,
but 'inputt' does not exist in type '{ input: {...}; ... }'. Did you mean to write 'input'?
Both reverted; pnpm typecheck is clean on HEAD.
Two notes on the ceiling of this, so the phrase "derived from the schemas" isn't read as stronger than it is. First, there's no zAutogrowInputSpec tuple to derive from — only zAutogrowOptions — so AutogrowInputSpec still hand-declares the 'COMFY_AUTOGROW_V3' tag, with the options half derived. Second, storing the result into ChildInputs (= Record<string, InputSpec>) erases the precise tuple back to the union, so the tightening binds the builders' own construction and parameters, not the value end-to-end. That residual is inherent to the passthrough schema, not something this PR can close.
| } | ||
|
|
||
| describe('dynamicCombo fixture builder', () => { | ||
| test('rejects option keys that a record reorders, and accepts those it does not', () => { |
There was a problem hiding this comment.
P3 Nit: This combines two reordered-key cases, duplicate rejection, and an accepted boundary in one test. Repository test guidance calls for named table rows when cases share one Arrange-Act-Assert shape and one behavior per test. Please table the rejected cases and give the accepted boundary its own named test so a failure identifies the broken contract directly.
There was a problem hiding this comment.
Done in 587187725a. The three rejected cases are now named table rows and the accepted boundary is its own test:
const rejectedCases: {
name: string
options: [ComboOption, ...ComboOption[]]
message: RegExp
}[] = [
{
name: 'a numeric key after a named one',
options: [option('Seedance'), option('0')],
message: /\[0, Seedance\] instead of \[Seedance, 0\]/
},
{
name: 'numeric keys out of ascending order',
options: [option('0'), option('2'), option('1')],
message: /\[0, 1, 2\] instead of \[0, 2, 1\]/
},
{
name: 'a key declared twice',
options: [option('a'), option('b'), option('a')],
message: /Duplicate option keys: a/
}
]
test.for(rejectedCases)('rejects $name', ({ options, message }) => {
expect(() => dynamicCombo(...options)).toThrow(message)
})
test('accepts a numeric key too large to be an array index', () => { ... })A failure now names the broken contract directly:
✓ dynamicCombo fixture builder > rejects 'a numeric key after a named one'
✓ dynamicCombo fixture builder > rejects 'numeric keys out of ascending order'
✓ dynamicCombo fixture builder > rejects 'a key declared twice'
✓ dynamicCombo fixture builder > accepts a numeric key too large to be an array index
Two incidental changes while restructuring. The row type is [ComboOption, ...ComboOption[]] rather than ComboOption[] so the spread into dynamicCombo(defaultOption, ...rest) typechecks. And I tightened the duplicate row's matcher from /Duplicate option keys/ to /Duplicate option keys: a/ — once the row is named for the duplicate, the assertion may as well pin which key is reported. I also trimmed the row names to fit under Vitest's title-truncation threshold, since $name was rendering as rejects 'a numeric key declared after a named …'.
I kept the boundary case rather than dropping it: '4294967295' is 2³²−1, one past the last valid array index, so it's the first numeric-looking key a record leaves in declared order — the exact boundary on the other side of the two rejected reorderings.
|
|
||
| describe('dynamicCombo fixture builder', () => { | ||
| test('rejects option keys that a record reorders, and accepts those it does not', () => { | ||
| const option = (key: string): ComboOption => [key, {}] |
There was a problem hiding this comment.
P3 Nit: docs/guidance/engineering.md prefers a function declaration whenever one works. This helper needs neither lexical this nor expression identity, so please use function option(key: string): ComboOption.
There was a problem hiding this comment.
Agreed and fixed in 587187725a:
function option(key: string): ComboOption {
return [key, {}]
}It also had to move out of the test body to be usable from the table rows for the sibling comment, which the declaration form handles cleanly via hoisting inside the describe callback.
| * (dynamic combos) name their child inputs `<group widget name>.<key>` and | ||
| * replace the whole set whenever their own value changes. | ||
| */ | ||
| /** Group widgets (dynamic combos) name children `<group widget name>.<key>`. */ |
There was a problem hiding this comment.
P3 Nit: This comment paraphrases groupNameOf() and the widget-name equality check immediately below it. It adds no timing, ownership, or rationale that the code cannot state, so I would delete it under the repository's redundant-comment rule.
There was a problem hiding this comment.
Agreed and deleted in 587187725a. groupNameOf() plus the widget.name === groupName comparison state it between them, and the "replaces the whole set on value change" half is already restated in realignGroupWidgetChildLinks's docblock below.
Reviewing the neighbouring trims against your rule, though, I found I'd over-trimmed the isAutogrowGroupInput docblock in an earlier commit on this branch and restored two things there (38e9137e17, a450301e83, 08c2c82094):
- The rejected alternative and its reason.
groupNameOfis a barelastIndexOf('.'), so an ordinary input namedfoo.baryields groupfoo. Both filters survive that only because they cross-check something else —node.widgetshere, the registry there. The trimmed text kept "a group is not a widget", which rules out the widget-name approach but says nothing against a plain dotted-name test, which is the simplification a reader would reach for next and which would returntruefor every dotted input. That's a rejected alternative, not a paraphrase of the code. - The pointer to the proving test.
grep -rn subgraphConvertAutogrowInputs src browser_tests docsreturned nothing after the trim, so nothing connected the filter to the E2E test that goes red without it. It's back as a bare@see. I'd initially written it as "the only test that fails when autogrow links are realigned early" and removed that claim — there are ten-plus autogrow-touching browser specs and I hadn't run them, so the exhaustiveness wasn't mine to assert.
Net on that block is still shorter than origin/main and now reads as two sentences instead of one run-on.
Verification across all four comments: pnpm typecheck clean, pnpm lint 0 errors (195 pre-existing warnings, unchanged), pnpm format:check clean, and the four affected suites pass 65/65. No Playwright or tmux verification — the production diff is comment-only (0 non-comment lines changed in either .ts file, so the emitted JS is unchanged), and the rest is test-file-internal, so there's no runtime surface to exercise. The real-usage check that does apply is the new lays out the default option, which the saved workflow does not select test, which drives a builder-produced fixture through registerNodeDef → LiteGraph.createNode and asserts the laid-out children.
DrJKL
left a comment
There was a problem hiding this comment.
Verdict: ready to approve with one documentation nit
All four prior findings are fixed at 08c2c82094ea941d5c9343092cf16741e48f3a68:
- The fixture helpers now derive their child, dynamic-combo, and autogrow shapes from the node-definition schemas.
- The rejected option-key cases are named table rows, with the accepted boundary in its own test.
optionis now a function declaration.- The redundant group-widget naming docblock is gone.
I found no behavioral regression in the follow-up. The focused suites pass 70/70, typecheck passes, scoped lint has only the three existing layer warnings, and formatting passes. The one new inline item is non-blocking.
| * @see browser_tests/tests/subgraph/subgraphConvertAutogrowInputs.spec.ts | ||
| * "loads with both reference images connected" |
There was a problem hiding this comment.
P3 Nit: Please drop this @see. The root guidance specifically calls out a guard comment that names the test file guarding it as redundant. The preceding prose already records the rejected dotted-name alternative and the runtime reason for this filter, while the test remains discoverable by the behavior and symbol names.
There was a problem hiding this comment.
Agreed and dropped in 685c323a60. AGENTS.md:252 — "A guard clause that mirrors another file does not need a comment naming that file" — covers this exactly, and I checked your discoverability point holds: find browser_tests -iname "*autogrow*" surfaces the spec, and it's the only autogrow file under browser_tests/tests/subgraph/. The prose keeps the durable half (the rejected dotted-name alternative, and which destroys them as the runtime reason), while the pointer was the rot-prone half — I'd already had to trim an unverifiable "the only test that fails" claim off it once.
For the record on how it got there: the @see was me over-correcting. I trimmed the original parenthetical early in the branch, then restored it as an @see when a review flagged that nothing linked the filter to its proving test. Your rule is the better tiebreak — it's gone.
Gates on 685c323a60: pnpm typecheck clean, pnpm lint 0 errors (195 pre-existing warnings, unchanged), pnpm format:check clean, 65/65 across the four affected suites. Production diff remains comment-only.
One observation I'd rather raise than act on unilaterally, since it's a consequence of your earlier comment (#4102556851) and reversing you on a P3 isn't my call:
Deleting the isGroupWidgetChildInput docblock removed the last definition of "group widget" in the tree. grep -rn "group widget\|dynamic combo" src/lib/litegraph/src/linkDeduplication.ts now returns only two hits, both uses in realignGroupWidgetChildLinks's docblock; repo-wide the remaining hits are uses too. The naming convention itself is recoverable exactly as you said — groupNameOf plus the widget.name === groupName comparison state it between them. What isn't recoverable from this layer is the identification group widget ≡ dynamic combo, because src/lib/litegraph/ can't reference src/core/graph/widgets/dynamicWidgets.ts (the import-x/no-restricted-paths warnings on this file are that boundary), and the term is baked into the exported name realignGroupWidgetChildLinks.
So it may be worth a bare /** Group widgets are dynamic combos. */ — the two-word gloss, not the paraphrase you correctly cut. Happy to add it or leave it as is; say the word and I'll push either way.
DrJKL
left a comment
There was a problem hiding this comment.
Verdict: ready to approve
The remaining documentation nit is fixed at 685c323a60c351f6d01477ef0485ec7de14f9b1f. The test-file pointer is gone, and this commit changes no executable code.
I would leave the deleted group-widget gloss out as well. The exported function name supplies the domain term, while groupNameOf() and the widget-name lookup show the relationship needed to understand this code. Restoring a separate "group widgets are dynamic combos" comment would start another prose dependency without changing how a reader follows the function.
I found no new findings. git diff --check passes for the follow-up commit.
|
Thanks — leaving the gloss out, and your reasoning convinced me rather than just settling it. No further changes. Branch is at |
PR Created by the Glary-Bot Agent
Addresses the two review objections from #17486 that were still open when it squash-merged, so neither is fixed on
main.dynamicWidgets.test.tslinkDeduplication.tsNo executable production line changes. Compiling both non-test files at
mainand at this branch withremoveComments: trueemits byte-identical output — every non-test change here is a comment.FE-2875: fixture nesting 13 levels deep, now 5
Three builders (
dynamicCombo,autogrow,testNodeDef) absorb theoptions[].inputs.requiredandtemplate.input.requiredenvelopes, leaving only the values the tests depend on at the call site:The three fixtures are byte-identical to
mainafter construction, property insertion order included — which matters, since input key order is slot order and the FE-258 scenarios are slot-index scenarios.Options are an ordered tuple list rather than a record, because option order decides which option is the default.
dynamicComboWidgetrealizes the list throughObject.fromEntries→Object.keys, so an array-index-like key is hoisted ahead of insertion order — and the sibling helper__fixtures__/dynamicInputHelpers.tskeys its options'0'/'1'/'2', exactly that shape.dynamicCombotherefore replays the same realization and rejects any key set the record would reorder, plus duplicate keys that would silently merge.FE-2876: 52 comment lines, now 27
Removed the rationale that belongs in a PR description, and the prose restating guard clauses sitting two lines below it. Kept what the code cannot state: that autogrow renumbers its own slots mid-configure, that the realign runs before widget values are applied, and why a saved link arrives on the wrong slot at all.
I did not restore the pointer to
subgraphConvertAutogrowInputs.spec.ts.AGENTS.mdline 252 says a guard clause does not need a comment naming the file that guards it, and I confirmed the guard is not relying on that breadcrumb: deletingisAutogrowGroupInputmovestarget_slotfrom 1 to 3 and failsleaves an autogrow child link where the group put itin the same unit file.Two tests added, both mutation-checked
mainasserted nothing about which option the resize fixture defaults to. If that flipped, the definition would lay outmultiplieritself, the realign would no-op, and the FE-258 test would stay green while testing nothing.lays out the default option, which the saved workflow does not selectpins the premise; reversing the two options in the fixture fails it.The order guard is pinned the same way — deleting it fails
rejects option keys that a record reorders, and accepts those it does not, so a future "simplify this redundant expression" pass cannot quietly drop it.Verification
pnpm typecheckclean,pnpm lint0 errors (195 pre-existing warnings, none in changed lines),pnpm format:checkclean, 78 tests pass acrosssrc/core/graph/widgets/,LGraph.inputSlotRealign.test.ts, andlinkDeduplication.conflictingOrigins.test.ts.No screenshots: this branch changes no executable production code, so there is no rendered surface to capture. Mutation testing stood in for manual QA — each new test was confirmed to fail against the specific regression it guards.