Repository navigation
test: cover remaining ECS widget persistence gaps - #17505
christian-byrne wants to merge 6 commits into
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:
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Team Run ID: 📒 Files selected for processing (2)
Included review availability: Your plan provides up to 10 included reviews per hour; 6 remain after this review. 📝 WalkthroughWalkthroughThe pull request adds browser coverage for promoted text persistence in LiteGraph and Vue Nodes modes. It adds legacy widget persistence tests for scheduler and multiline text values, plus helpers for saved userdata access and workflow draft cleanup. ChangesWidget persistence coverage
Priority: ⬇️ Low Estimated code review effort: 3 (Moderate) | ~20 minutes Change: Other Suggested reviewers: Merge Risk: ⚪ Minimal · up to The changes add persistence coverage without an identified merge-blocking defect. 🚥 Pre-merge checks | ✅ 7✅ Passed checks (7 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
🎭 Playwright: ✅ 2033 passed, 0 failed · 3 flaky📊 Browser Reports
📦 Bundle: 10 MB gzip ⚪ 0 BDetailsSummary
Category Glance App Entry Points — 3.73 kB (baseline 3.73 kB) • ⚪ 0 BMain entry bundles and manifests Status: 1 unchanged Graph Workspace — 1.44 MB (baseline 1.44 MB) • ⚪ 0 BGraph editor runtime, canvas, workflow orchestration Status: 3 unchanged Views & Navigation — 132 kB (baseline 132 kB) • ⚪ 0 BTop-level views, pages, and routed surfaces Status: 18 unchanged Panels & Settings — 623 kB (baseline 623 kB) • ⚪ 0 BConfiguration panels, inspectors, and settings screens Status: 27 unchanged User & Accounts — 32.3 kB (baseline 32.3 kB) • ⚪ 0 BAuthentication, profile, and account management bundles Status: 11 unchanged Editors & Dialogs — 126 kB (baseline 126 kB) • ⚪ 0 BModals, dialogs, drawers, and in-app editors Status: 8 unchanged UI Components — 124 kB (baseline 124 kB) • ⚪ 0 BReusable component library chunks Status: 16 unchanged Data & Services — 3.81 MB (baseline 3.81 MB) • ⚪ 0 BStores, services, APIs, and repositories Status: 16 unchanged Utilities & Hooks — 561 kB (baseline 561 kB) • ⚪ 0 BHelpers, composables, and utility bundles Status: 37 unchanged Vendor & Third-Party — 18.2 MB (baseline 18.2 MB) • ⚪ 0 BExternal libraries and shared vendor chunks Status: 18 unchanged Other — 17.5 MB (baseline 17.5 MB) • ⚪ 0 BBundles that do not match a named category Status: 297 unchanged ⚡ Performance Report
Show regressions
All metrics
Historical variance (last 15 runs)
Trend (last 15 commits on main)
Raw data{
"timestamp": "2026-09-17T03:16:24.724Z",
"gitSha": "cf609ab4088ee05221e1258f51b653e3e274a131",
"branch": "test/ecs-qa-widget-gaps",
"measurements": [
{
"name": "canvas-idle",
"durationMs": 2010.4970000000435,
"styleRecalcs": 10,
"styleRecalcDurationMs": 5.401000000000002,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 264.481,
"heapDeltaBytes": 25330552,
"heapUsedBytes": 87522916,
"domNodes": 20,
"jsHeapTotalBytes": 5505024,
"scriptDurationMs": 6.573999999999998,
"eventListeners": 4,
"totalBlockingTimeMs": 0,
"frameDurationMs": 24.440000000000023,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "canvas-idle",
"durationMs": 1990.3969999999163,
"styleRecalcs": 10,
"styleRecalcDurationMs": 6.42,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 318.92900000000003,
"heapDeltaBytes": 9666152,
"heapUsedBytes": 72155672,
"domNodes": -259,
"jsHeapTotalBytes": 4710400,
"scriptDurationMs": 8.256999999999998,
"eventListeners": -152,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1811.1260000000016,
"styleRecalcs": 72,
"styleRecalcDurationMs": 24.475,
"layouts": 14,
"layoutDurationMs": 4.1530000000000005,
"taskDurationMs": 638.295,
"heapDeltaBytes": 9511832,
"heapUsedBytes": 71837416,
"domNodes": -257,
"jsHeapTotalBytes": 5234688,
"scriptDurationMs": 97.61600000000001,
"eventListeners": -154,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1729.0490000000318,
"styleRecalcs": 74,
"styleRecalcDurationMs": 25.403999999999996,
"layouts": 14,
"layoutDurationMs": 4.259,
"taskDurationMs": 593.583,
"heapDeltaBytes": -2499924,
"heapUsedBytes": 59935952,
"domNodes": -257,
"jsHeapTotalBytes": 6283264,
"scriptDurationMs": 112.571,
"eventListeners": -184,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1707.0799999999622,
"styleRecalcs": 31,
"styleRecalcDurationMs": 10.129,
"layouts": 6,
"layoutDurationMs": 1.0230000000000001,
"taskDurationMs": 192.726,
"heapDeltaBytes": 2170100,
"heapUsedBytes": 64686156,
"domNodes": 76,
"jsHeapTotalBytes": 4718592,
"scriptDurationMs": 6.9769999999999985,
"eventListeners": 19,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1708.6060000000316,
"styleRecalcs": 30,
"styleRecalcDurationMs": 13.784,
"layouts": 6,
"layoutDurationMs": 1.141,
"taskDurationMs": 248.08999999999997,
"heapDeltaBytes": 2077932,
"heapUsedBytes": 64621420,
"domNodes": 76,
"jsHeapTotalBytes": 5242880,
"scriptDurationMs": 11.649000000000001,
"eventListeners": 19,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "dom-widget-clipping",
"durationMs": 415.017999999975,
"styleRecalcs": 11,
"styleRecalcDurationMs": 5.703999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 273.53499999999997,
"heapDeltaBytes": 17555404,
"heapUsedBytes": 80303020,
"domNodes": 17,
"jsHeapTotalBytes": 5767168,
"scriptDurationMs": 99.479,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "dom-widget-clipping",
"durationMs": 449.4329999999991,
"styleRecalcs": 13,
"styleRecalcDurationMs": 7.247,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 301.97200000000004,
"heapDeltaBytes": 17512312,
"heapUsedBytes": 79752352,
"domNodes": 21,
"jsHeapTotalBytes": 5242880,
"scriptDurationMs": 106.128,
"eventListeners": 2,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.663333333333338,
"p95FrameDurationMs": 16.700000000000273
},
{
"name": "large-graph-idle",
"durationMs": 2007.5199999999995,
"styleRecalcs": 7,
"styleRecalcDurationMs": 3.7839999999999994,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 333.97800000000007,
"heapDeltaBytes": 4445048,
"heapUsedBytes": 83008536,
"domNodes": -242,
"jsHeapTotalBytes": -1056768,
"scriptDurationMs": 23.433999999999997,
"eventListeners": -150,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-idle",
"durationMs": 2008.6949999999888,
"styleRecalcs": 8,
"styleRecalcDurationMs": 6.714999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 470.792,
"heapDeltaBytes": -564108,
"heapUsedBytes": 78981792,
"domNodes": -247,
"jsHeapTotalBytes": -8192,
"scriptDurationMs": 29.002,
"eventListeners": -182,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "large-graph-pan",
"durationMs": 2114.4709999999804,
"styleRecalcs": 69,
"styleRecalcDurationMs": 13.858,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1041.384,
"heapDeltaBytes": -4388440,
"heapUsedBytes": 75482336,
"domNodes": -241,
"jsHeapTotalBytes": 1003520,
"scriptDurationMs": 528.702,
"eventListeners": -150,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-pan",
"durationMs": 2177.338999999961,
"styleRecalcs": 69,
"styleRecalcDurationMs": 16.023000000000003,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1190.1730000000002,
"heapDeltaBytes": -13241644,
"heapUsedBytes": 67333920,
"domNodes": -251,
"jsHeapTotalBytes": 516096,
"scriptDurationMs": 595.24,
"eventListeners": -182,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "large-graph-zoom",
"durationMs": 3169.416999999953,
"styleRecalcs": 66,
"styleRecalcDurationMs": 16.056,
"layouts": 60,
"layoutDurationMs": 14.719999999999999,
"taskDurationMs": 1234.912,
"heapDeltaBytes": -12360616,
"heapUsedBytes": 69318068,
"domNodes": -248,
"jsHeapTotalBytes": -794624,
"scriptDurationMs": 532.51,
"eventListeners": -178,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "large-graph-zoom",
"durationMs": 3087.0590000000675,
"styleRecalcs": 63,
"styleRecalcDurationMs": 12.402999999999997,
"layouts": 60,
"layoutDurationMs": 13.719999999999999,
"taskDurationMs": 1127.107,
"heapDeltaBytes": 3886652,
"heapUsedBytes": 70473428,
"domNodes": 8,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 518.99,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "legacy-node-drag",
"durationMs": 2282.950000000028,
"styleRecalcs": 46,
"styleRecalcDurationMs": 8.982999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1286.664,
"heapDeltaBytes": -17025616,
"heapUsedBytes": 66277172,
"domNodes": -223,
"jsHeapTotalBytes": 2088960,
"scriptDurationMs": 642.9599999999999,
"eventListeners": -2,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "legacy-node-drag",
"durationMs": 2360.584000000017,
"styleRecalcs": 45,
"styleRecalcDurationMs": 9.039000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1305.733,
"heapDeltaBytes": 5121020,
"heapUsedBytes": 71495216,
"domNodes": 10,
"jsHeapTotalBytes": 1310720,
"scriptDurationMs": 651.463,
"eventListeners": 186,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "minimap-idle",
"durationMs": 2008.2780000000184,
"styleRecalcs": 7,
"styleRecalcDurationMs": 3.9660000000000006,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 322.28599999999994,
"heapDeltaBytes": -4320528,
"heapUsedBytes": 84037264,
"domNodes": -243,
"jsHeapTotalBytes": -1544192,
"scriptDurationMs": 23.416000000000007,
"eventListeners": -152,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "minimap-idle",
"durationMs": 2020.0320000000147,
"styleRecalcs": 8,
"styleRecalcDurationMs": 6.502000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 414.315,
"heapDeltaBytes": -10857496,
"heapUsedBytes": 76332384,
"domNodes": -259,
"jsHeapTotalBytes": -233472,
"scriptDurationMs": 28.833,
"eventListeners": -182,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 377.3540000000253,
"styleRecalcs": 48,
"styleRecalcDurationMs": 10.129000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 260.92,
"heapDeltaBytes": 13625944,
"heapUsedBytes": 77172744,
"domNodes": 22,
"jsHeapTotalBytes": 5505024,
"scriptDurationMs": 91.623,
"eventListeners": 8,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.669999999999998,
"p95FrameDurationMs": 16.700000000000273
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 447.31899999999314,
"styleRecalcs": 47,
"styleRecalcDurationMs": 9.822,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 278.327,
"heapDeltaBytes": 13440508,
"heapUsedBytes": 76966484,
"domNodes": 20,
"jsHeapTotalBytes": 5242880,
"scriptDurationMs": 96.47,
"eventListeners": 8,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000273
},
{
"name": "subgraph-idle",
"durationMs": 1992.274000000009,
"styleRecalcs": 11,
"styleRecalcDurationMs": 7.547999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 293.502,
"heapDeltaBytes": -1845128,
"heapUsedBytes": 61487596,
"domNodes": 21,
"jsHeapTotalBytes": 4718592,
"scriptDurationMs": 7.436999999999999,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-idle",
"durationMs": 1995.2359999999771,
"styleRecalcs": 10,
"styleRecalcDurationMs": 5.190000000000002,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 299.152,
"heapDeltaBytes": -1579364,
"heapUsedBytes": 61545392,
"domNodes": 19,
"jsHeapTotalBytes": 4980736,
"scriptDurationMs": 8.681999999999999,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1677.2830000000454,
"styleRecalcs": 77,
"styleRecalcDurationMs": 20.993,
"layouts": 18,
"layoutDurationMs": 5.116,
"taskDurationMs": 516.508,
"heapDeltaBytes": 5547156,
"heapUsedBytes": 68519676,
"domNodes": -273,
"jsHeapTotalBytes": 5496832,
"scriptDurationMs": 77.517,
"eventListeners": -200,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1691.551000000004,
"styleRecalcs": 78,
"styleRecalcDurationMs": 24.773,
"layouts": 18,
"layoutDurationMs": 5.699999999999999,
"taskDurationMs": 580.073,
"heapDeltaBytes": 8182508,
"heapUsedBytes": 71779024,
"domNodes": -272,
"jsHeapTotalBytes": 5496832,
"scriptDurationMs": 88.967,
"eventListeners": -200,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "subgraph-transition-enter",
"durationMs": 784.3500000000176,
"styleRecalcs": 18,
"styleRecalcDurationMs": 19.77899999999999,
"layouts": 5,
"layoutDurationMs": 8.318999999999999,
"taskDurationMs": 570.6040000000002,
"heapDeltaBytes": 20020112,
"heapUsedBytes": 103388556,
"domNodes": 13587,
"jsHeapTotalBytes": 9961472,
"scriptDurationMs": 21.535,
"eventListeners": 2383,
"totalBlockingTimeMs": 75,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "viewport-pan-sweep",
"durationMs": 8212.550999999963,
"styleRecalcs": 249,
"styleRecalcDurationMs": 41.193999999999996,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 3770.8350000000005,
"heapDeltaBytes": -12301924,
"heapUsedBytes": 66728800,
"domNodes": -239,
"jsHeapTotalBytes": 3923968,
"scriptDurationMs": 1861.8970000000002,
"eventListeners": -134,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "viewport-pan-sweep",
"durationMs": 8367.230999999947,
"styleRecalcs": 250,
"styleRecalcDurationMs": 42.42700000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 3974.512,
"heapDeltaBytes": 13579448,
"heapUsedBytes": 93578496,
"domNodes": -212,
"jsHeapTotalBytes": 2052096,
"scriptDurationMs": 1907.0320000000002,
"eventListeners": -166,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-large-graph-idle",
"durationMs": 11876.781999999992,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 11274.122,
"heapDeltaBytes": -47848312,
"heapUsedBytes": 182817028,
"domNodes": -8061,
"jsHeapTotalBytes": -8183808,
"scriptDurationMs": 159.964,
"eventListeners": -14928,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.219999999999953,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-large-graph-idle",
"durationMs": 12195.71499999995,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 11654.441,
"heapDeltaBytes": -46911736,
"heapUsedBytes": 176150728,
"domNodes": -8062,
"jsHeapTotalBytes": -4415488,
"scriptDurationMs": 170.14399999999998,
"eventListeners": -14926,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-large-graph-pan",
"durationMs": 14300.44700000002,
"styleRecalcs": 72,
"styleRecalcDurationMs": 13.872999999999996,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 14045.039999999999,
"heapDeltaBytes": -26036860,
"heapUsedBytes": 188957236,
"domNodes": -8062,
"jsHeapTotalBytes": -18255872,
"scriptDurationMs": 599.5429999999999,
"eventListeners": -14928,
"totalBlockingTimeMs": 11,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "vue-large-graph-pan",
"durationMs": 14160.04399999997,
"styleRecalcs": 69,
"styleRecalcDurationMs": 13.821,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 13869.619,
"heapDeltaBytes": -46089152,
"heapUsedBytes": 185518544,
"domNodes": -8062,
"jsHeapTotalBytes": -15593472,
"scriptDurationMs": 616.675,
"eventListeners": -14926,
"totalBlockingTimeMs": 7,
"frameDurationMs": 17.219999999999953,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-node-resize-workload",
"durationMs": 4440.360000000055,
"styleRecalcs": 20,
"styleRecalcDurationMs": 55.76700000000001,
"layouts": 20,
"layoutDurationMs": 22.357000000000003,
"taskDurationMs": 4343.932000000001,
"heapDeltaBytes": -34716464,
"heapUsedBytes": 197120408,
"domNodes": -8062,
"jsHeapTotalBytes": 19865600,
"scriptDurationMs": 70.843,
"eventListeners": -14928,
"totalBlockingTimeMs": 73,
"frameDurationMs": 17.220000000000013,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-node-resize-workload",
"durationMs": 4019.044000000008,
"styleRecalcs": 20,
"styleRecalcDurationMs": 58.52199999999999,
"layouts": 20,
"layoutDurationMs": 21.371000000000002,
"taskDurationMs": 3925.15,
"heapDeltaBytes": -53510620,
"heapUsedBytes": 176140576,
"domNodes": -8062,
"jsHeapTotalBytes": 12075008,
"scriptDurationMs": 86.46799999999999,
"eventListeners": -14926,
"totalBlockingTimeMs": 1,
"frameDurationMs": 16.66333333333338,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-zoom-out-idle",
"durationMs": 5975.526999999943,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 5809.338000000001,
"heapDeltaBytes": 2167184,
"heapUsedBytes": 182938472,
"domNodes": 0,
"jsHeapTotalBytes": -11534336,
"scriptDurationMs": 83.63199999999999,
"eventListeners": -8,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.219999999999953,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-zoom-out-idle",
"durationMs": 5796.338999999989,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 5381.901000000001,
"heapDeltaBytes": 2439664,
"heapUsedBytes": 165490224,
"domNodes": 0,
"jsHeapTotalBytes": -24117248,
"scriptDurationMs": 72.14199999999998,
"eventListeners": -8,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.219999999999953,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "workflow-execution",
"durationMs": 117.13400000007823,
"styleRecalcs": 8,
"styleRecalcDurationMs": 16.064999999999998,
"layouts": 3,
"layoutDurationMs": 1.39,
"taskDurationMs": 97.65899999999999,
"heapDeltaBytes": 3180648,
"heapUsedBytes": 65454568,
"domNodes": 109,
"jsHeapTotalBytes": 262144,
"scriptDurationMs": 12.617999999999999,
"eventListeners": 33,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000273
},
{
"name": "workflow-execution",
"durationMs": 73.28400000005786,
"styleRecalcs": 5,
"styleRecalcDurationMs": 9.943999999999999,
"layouts": 3,
"layoutDurationMs": 1.375,
"taskDurationMs": 47.89399999999999,
"heapDeltaBytes": 3012616,
"heapUsedBytes": 65215632,
"domNodes": 122,
"jsHeapTotalBytes": 0,
"scriptDurationMs": 3.638000000000001,
"eventListeners": 17,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000273
}
]
}🔬 E2E Coverage
Top 10 files by uncovered lines
|
Codecov Report✅ All modified and coverable lines are covered by tests. @@ Coverage Diff @@
## main #17505 +/- ##
==========================================
+ Coverage 82.69% 83.70% +1.01%
==========================================
Files 2782 2201 -581
Lines 139995 126230 -13765
Branches 37536 34372 -3164
==========================================
- Hits 115764 105665 -10099
+ Misses 23144 20277 -2867
+ Partials 1087 288 -799
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
|
The The ECS Area 3 row is "clear a promoted text widget, save, reload. It is still empty — not
The fix is one extra save/reload: await promotedText.fill('value that must be cleared')
await comfyPage.menu.topbar.saveWorkflow(name)
await comfyPage.workflow.reloadAndWaitForApp()
await expect(promotedText).toHaveValue('value that must be cleared') // control
await promotedText.fill('')
await comfyPage.menu.topbar.saveWorkflow(name) // or re-save
await comfyPage.workflow.reloadAndWaitForApp()
await expect(promotedText).toHaveValue('') // now discriminatingThat first round does two jobs: it proves promoted text is persisted at all (so a green run Everything else here is stronger than what I had open in parallel: the renderer matrix, the Overlap note so we don't merge both: this duplicates |
|
Applied the control in What changed, only in the
The With this in, #17524 and Verified locally: |
DrJKL
left a comment
There was a problem hiding this comment.
Approved under the christian-byrne review policy. I found no security issue or shipped-code correctness defect. Four nonblocking test follow-ups are ticketed:
- FE-2376: distinguish durable saves from draft restoration, add the combo fixture precondition, and remove the redundant
isModifieddurability poll. - FE-2377: restore the persisted renderer setting and add named phases to the multi-action persistence flows.
Coverage and claim verification
- Exact empty promoted text is covered in both renderers. Removing the second save made both variants restore
value that must be cleared, so those assertions falsify the missing-clear persistence behavior. - Legacy multiline uses exact paragraph and special-character assertions. Saving the fixture baseline instead of the edited value made the test restore
Resize alignment probe text, so the assertion discriminates. - The legacy combo fixture currently starts at
simpleand selectskarras, but the test can pass through local draft restoration without the post-edit durable save. That limitation is tracked in FE-2376. - PR #17460 contains the claimed Vue multiline and supported connect/disconnect coverage. PR #17264 overlaps in subject matter, but its missing-node, partial named/positional, write-order, and reconciliation cases are outside this PR's explicit QA 022/025/026 scope.
- The PR description's "Knip green" claim is unverified and was not counted. This checkout has no
DATABASE_URL, and Knip can exit 0 without performing the analysis.
Local checks at ea56ed4
pnpm typecheck: passed.pnpm typecheck:browser: passed.- Focused Chromium for the four added cases: 4/4 passed after the cold Vite module graph warmed.
- Focused Vitest over the changed paths: not applicable; Vitest discovered no tests because both files are Playwright specs, so this is not counted as a pass.
- Scoped ESLint, type-aware Oxlint, and Oxfmt: passed.
git diff --check: passed.- Diff-scoped
pnpm fallow:audit: zero introduced findings. Stored baselines did not match, so this is not reported as a generic Fallow pass.
REST CI snapshot
At the reviewed head, REST returned 94 status contexts: 77 success, 16 skipped, 0 failed, and 1 pending. This combines 91 check runs with 3 commit-status contexts.
Ledger: posted-inline 3 + posted-body/ticketed 1 + confirmed-existing 1 + merged 1 + rejected-with-evidence 2 + question 0 + unverified 1 = 9 total candidates.
| await expect.poll(() => scheduler.getValue()).toBe('karras') | ||
| await comfyPage.menu.topbar.saveWorkflow('legacy-combo-persistence') | ||
|
|
||
| await comfyPage.workflow.reloadAndWaitForApp() |
There was a problem hiding this comment.
[P2] This reload does not isolate the durable save. At this exact head, a mutation that saved the fixture baseline, switched to karras, removed the post-edit save, disabled autosave, and cleared existing draft keys still passed because navigation flushed and restored the pending local draft. Please reopen the saved workflow through a draft-independent path and assert that the fixture starts at a value other than karras; otherwise this test does not prove the stated full save/reload boundary. Tracked in FE-2376.
| .activeWorkflow?.isModified | ||
| ) | ||
| ) | ||
| .toBe(false) |
There was a problem hiding this comment.
[P3] isModified is UI dirty state, not an independent durability signal. executeCommand already awaits Comfy.SaveWorkflow, whose command path awaits workflowService.saveWorkflow(), and this value can already be false. Please remove this poll and the WorkspaceStore dependency, or assert the saved artifact if separate durability proof is needed. Tracked in FE-2376.
| comfyPage | ||
| }) => { | ||
| await comfyPage.settings.setSetting( | ||
| 'Comfy.VueNodes.Enabled', |
There was a problem hiding this comment.
[P2] This server-persisted setting is left enabled by the final loop iteration. The base comfyPage fixture sets it for @vue-nodes tests but does not reset it for untagged tests, so a later test using the same worker user can inherit the wrong renderer. The new legacy spec likewise writes the setting without restoring it. Please reset it in cleanup or move renderer selection into fixture-owned setup that restores the prior value. Tracked in FE-2377.
The fixture ships this widget empty on the host AND on the interior
CLIPTextEncode, and the test cleared the value before its only save. So "empty
after reload" was satisfied by a persisted empty value and by a total reset to
the interior default alike — which is exactly the outcome the ECS plan row says
to watch for ("still empty, NOT reset to a default").
Now persists the non-empty value and reloads once first. That reload is a
control (promoted text is recorded at all, so a green run is not green because
the mechanism never ran) and it leaves something in storage that a falsy drop
brings back — the failure then names the bug instead of passing quietly.
The second save goes through Comfy.SaveWorkflow rather than the topbar helper:
the workflow is already persisted, so it saves in place, while a second topbar
Save reopens the dialog and hits the overwrite confirmation.
ea56ed4 to
c019435
Compare
Playwright videos for new testsDownload the recorded videos and HTML report Recorded projects: chromium Recorded test titles:
|
DrJKL
left a comment
There was a problem hiding this comment.
Verdict: changes required
Two blockers remain at c0194355d7d216cc8f3320cc7845b39c2e18b402:
- The legacy combo case does not yet prove durable save persistence. At this exact head I saved the baseline, disabled autosave, changed
simpletokarras, omitted the post-edit save, cleared the current persisted draft keys, and reloaded; the test still passed becausereloadAndWaitForApp()flushes/restores the pending local draft. The assertion therefore cannot distinguish the intended save path from draft restoration. This is the unresolved issue in the existing thread, tracked by FE-2376. - The current base resets
Comfy.VueNodes.Enabledto false before each test, so the added runtime setter is now redundant and fails the type-awarecomfy(prefer-initial-settings)rule. The PR'slint-and-formatcheck is red for the same reason. See the inline comment; this overlaps FE-2377, so I did not create a duplicate ticket.
Nonblocking prior findings remain: the isModified === false poll after awaiting Comfy.SaveWorkflow is not an independent durability signal (FE-2376), and the multi-phase scenarios still lack test.step phases (FE-2377). The prior renderer-setting leak itself is fixed by the current base's per-test reset. The missing combo-fixture precondition is folded into FE-2376.
Coverage falsification remains mixed: removing the second save makes both empty-promoted-text variants restore the nonempty sentinel, and persisting the legacy multiline baseline makes its exact-value assertion fail. The combo mutation above still passes, so that claimed gap is not covered. PR #17460 contains the referenced Vue multiline and supported connect/disconnect coverage; PR #17264's missing-node/reconciliation cases remain outside this PR's stated QA 022/025/026 scope. The PR description's Knip claim remains unverified because Knip can exit 0 without DATABASE_URL.
Since the last DrJKL review, the author rebased the unchanged patch hunks onto the current base, producing this head; there were no human replies, review comments, or reviews.
Local checks at the exact head:
pnpm typecheckandpnpm typecheck:browserpassed.- Scoped ESLint, scoped Oxfmt, and
git diff --checkpassed. - Scoped type-aware Oxlint failed only on the inline
prefer-initial-settingsviolation. - Focused Chromium: three cases passed on the first run; the Vue-disabled case hit a cold fixture startup timeout before its test body, then passed on the exact warm rerun (four cases passing after warmup).
- Diff-scoped
pnpm fallow:auditfound zero introduced findings. Stored baselines did not match, so this is not a generic Fallow pass. - REST CI snapshot immediately before this review: 76 success, 16 skipped, 1 failed, 1 pending (94 contexts). The failure is
lint-and-format; CodeRabbit is pending.
Ledger: posted-inline 1 + posted-body/ticketed 2 + confirmed-existing 2 + merged 1 + rejected-with-evidence 2 + question 0 + unverified 1 = 9 total candidates. Mapping: inline = redundant setting/lint; body/ticketed = combo durability and missing phases; confirmed-existing = redundant poll and #17460 coverage; merged = combo precondition; rejected-with-evidence = old setting leak and #17264 overlap; unverified = Knip.
| { tag: ['@widget', '@workflow'] }, | ||
| () => { | ||
| test.beforeEach(async ({ comfyPage }) => { | ||
| await comfyPage.settings.setSetting('Comfy.VueNodes.Enabled', false) |
There was a problem hiding this comment.
Blocking: The current base already resets Comfy.VueNodes.Enabled to false in the shared per-test setup. This extra runtime setter now fails the type-aware comfy(prefer-initial-settings) rule and keeps lint-and-format red. Please remove this line; no initialSettings replacement is needed because false is already the fixture default.
There was a problem hiding this comment.
Removed in 32d83cb. Local checks green.
Full context for agent readers
The shared fixture already sets Comfy.VueNodes.Enabled to false, so no initialSettings replacement is needed. Both legacyWidgetPersistence.spec.ts browser cases pass against a disposable CPU backend. Repository lint, application and browser typechecks, formatting, and pre-push Knip passed. Remote checks are running. This addresses the redundant setter only; the earlier durable-save follow-ups remain open. Leaving this thread for your verification.
Addresses review feedback: #17505 (comment)
DrJKL
left a comment
There was a problem hiding this comment.
Verdict: changes required
The follow-up fixes the new lint blocker. Commit 32d83cb6ab59046480b3a48f6594b130131fa5c9 removes the redundant Comfy.VueNodes.Enabled setter, the shared fixture still sets the value to false, and scoped type-aware Oxlint now passes. I verified the author's reply against the diff and fixture.
The legacy combo blocker is unchanged. The one-line follow-up does not touch that test's save and reload path, so the prior exact-head falsification still applies: after saving only the baseline, disabling autosave, changing simple to karras, omitting the post-edit save, clearing persisted draft keys, and reloading, the test passed through reloadAndWaitForApp()'s pending-draft path. It still cannot distinguish a durable save from draft restoration. The author also says the earlier durable-save follow-ups remain open. This remains tracked by FE-2376 and the existing discussion.
The prior nonblocking findings also remain unchanged: the isModified === false poll after awaited Comfy.SaveWorkflow is not an independent durability signal (FE-2376), and the multi-phase scenarios still lack test.step phases (FE-2377). I found no new correctness, maintainability, idiomaticity, test, or documentation issue in the one-line follow-up.
Local checks at the exact head:
pnpm typecheck,pnpm typecheck:browser, scoped ESLint, scoped type-aware Oxlint, scoped Oxfmt, andgit diff --checkpassed.- The two affected Chromium cases exceeded the project's 15-second timeout while reloading through this orb's dev server. After resetting the disposable backend and allowing 60 seconds, both passed in 43.4 seconds. The timeout is not introduced by the one-line deletion, so I did not add a finding for it.
- Diff-scoped
pnpm fallow:auditfound zero introduced findings. Stored baselines did not match, so this is not a generic Fallow pass. - The PR description and reply's Knip claim remains unverified because Knip can exit 0 without
DATABASE_URL. - REST CI snapshot immediately before this review: 78 success, 18 skipped, 0 failed, 1 pending (97 contexts).
Ledger: posted-inline 0 + posted-body/ticketed 2 + confirmed-existing 2 + merged 1 + rejected-with-evidence 3 + question 0 + unverified 1 = 9 total candidates. The fixed setter moved from posted-inline to rejected-with-evidence; all other dispositions are unchanged.
`reloadAndWaitForApp()` deliberately preserves localStorage, and the unload flush rewrites the pending draft that the next boot reads back, so the combo case passed even with the post-edit save removed: it could not tell a durable save from draft restoration. It now asserts the fixture precondition (`simple`, not `karras`), asserts the save cleared the dirty flag, closes the tab, clears the persisted drafts, and reopens the workflow from the saved list, asserting the active path is the saved file. Adds `WorkflowHelper.clearPersistedDrafts()` for that. Also moves the empty-promoted-text renderer selection onto the `@vue-nodes` tag so the base fixture owns `Comfy.VueNodes.Enabled` and no longer leaves it enabled for whatever runs next on the same worker user, and drops the `isModified` poll after `Comfy.SaveWorkflow`, which `executeCommand` already awaits through `workflowService.saveWorkflow()`. Addresses review threads r4010635668 and the two nonblocking findings on PR #17505 (FE-2376, FE-2377).
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
DrJKL
left a comment
There was a problem hiding this comment.
Verdict: ready to approve
The durable-save blocker is fixed at 39c584e4c561d17b295d0e8a8e16a003b287d803. The combo case now checks the simple fixture precondition, waits for the save to clear the dirty state, closes the tab, removes draft payload and index keys, reloads, reopens the named workflow from the saved list, and verifies the server workflow path before asserting karras.
I reran the original falsifier. The exact-head test passed. Removing only saveWorkflow(SAVED_COMBO_WORKFLOW) made it fail at the dirty-state assertion with Expected: false, Received: true; the old draft-restoration false positive no longer passes. This addresses FE-2376 and the prior blocker.
The other prior findings are reconciled:
- The redundant
isModifiedpoll after awaitedComfy.SaveWorkflowis removed. - Renderer selection now uses the
@vue-nodestag, so the shared fixture owns setup and reset. Test listing confirms only the Vue variant receives that tag. - The redundant legacy renderer setter remains removed.
- Named
test.stepphases are still absent from the multi-action scenarios. This remains a nonblocking cleanup in FE-2377. - PR #17460 still contains the referenced Vue multiline and supported connect/disconnect coverage. PR #17264's missing-node and reconciliation cases remain outside this PR's QA 022/025/026 scope.
- The Knip claim remains unverified because Knip can exit 0 without
DATABASE_URL.
No new correctness, maintainability, idiomaticity, test, or documentation finding emerged. The new clearPersistedDrafts() helper owns one test concern, and its comments explain the unload/draft invariant that made the original test unfalsifiable.
Checks at the exact head:
pnpm typecheck,pnpm typecheck:browser, scoped ESLint, full type-awarepnpm oxlint:main, fullpnpm format:check, andgit diff --checkpassed.- Playwright listed all 28 tests in the two changed specs.
- The four affected Chromium cases passed in 1.9 minutes with a 90-second local verification timeout. The combo save-removal mutation failed as expected.
- Diff-scoped
pnpm fallow:auditfound zero introduced findings across all three changed files. Stored baselines did not match, so this is not a generic Fallow pass. - REST CI snapshot immediately before this review: 80 success, 18 skipped, 2 failed, 1 pending (101 contexts).
Ledger: posted-inline 0 + posted-body/ticketed 1 + confirmed-existing 1 + merged 1 + rejected-with-evidence 5 + question 0 + unverified 1 = 9 total candidates. The only remaining finding is the nonblocking FE-2377 phase-label cleanup.
… app boot Both renderer variants failed hosted shard 13 at the chromium project's 15s timeout, each exhausting retries inside `reloadAndWaitForApp()`. The test did two full app boots, and two boots do not fit that budget on a loaded runner. The first reload only served as a control that the sentinel was persisted at all, which never needed a browser. It now reads the saved workflow through a new `UserDataHelper.readText()`, and a second artifact assertion checks the sentinel is gone after the clear, so a dropped-as-falsy clear fails before the reload and localises the defect to the save rather than the load. Reading the file is also the stronger control, because a reload can be satisfied by a restored draft. One reload remains, for the browser-observable projection. No timeout was raised. Verified at the default 15s timeout: both cases pass in 10.7s and 11.1s single-worker, down from the 43s-90s that previously needed an inflated local timeout. Removing the post-clear save fails the new artifact assertion, so it has teeth. Under `--workers=2` on a loaded host they still exceed 15s, but so does the untouched `Promoted widget tuples are stable after full page reload boot path`, so that budget pressure is not specific to this change.
DrJKL
left a comment
There was a problem hiding this comment.
Verdict: ready to approve
The prior Chromium timeout blocker is fixed at 4763b231118c6896f99cabb30b7a37e67d21b533. The promoted-text cases now prove the nonempty sentinel reached the saved workflow by reading the server artifact, save the empty value, prove the sentinel left that artifact, and perform one browser reload to verify the rendered empty value. Hosted shard 13 and the aggregate e2e-status gate both pass at the unchanged 15-second project timeout.
The changed test still has a working falsifier. Both exact-head renderer variants pass. Removing only the post-clear Comfy.SaveWorkflow command makes the artifact assertion fail before reload because the saved JSON still contains value that must be cleared. A restored local draft cannot satisfy either server-artifact check.
Prior findings are reconciled:
- FE-2376's durable-save isolation and fixture precondition are fixed.
- The redundant post-save
isModifiedpoll is gone. - Renderer setup and cleanup remain owned by the
@vue-nodestag and shared fixture. - The redundant legacy renderer setter remains removed.
- Named
test.stepphases remain absent from the multi-action scenarios. This is still nonblocking and tracked by FE-2377. - PR #17460 still contains the referenced Vue multiline and supported connect/disconnect coverage. PR #17264's missing-node and reconciliation cases remain outside this PR's QA 022/025/026 scope.
- The Knip claim remains unverified because Knip can exit 0 without
DATABASE_URL.
No new correctness, maintainability, idiomaticity, test, or documentation finding emerged. UserDataHelper.readText() uses the existing userdata request boundary and gives the test a direct durable-state oracle without another app boot.
Checks at the exact head:
pnpm install --frozen-lockfilecompleted. It removed one stale catalog line, which I restored before verification; the reviewed worktree remained clean.pnpm typecheck,pnpm typecheck:browser, scoped ESLint, full type-awarepnpm oxlint:main, fullpnpm format:check, andgit diff --checkpassed.- Playwright listed both renderer variants. Both passed locally in 44.0 seconds with a 90-second verification timeout. This orb's Vite startup still exceeds 15 seconds, but the exact hosted shard that previously failed now passes at the project timeout.
- Removing the post-clear save failed at the new artifact assertion as expected.
- Diff-scoped
pnpm fallow:auditfound zero introduced findings across all four changed files. Stored baselines did not match, so this is not a generic Fallow pass. - REST CI snapshot immediately before this review: 82 success, 18 skipped, 0 failed, 1 pending (101 contexts).
Ledger: posted-inline 0 + posted-body/ticketed 1 + confirmed-existing 1 + merged 1 + rejected-with-evidence 6 + question 0 + unverified 1 = 10 total candidates. The prior timeout candidate moved to rejected-with-evidence after the exact hosted shard passed.
Summary
Adds the non-duplicated renderer boundaries for ECS widget persistence QA cases 022, 025, and 026.
Changes
Review Focus
Exact empty-string, non-default combo, and paragraph-preserving assertions after a full save/reload.
Feature flag
Comfy.VueNodes.Enabled(test matrix only; no runtime behavior change)Full context for agent readers
Case 022 uses the promoted subgraph text fixture, proves a non-empty positive control before clearing, and asserts both the rendered textbox and graph widget remain exactly empty after save/reload with Vue Nodes enabled and disabled. Case 025 adds only the missing legacy boundary because main already covers the Vue non-default combo round trip. Case 026 adds only the missing legacy boundary because the existing widget QA pull request covers Vue multiline persistence. Case 020 is not duplicated: that same pull request owns the supported connect/disconnect behavior, while current main explicitly deprecates conversion because widget sockets and controls coexist.
Verification: focused Chromium 4/4; Oxfmt, Oxlint, ESLint, typecheck, and Knip green. A mutation changing the promoted graph expectation from empty to
default restoredfailed as expected.