// @vitest-environment happy-dom import { describe, expect, it } from "vitest"; import { mount } from "@vue/test-utils"; import { normalizeGenerationMetadata, normalizePromptsForDisplay, sanitizePromptForDisplay, } from "../components/sidebar/parsers/geninfoParser.js"; import { parseComfyUIWorkflow } from "../components/sidebar/parsers/comfyWorkflowParser.js"; import SidebarGenerationSection from "../vue/components/panel/sidebar/SidebarGenerationSection.vue"; import { buildGenerationSectionState } from "../vue/components/panel/sidebar/generationSectionState.js"; describe("generation prompt sanitization", () => { it("drops filepath-like positive prompts from display", () => { expect( sanitizePromptForDisplay( "d:/__comfy_outputs/projects/veille/02_out/videos/260402/lidox.mp4", ), ).toBe(""); }); it("keeps natural-language prompts intact", () => { expect(sanitizePromptForDisplay("sunset over mountains, cinematic lighting")).toBe( "sunset over mountains, cinematic lighting", ); }); it("prevents alignment UI when the only prompt is a filepath", () => { const prompts = normalizePromptsForDisplay( "d:/__comfy_outputs/projects/veille/02_out/videos/260402/lidox.mp4", "", ); expect(prompts.positive).toBe(""); const state = buildGenerationSectionState({ id: 8050, kind: "video", prompt: "d:/__comfy_outputs/projects/veille/02_out/videos/260402/lidox.mp4", }); expect(state.kind).toBe("empty"); expect(state.positivePrompt).toBe(""); expect(state.showAlignment).toBe(false); }); it("builds a readable workflow label for API nodes", () => { const state = buildGenerationSectionState({ id: 13, kind: "video", geninfo: { engine: { type: "img2vid", sampler_mode: "api", api_provider: "happy_horse", }, checkpoint: { name: "happyhorse-1.0-i2v" }, positive: { value: "steampunk ship girl with binoculars" }, }, }); expect(state.workflowType).toBe("img2vid"); expect(state.workflowLabel).toBe("API Image-to-Video"); expect(state.workflowBadge).toBe("Happy Horse"); }); it("keeps prompt from metadata_raw when top-level geninfo is partial", () => { const state = buildGenerationSectionState({ id: 13008, kind: "image", filename: "Krea2_turbo_00001_.png", geninfo: { models: { unet: { name: "krea2_turbo_fp8_scaled" }, clip: { name: "qwen3vl_4b_fp8_scaled" }, vae: { name: "qwen_image_vae" }, }, seed: { value: 397747871968973 }, sampler: { name: "euler" }, steps: { value: 8 }, }, metadata_raw: { geninfo: { positive: { value: "A high-resolution, surreal digital illustration showing a hand holding a martini glass.", }, models: { unet: { name: "krea2_turbo_fp8_scaled" }, clip: { name: "qwen3vl_4b_fp8_scaled" }, vae: { name: "qwen_image_vae" }, }, seed: { value: 397747871968973 }, sampler: { name: "euler" }, steps: { value: 8 }, }, }, }); expect(state.kind).toBe("full"); expect(state.positivePrompt).toBe( "A high-resolution, surreal digital illustration showing a hand holding a martini glass.", ); expect(state.modelFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "krea2_turbo_fp8_scaled" }), ]), ); }); it("prefers canonical metadata_raw over stale top-level geninfo mirrors", () => { const state = buildGenerationSectionState({ id: 13009, kind: "video", geninfo: { positive: { value: "stale top-level prompt" }, seed: { value: 111 }, sampler: { name: "stale_sampler" }, steps: { value: 1 }, }, metadata_raw: { geninfo: { positive: { value: "fresh metadata_raw prompt" }, seed: { value: 42 }, sampler: { name: "euler_cfg_pp" }, steps: { value: 3 }, }, }, }); expect(state.positivePrompt).toBe("fresh metadata_raw prompt"); expect(state.samplingFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Seed", value: 42 }), expect.objectContaining({ label: "Sampler", value: "euler_cfg_pp" }), expect.objectContaining({ label: "Steps", value: 3 }), ]), ); }); it("deduplicates sampler passes when prompt and workflow contain the same sampler", () => { const prompt = { "1": { class_type: "KSampler", inputs: { seed: 42, steps: 3, cfg: 1, sampler_name: "euler_cfg_pp", scheduler: "simple", denoise: 1, }, }, "2": { class_type: "KSampler", inputs: { seed: 43, steps: 8, cfg: 1, sampler_name: "euler_ancestral_cfg_pp", scheduler: "simple", denoise: 1, }, }, }; const workflow = { nodes: [ { id: 1, type: "KSampler", inputs: prompt["1"].inputs }, { id: 2, type: "KSampler", inputs: prompt["2"].inputs }, ], }; const parsed = parseComfyUIWorkflow({ prompt, workflow }); expect(parsed?.all_samplers).toHaveLength(2); expect(parsed?.all_samplers).toEqual( expect.arrayContaining([ expect.objectContaining({ sampler_name: "euler_cfg_pp", seed: 42, steps: 3 }), expect.objectContaining({ sampler_name: "euler_ancestral_cfg_pp", seed: 43, steps: 8 }), ]), ); }); it("deduplicates a prompt sampler repeated inside a workflow subgraph", () => { const prompt = { "30:3": { class_type: "KSampler", inputs: { seed: 1103244806536411, steps: 8, cfg: 1, sampler_name: "euler", scheduler: "simple", denoise: 1, model: ["30:22", 0], }, }, "30:22": { class_type: "UNETLoader", inputs: { unet_name: "krea2_turbo_fp8_scaled.safetensors" }, }, }; const workflow = { definitions: { subgraphs: [ { nodes: [ { id: 3, type: "KSampler", inputs: [ { name: "seed", type: "INT", widget: { name: "seed" } }, { name: "steps", type: "INT", widget: { name: "steps" } }, { name: "cfg", type: "FLOAT", widget: { name: "cfg" } }, { name: "sampler_name", type: "COMBO", widget: { name: "sampler_name" } }, { name: "scheduler", type: "COMBO", widget: { name: "scheduler" } }, { name: "denoise", type: "FLOAT", widget: { name: "denoise" } }, ], widgets_values: [ 1103244806536411, 8, 1, "euler", "simple", 1, ], }, ], }, ], }, }; const parsed = parseComfyUIWorkflow({ prompt, workflow }); const state = buildGenerationSectionState({ id: 13013, kind: "image", metadata_raw: { prompt: JSON.stringify(prompt), workflow, }, }); expect(parsed?.all_samplers).toBeUndefined(); expect(state.pipelineTabs).toHaveLength(0); expect(state.branchCards.filter((branch) => branch.samplingFields.length)).toHaveLength(0); expect(state.samplingFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Seed", value: 1103244806536411 }), expect.objectContaining({ label: "Steps", value: 8 }), expect.objectContaining({ label: "Sampler", value: "euler" }), ]), ); }); it("keeps shared base CLIP VAE and LoRAs visible when branch cards are used", () => { const state = buildGenerationSectionState({ id: 13010, kind: "video", metadata_raw: { geninfo: { models: { checkpoint: { name: "ltx-2.3-22b-dev-fp8.safetensors" }, clip: { name: "t5xxl_fp8_e4m3fn.safetensors" }, vae: { name: "ltxv_vae.safetensors" }, }, loras: [ { name: "ltx2.3_audio_reactive_lora_v2.safetensors", strength_model: 1.2 }, { name: "ltx-2.3-22b-distilled-lora-384.safetensors", strength_model: 0.5 }, ], }, }, }); expect(state.branchCards).toEqual( expect.arrayContaining([ expect.objectContaining({ key: "base", label: "Base", modelFields: expect.arrayContaining([ expect.objectContaining({ label: "Model", value: "ltx-2.3-22b-dev-fp8" }), expect.objectContaining({ label: "CLIP", value: "t5xxl_fp8_e4m3fn" }), expect.objectContaining({ label: "VAE", value: "ltxv_vae" }), ]), loras: expect.arrayContaining([ "ltx2.3_audio_reactive_lora_v2 (m=1.2)", "ltx-2.3-22b-distilled-lora-384 (m=0.5)", ]), }), ]), ); }); it("keeps repeated base sampler passes in separate sidebar cards", () => { const state = buildGenerationSectionState({ id: 13011, kind: "video", metadata_raw: { geninfo: { chained_passes: [ { pass_stage: "base", sampler_name: "euler_ancestral_cfg_pp", steps: 8, cfg: 1, seed: 767185649304890, }, { pass_stage: "base", sampler_name: "euler_cfg_pp", steps: 3, seed: 42, }, ], }, }, }); const samplingCards = state.branchCards.filter((branch) => branch.samplingFields.length); expect(samplingCards).toHaveLength(2); expect(samplingCards.map((branch) => branch.key)).toEqual(["pass_1", "pass_2"]); expect(samplingCards.map((branch) => branch.label)).toEqual(["Pass 1", "Pass 2"]); expect(state.branchCards.filter((branch) => branch.modelFields.length)).toHaveLength(0); expect(samplingCards).toEqual( expect.arrayContaining([ expect.objectContaining({ key: "pass_1", samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_ancestral_cfg_pp" }), expect.objectContaining({ label: "Steps", value: "8" }), expect.objectContaining({ label: "Seed", value: "767185649304890" }), ]), }), expect.objectContaining({ key: "pass_2", samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_cfg_pp" }), expect.objectContaining({ label: "Steps", value: "3" }), expect.objectContaining({ label: "Seed", value: "42" }), ]), }), ]), ); }); it("uses the first positive sampler seed for the isolated seed block", () => { const state = buildGenerationSectionState({ id: 13012, kind: "video", metadata_raw: { geninfo: { seed: { value: 0 }, all_samplers: [ { pass_stage: "low", sampler_name: "euler", seed: 0 }, { pass_stage: "high", sampler_name: "euler", seed: 675458056445911 }, ], }, }, }); expect(state.seed).toBe(675458056445911); }); it("keeps explicit WAN high/low noise models distinct in generation state", () => { const state = buildGenerationSectionState({ id: 99, kind: "video", geninfo: { positive: { value: "living gelatinous blob" }, checkpoint: { name: "wan2.2_i2v_high_noise_14B_fp8_scaled" }, models: { unet: { name: "wan2.2_i2v_high_noise_14B_fp8_scaled" }, unet_high_noise: { name: "wan2.2_i2v_high_noise_14B_fp8_scaled" }, unet_low_noise: { name: "wan2.2_i2v_low_noise_14B_fp8_scaled" }, clip: { name: "umt5_xxl_fp8_e4m3fn_scaled" }, vae: { name: "wan_2.1_vae" }, }, model_groups: [ { key: "high_noise", label: "High Noise", model: { name: "wan2.2_i2v_high_noise_14B_fp8_scaled" }, loras: [{ name: "high_noise_model_rank64", strength_model: 0.5 }], }, { key: "low_noise", label: "Low Noise", model: { name: "wan2.2_i2v_low_noise_14B_fp8_scaled" }, loras: [{ name: "low_noise_model_rank64", strength_model: 1.0000000000000002 }], }, ], loras: [ { name: "high_noise_model_rank64", strength_model: 0.5 }, { name: "low_noise_model_rank64", strength_model: 1.0000000000000002 }, ], steps: { value: 8 }, cfg_high_noise: { value: 1.0 }, cfg_low_noise: { value: 1.0 }, }, }); expect(state.kind).toBe("full"); expect(state.modelGroups).toEqual([ { key: "high_noise", label: "High Noise", model: "wan2.2_i2v_high_noise_14B_fp8_scaled", loras: ["high_noise_model_rank64 (m=0.5)"], }, { key: "low_noise", label: "Low Noise", model: "wan2.2_i2v_low_noise_14B_fp8_scaled", loras: ["low_noise_model_rank64 (m=1)"], }, ]); expect(state.modelFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "CLIP", value: "umt5_xxl_fp8_e4m3fn_scaled" }), expect.objectContaining({ label: "VAE", value: "wan_2.1_vae" }), ]), ); expect(state.modelFields.find((item) => item.label === "UNet High Noise")).toBeFalsy(); expect(state.modelFields.find((item) => item.label === "UNet Low Noise")).toBeFalsy(); expect(state.modelFields.find((item) => item.label === "UNet")).toBeFalsy(); expect(state.samplingFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Steps", value: 8 }), expect.objectContaining({ label: "CFG High Noise", value: 1 }), expect.objectContaining({ label: "CFG Low Noise", value: 1 }), ]), ); expect(state.modelFields.find((item) => item.label === "LoRAs")?.value).toContain( "high_noise_model_rank64 (m=0.5)", ); expect(state.branchCards).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "High", modelFields: expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "wan2.2_i2v_high_noise_14B_fp8_scaled" }), ]), samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "CFG", value: "1" }), ]), loras: ["high_noise_model_rank64 (m=0.5)"], }), expect.objectContaining({ label: "Low", modelFields: expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "wan2.2_i2v_low_noise_14B_fp8_scaled" }), ]), loras: ["low_noise_model_rank64 (m=1)"], }), expect.objectContaining({ label: "Shared", modelFields: expect.arrayContaining([ expect.objectContaining({ label: "CLIP", value: "umt5_xxl_fp8_e4m3fn_scaled" }), expect.objectContaining({ label: "VAE", value: "wan_2.1_vae" }), ]), }), ]), ); expect(state.moduleBlocks.find((module) => module.title === "Upscale")).toBeFalsy(); }); it("summarizes multi-model geninfo for normalized viewer consumers", () => { const normalized = normalizeGenerationMetadata({ geninfo: { checkpoint: { name: "wan2.2_i2v_high_noise_14B_fp8_scaled" }, models: { unet_high_noise: { name: "wan2.2_i2v_high_noise_14B_fp8_scaled" }, unet_low_noise: { name: "wan2.2_i2v_low_noise_14B_fp8_scaled" }, clip: { name: "umt5_xxl_fp8_e4m3fn_scaled" }, }, model_groups: [{ key: "high_noise", label: "High Noise", model: { name: "wan2.2_i2v_high_noise_14B_fp8_scaled" }, loras: [] }], }, }); expect(normalized.models.unet_high_noise.name).toBe( "wan2.2_i2v_high_noise_14B_fp8_scaled", ); expect(normalized.models.unet_low_noise.name).toBe( "wan2.2_i2v_low_noise_14B_fp8_scaled", ); expect(normalized.model_groups[0].label).toBe("High Noise"); }); it("labels ComfyUI native sampler passes from upstream latent nodes", () => { const state = buildGenerationSectionState({ id: 101, kind: "image", metadata: { prompt: { "3": { class_type: "KSampler", inputs: { seed: 1, steps: 8, cfg: 1, sampler_name: "lcm", scheduler: "simple", denoise: 1, latent_image: ["5", 0], }, }, "5": { class_type: "EmptyLatentImage", inputs: { width: 512, height: 512, batch_size: 1 }, }, "9": { class_type: "LatentUpscale", inputs: { samples: ["3", 0], width: 1024, height: 1024 }, }, "10": { class_type: "KSampler", inputs: { seed: 2, steps: 4, cfg: 1, sampler_name: "lcm", scheduler: "simple", denoise: 0.35, latent_image: ["9", 0], }, }, }, }, }); expect(state.pipelineTabs.map((tab) => tab.label)).toEqual([ "Text-to-Image", "Upscale", ]); }); it("extracts sampler custom stages and models from linked ComfyUI nodes", () => { const state = buildGenerationSectionState({ id: 102, kind: "image", metadata: { prompt: { "244": { class_type: "LoadImage", inputs: { image: "source.png" } }, "95:29": { class_type: "VAELoader", inputs: { vae_name: "FLUX/ae.safetensors" } }, "95:28": { class_type: "UNETLoader", inputs: { unet_name: "Z-IMAGE/TURBO/z_image_turbo_bf16.safetensors" }, }, "95:13": { class_type: "EmptySD3LatentImage", inputs: { width: 1024, height: 1024 }, }, "95:242": { class_type: "VAEEncode", inputs: { pixels: ["244", 0], vae: ["95:29", 0] }, }, "95:243": { class_type: "ComfySwitchNode", inputs: { switch: true, on_false: ["95:13", 0], on_true: ["95:242", 0] }, }, "95:11": { class_type: "ModelSamplingAuraFlow", inputs: { model: ["95:28", 0] }, }, "95:3": { class_type: "KSampler", inputs: { seed: 100, steps: 8, cfg: 1, sampler_name: "res_multistep", scheduler: "simple", denoise: 1, model: ["95:11", 0], latent_image: ["95:243", 0], }, }, "94:82": { class_type: "UNETLoader", inputs: { unet_name: "pid_flux1_1024_to_4096_4step_bf16.safetensors" }, }, "94:76": { class_type: "KSamplerSelect", inputs: { sampler_name: "lcm" } }, "94:77": { class_type: "BasicScheduler", inputs: { scheduler: "simple", steps: 4, denoise: 1, model: ["94:82", 0] }, }, "94:84": { class_type: "EmptyChromaRadianceLatentImage", inputs: { width: 4096, height: 4096 }, }, "94:86": { class_type: "PiDConditioning", inputs: { latent: ["95:3", 0] }, }, "94:75": { class_type: "SamplerCustom", inputs: { noise_seed: 3, cfg: 1, model: ["94:82", 0], positive: ["94:86", 0], sampler: ["94:76", 0], sigmas: ["94:77", 0], latent_image: ["94:84", 0], }, }, }, }, }); expect(state.pipelineTabs.map((tab) => tab.label)).toEqual(["Image-to-Image", "Upscale"]); expect(state.pipelineTabs[0].fields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Model", value: "Z-IMAGE/TURBO/z_image_turbo_bf16.safetensors", }), expect.objectContaining({ label: "Sampler", value: "res_multistep" }), ]), ); expect(state.pipelineTabs[1].fields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Model", value: "pid_flux1_1024_to_4096_4step_bf16.safetensors", }), expect.objectContaining({ label: "Sampler", value: "lcm" }), expect.objectContaining({ label: "Steps", value: "4" }), ]), ); expect(state.moduleBlocks).toEqual( expect.arrayContaining([ expect.objectContaining({ title: "Upscale", fields: expect.arrayContaining([ expect.objectContaining({ label: "Model", value: "pid_flux1_1024_to_4096_4step_bf16.safetensors", }), ]), }), ]), ); }); it("resolves SamplerCustomAdvanced values through guider sampler noise and sigmas links", () => { const state = buildGenerationSectionState({ id: 103, kind: "video", metadata: { prompt: { "276": { class_type: "RandomNoise", inputs: { noise_seed: 42 } }, "277": { class_type: "RandomNoise", inputs: { noise_seed: 221923341291547 } }, "278": { class_type: "LTXVConcatAVLatent", inputs: { video_latent: ["288", 0], audio_latent: ["307", 1] } }, "280": { class_type: "KSamplerSelect", inputs: { sampler_name: "euler_cfg_pp" } }, "281": { class_type: "ManualSigmas", inputs: { sigmas: "0.85, 0.7250, 0.4219, 0.0" } }, "282": { class_type: "CFGGuider", inputs: { cfg: 1, model: ["285", 0] } }, "283": { class_type: "SamplerCustomAdvanced", inputs: { noise: ["277", 0], guider: ["314", 0], sampler: ["291", 0], sigmas: ["306", 0], latent_image: ["318", 0], }, }, "285": { class_type: "LoraLoaderModelOnly", inputs: { lora_name: "ltx-2.3-22b-distilled-lora-384.safetensors", strength_model: 0.5, model: ["316", 0], }, }, "291": { class_type: "KSamplerSelect", inputs: { sampler_name: "euler_ancestral_cfg_pp" } }, "287": { class_type: "LTXVLatentUpsampler", inputs: { samples: ["307", 0], upscale_model: ["311", 0] } }, "288": { class_type: "LTXVImgToVideoInplace", inputs: { latent: ["287", 0] } }, "296": { class_type: "LTXVImgToVideoInplace", inputs: {} }, "306": { class_type: "ManualSigmas", inputs: { sigmas: "1.0, 0.99375, 0.0" } }, "307": { class_type: "LTXVSeparateAVLatent", inputs: { av_latent: ["283", 0] } }, "308": { class_type: "SamplerCustomAdvanced", inputs: { noise: ["276", 0], guider: ["282", 0], sampler: ["280", 0], sigmas: ["281", 0], latent_image: ["278", 0], }, }, "311": { class_type: "LatentUpscaleModelLoader", inputs: { model_name: "ltx-2.3-spatial-upscaler-x2-1.1.safetensors" }, }, "314": { class_type: "CFGGuider", inputs: { cfg: 1, model: ["285", 0] } }, "316": { class_type: "CheckpointLoaderSimple", inputs: { ckpt_name: "ltx-2.3-22b-dev-fp8.safetensors" } }, "318": { class_type: "LTXVConcatAVLatent", inputs: { video_latent: ["296", 0], audio_latent: ["305", 0] } }, }, }, }); expect(state.modelFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Model", value: "ltx-2.3-22b-dev-fp8" }), expect.objectContaining({ label: "LoRA", value: expect.stringContaining("ltx-2.3-22b-distilled-lora-384") }), expect.objectContaining({ label: "Upscaler", value: "ltx-2.3-spatial-upscaler-x2-1.1" }), ]), ); expect(state.branchCards).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Low", samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_cfg_pp" }), expect.objectContaining({ label: "Seed", value: "42" }), expect.objectContaining({ label: "CFG", value: "1" }), ]), }), expect.objectContaining({ label: "High", samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_ancestral_cfg_pp" }), expect.objectContaining({ label: "Seed", value: "221923341291547" }), ]), }), ]), ); }); it("keeps LTX two-pass SamplerCustomAdvanced visible when steps come from ManualSigmas", () => { const state = buildGenerationSectionState({ id: 105, kind: "video", metadata: { prompt: { "285": { class_type: "RandomNoise", inputs: { noise_seed: 42 } }, "286": { class_type: "RandomNoise", inputs: { noise_seed: 89173056758850 } }, "288": { class_type: "KSamplerSelect", inputs: { sampler_name: "euler_cfg_pp" } }, "289": { class_type: "ManualSigmas", inputs: { sigmas: "0.85, 0.7250, 0.4219, 0.0" } }, "290": { class_type: "CFGGuider", inputs: { cfg: 1, model: ["349", 0] } }, "291": { class_type: "SamplerCustomAdvanced", inputs: { noise: ["286", 0], guider: ["315", 0], sampler: ["298", 0], sigmas: ["308", 0], latent_image: ["326", 0], }, }, "298": { class_type: "KSamplerSelect", inputs: { sampler_name: "euler_ancestral_cfg_pp" } }, "308": { class_type: "ManualSigmas", inputs: { sigmas: "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0" } }, "310": { class_type: "SamplerCustomAdvanced", inputs: { noise: ["285", 0], guider: ["290", 0], sampler: ["288", 0], sigmas: ["289", 0], latent_image: ["287", 0], }, }, "315": { class_type: "CFGGuider", inputs: { cfg: 1, model: ["349", 0] } }, "317": { class_type: "CheckpointLoaderSimple", inputs: { ckpt_name: "ltx-2.3-22b-dev-fp8.safetensors" } }, "349": { class_type: "LoraLoaderModelOnly", inputs: { lora_name: "ltx2.3_audio_reactive_lora_v2.safetensors", strength_model: 1.5, model: ["317", 0], }, }, }, }, }); expect(state.kind).toBe("full"); expect(state.branchCards.some((branch) => branch.modelFields.length || branch.loras.length)).toBe(true); expect(state.branchCards.some((branch) => branch.samplingFields.length)).toBe(true); expect(state.branchCards).toEqual( expect.arrayContaining([ expect.objectContaining({ samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_cfg_pp" }), expect.objectContaining({ label: "Steps", value: "3" }), expect.objectContaining({ label: "CFG", value: "1" }), ]), }), expect.objectContaining({ samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_ancestral_cfg_pp" }), expect.objectContaining({ label: "Steps", value: "8" }), ]), }), ]), ); }); it("keeps embedded two-pass sampler data when geninfo only exposes one pass", () => { const promptGraph = { "285": { class_type: "RandomNoise", inputs: { noise_seed: 42 } }, "286": { class_type: "RandomNoise", inputs: { noise_seed: 89173056758850 } }, "288": { class_type: "KSamplerSelect", inputs: { sampler_name: "euler_cfg_pp" } }, "289": { class_type: "ManualSigmas", inputs: { sigmas: "0.85, 0.7250, 0.4219, 0.0" } }, "290": { class_type: "CFGGuider", inputs: { cfg: 1, model: ["349", 0] } }, "291": { class_type: "SamplerCustomAdvanced", inputs: { noise: ["286", 0], guider: ["315", 0], sampler: ["298", 0], sigmas: ["308", 0], latent_image: ["326", 0], }, }, "298": { class_type: "KSamplerSelect", inputs: { sampler_name: "euler_ancestral_cfg_pp" } }, "308": { class_type: "ManualSigmas", inputs: { sigmas: "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0" } }, "310": { class_type: "SamplerCustomAdvanced", inputs: { noise: ["285", 0], guider: ["290", 0], sampler: ["288", 0], sigmas: ["289", 0], latent_image: ["287", 0], }, }, "315": { class_type: "CFGGuider", inputs: { cfg: 1, model: ["349", 0] } }, "317": { class_type: "CheckpointLoaderSimple", inputs: { ckpt_name: "ltx-2.3-22b-dev-fp8.safetensors" } }, "349": { class_type: "LoraLoaderModelOnly", inputs: { lora_name: "ltx2.3_audio_reactive_lora_v2.safetensors", strength_model: 1.5, model: ["317", 0], }, }, }; const state = buildGenerationSectionState({ id: 106, kind: "video", metadata_raw: { geninfo: { seed: { value: 42 }, sampler: { name: "euler_cfg_pp" }, steps: { value: 3 }, cfg: { value: 1 }, denoise: { value: 1 }, }, raw_ffprobe: { format: { tags: { prompt: JSON.stringify(promptGraph), }, }, }, }, }); expect(state.pipelineTabs).toHaveLength(2); expect(state.branchCards.some((branch) => branch.samplingFields.length)).toBe(true); expect(state.branchCards).toEqual( expect.arrayContaining([ expect.objectContaining({ samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_cfg_pp" }), expect.objectContaining({ label: "Steps", value: "3" }), expect.objectContaining({ label: "Seed", value: "42" }), ]), }), expect.objectContaining({ samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Sampler", value: "euler_ancestral_cfg_pp" }), expect.objectContaining({ label: "Steps", value: "8" }), expect.objectContaining({ label: "Seed", value: "89173056758850" }), ]), }), ]), ); }); it("renders generic sampling when pipeline tabs exist without branch sampling cards", () => { const wrapper = mount(SidebarGenerationSection, { props: { asset: { kind: "video", metadata: { prompt: "generated scene", seed: 42, sampler: "euler_cfg_pp", scheduler: "simple", steps: 8, cfg: 1, chained_passes: [{ pass_stage: "base" }, { pass_stage: "upscale" }], }, }, }, global: { stubs: { MButton: { template: "" }, GenerationInputThumb: true, }, }, }); const text = wrapper.text(); expect(text).toContain("Sampling"); expect(text).toContain("euler_cfg_pp"); expect(text).toContain("42"); wrapper.unmount(); }); it("auto-builds sidebar fields from embedded ComfyUI prompt tags for any asset kind", () => { const promptGraph = { "1": { class_type: "CheckpointLoaderSimple", inputs: { ckpt_name: "universal-model.safetensors" }, }, "2": { class_type: "CLIPTextEncode", inputs: { text: "cinematic universal prompt", clip: ["1", 1] }, _meta: { title: "CLIP Text Encode (Prompt)" }, }, "3": { class_type: "KSampler", inputs: { seed: 123, steps: 9, cfg: 1.5, sampler_name: "euler", scheduler: "simple", model: ["1", 0], positive: ["2", 0], }, }, }; for (const kind of ["image", "video", "audio", "3d"]) { const state = buildGenerationSectionState({ id: `${kind}-embedded`, kind, metadata_raw: { prompt: "D:/outputs/path-like-preview.mp4", raw_ffprobe: { format: { tags: { prompt: JSON.stringify(promptGraph), }, }, }, }, }); expect(state.kind, kind).toBe("full"); expect(state.positivePrompt, kind).toBe("cinematic universal prompt"); expect(state.samplingFields, kind).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Seed", value: 123 }), expect.objectContaining({ label: "Sampler", value: "euler" }), expect.objectContaining({ label: "Steps", value: 9 }), expect.objectContaining({ label: "CFG Scale", value: 1.5 }), ]), ); expect(state.modelFields, kind).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Model", value: "universal-model" }), ]), ); } }); it("resolves Wan high low samplers through Comfy switches", () => { const prompt = { "131": { class_type: "PrimitiveBoolean", inputs: { value: true } }, "118": { class_type: "PrimitiveInt", inputs: { value: 4 } }, "122": { class_type: "PrimitiveFloat", inputs: { value: 1.0 } }, "124": { class_type: "PrimitiveInt", inputs: { value: 2 } }, "128": { class_type: "PrimitiveInt", inputs: { value: 20 } }, "119": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_false: ["128", 0], on_true: ["118", 0] }, }, "120": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_false: 3.5, on_true: ["122", 0] }, }, "125": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_false: 10, on_true: ["124", 0] }, }, "95": { class_type: "UNETLoader", inputs: { unet_name: "WAN2.2/I2V/wan2.2_i2v_high_noise_14B_fp8_scaled.safetensors" }, }, "96": { class_type: "UNETLoader", inputs: { unet_name: "WAN2.2/I2V/wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors" }, }, "101": { class_type: "LoraLoaderModelOnly", inputs: { lora_name: "WAN2.2/I2V/high_noise_lora.safetensors", strength_model: 1.0000000000000002, model: ["95", 0], }, }, "102": { class_type: "LoraLoaderModelOnly", inputs: { lora_name: "WAN2.2/I2V/low_noise_lora.safetensors", strength_model: 1.0000000000000002, model: ["96", 0], }, }, "116": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_false: ["95", 0], on_true: ["101", 0] }, }, "117": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_false: ["96", 0], on_true: ["102", 0] }, }, "103": { class_type: "ModelSamplingSD3", inputs: { shift: 5, model: ["117", 0] } }, "104": { class_type: "ModelSamplingSD3", inputs: { shift: 5, model: ["116", 0] } }, "85": { class_type: "KSamplerAdvanced", inputs: { noise_seed: 0, steps: ["119", 0], cfg: ["120", 0], sampler_name: "euler", scheduler: "simple", start_at_step: ["125", 0], end_at_step: ["119", 0], model: ["103", 0], }, }, "86": { class_type: "KSamplerAdvanced", inputs: { noise_seed: 555, steps: ["119", 0], cfg: ["120", 0], sampler_name: "euler", scheduler: "simple", start_at_step: 0, end_at_step: ["125", 0], model: ["104", 0], }, }, }; const state = buildGenerationSectionState({ id: 104, kind: "video", metadata: { prompt: JSON.stringify(prompt), }, }); expect(state.branchCards).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Low", modelFields: expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "wan2.2_i2v_low_noise_14B_fp8_scaled" }), ]), samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Steps", value: "4" }), expect.objectContaining({ label: "CFG", value: "1" }), expect.objectContaining({ label: "Start", value: "2" }), expect.objectContaining({ label: "End", value: "4" }), ]), loras: ["low_noise_lora (m=1)"], }), expect.objectContaining({ label: "High", modelFields: expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "wan2.2_i2v_high_noise_14B_fp8_scaled" }), ]), samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Start", value: "0" }), expect.objectContaining({ label: "End", value: "2" }), ]), loras: ["high_noise_lora (m=1)"], }), ]), ); }); it("keeps Wan high and low branches when geninfo is low-only but embedded prompt has both", () => { const prompt = { "131": { class_type: "PrimitiveBoolean", inputs: { value: true } }, "118": { class_type: "PrimitiveInt", inputs: { value: 4 } }, "122": { class_type: "PrimitiveFloat", inputs: { value: 1.0 } }, "124": { class_type: "PrimitiveInt", inputs: { value: 2 } }, "119": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_true: ["118", 0] }, }, "120": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_true: ["122", 0] }, }, "125": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_true: ["124", 0] }, }, "95": { class_type: "UNETLoader", inputs: { unet_name: "WAN2.2/I2V/wan2.2_i2v_high_noise_14B_fp8_scaled.safetensors" }, }, "96": { class_type: "UNETLoader", inputs: { unet_name: "WAN2.2/I2V/wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors" }, }, "101": { class_type: "LoraLoaderModelOnly", inputs: { lora_name: "WAN2.2/I2V/wan2.2_i2v_lightx2v_4steps_lora_v1_high_noise.safetensors", strength_model: 1, model: ["95", 0], }, }, "102": { class_type: "LoraLoaderModelOnly", inputs: { lora_name: "WAN2.2/I2V/wan2.2_i2v_lightx2v_4steps_lora_v1_low_noise.safetensors", strength_model: 1, model: ["96", 0], }, }, "116": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_false: ["95", 0], on_true: ["101", 0] }, }, "117": { class_type: "ComfySwitchNode", inputs: { switch: ["131", 0], on_false: ["96", 0], on_true: ["102", 0] }, }, "103": { class_type: "ModelSamplingSD3", inputs: { model: ["117", 0] } }, "104": { class_type: "ModelSamplingSD3", inputs: { model: ["116", 0] } }, "85": { class_type: "KSamplerAdvanced", inputs: { noise_seed: 0, steps: ["119", 0], cfg: ["120", 0], sampler_name: "euler", scheduler: "simple", start_at_step: ["125", 0], model: ["103", 0], }, }, "86": { class_type: "KSamplerAdvanced", inputs: { noise_seed: 675458056445911, steps: ["119", 0], cfg: ["120", 0], sampler_name: "euler", scheduler: "simple", end_at_step: ["125", 0], model: ["104", 0], }, }, }; const state = buildGenerationSectionState({ id: 260415, kind: "video", metadata_raw: { geninfo: { models: { unet: { name: "wan2.2_i2v_low_noise_14B_fp8_scaled" }, clip: { name: "umt5_xxl_fp8_e4m3fn_scaled" }, vae: { name: "wan_2.1_vae" }, }, loras: [ { name: "wan2.2_i2v_lightx2v_4steps_lora_v1_low_noise", strength_model: 1 }, ], all_samplers: [ { pass_stage: "low", sampler_name: "euler", steps: 4, seed: 0 }, { pass_stage: "base", sampler_name: "euler", seed: 675458056445911 }, ], }, raw_ffprobe: { format: { tags: { prompt: JSON.stringify(prompt), }, }, }, }, }); expect(state.branchCards.find((branch) => branch.label === "Base")).toBeFalsy(); expect(state.moduleBlocks.find((module) => module.title === "Upscale")).toBeFalsy(); expect(state.branchCards).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Shared", modelFields: expect.arrayContaining([ expect.objectContaining({ label: "CLIP", value: "umt5_xxl_fp8_e4m3fn_scaled" }), expect.objectContaining({ label: "VAE", value: "wan_2.1_vae" }), ]), }), ]), ); const highCard = state.branchCards.find((branch) => branch.label === "High"); const lowCard = state.branchCards.find((branch) => branch.label === "Low"); expect(highCard).toEqual( expect.objectContaining({ modelFields: expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "wan2.2_i2v_high_noise_14B_fp8_scaled" }), ]), samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Seed", value: "675458056445911" }), ]), loras: ["wan2.2_i2v_lightx2v_4steps_lora_v1_high_noise (m=1)"], }), ); expect(lowCard).toEqual( expect.objectContaining({ modelFields: expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "wan2.2_i2v_low_noise_14B_fp8_scaled" }), ]), samplingFields: expect.arrayContaining([ expect.objectContaining({ label: "Seed", value: "0" }), ]), loras: ["wan2.2_i2v_lightx2v_4steps_lora_v1_low_noise (m=1)"], }), ); expect(highCard?.modelFields.some((field) => ["CLIP", "VAE"].includes(field.label))).toBe(false); expect(lowCard?.modelFields.some((field) => ["CLIP", "VAE"].includes(field.label))).toBe(false); }); it("builds LTX Director prompt boxes with global prompt and segment in/out values", () => { const timeline = { global_prompt: "golden glitter portrait, cinematic sunset light", segments: [ { id: "shot-a", start: 0, length: 50, prompt: "camera turns to the other side of the model face", type: "image", imageFile: "Krea2_turbo_00020_.png", }, { id: "shot-b", start: 50, end: 100, prompt: "the light starts shifting across her skin", type: "image", imageFile: "Krea2_turbo_00021_.png", }, ], }; const promptGraph = { "131": { class_type: "LTXDirector", inputs: { model: ["35", 0], clip: ["12", 0], frame_rate: 25, custom_width: 768, custom_height: 1344, duration_frames: 100, timeline_data: JSON.stringify(timeline), }, }, "35": { class_type: "UNETLoader", inputs: { unet_name: "ltx-2.3-22b-distilled.safetensors" } }, "12": { class_type: "DualCLIPLoader", inputs: { clip_name1: "gemma_3_12B_it.safetensors" } }, "17": { class_type: "CFGGuider", inputs: { cfg: 1, model: ["131", 0] } }, "19": { class_type: "SamplerCustomAdvanced", inputs: { noise: ["30", 0], guider: ["17", 0], sampler: ["20", 0], sigmas: ["21", 0], latent_image: ["18", 0], }, }, "20": { class_type: "KSamplerSelect", inputs: { sampler_name: "euler" } }, "21": { class_type: "BasicScheduler", inputs: { scheduler: "linear_quadratic", steps: 4, denoise: 0.42 } }, "30": { class_type: "RandomNoise", inputs: { noise_seed: 0 } }, }; const state = buildGenerationSectionState({ id: 10645, kind: "video", metadata_raw: { prompt: "D:/____COMFY_OUTPUTS/video/LTX_Director_00005_.mp4", raw_ffprobe: { format: { tags: { prompt: JSON.stringify(promptGraph), }, }, }, }, }); expect(state.kind).toBe("full"); expect(state.ltxDirector?.globalPrompt).toBe("golden glitter portrait, cinematic sunset light"); expect(state.ltxDirector?.fields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "FPS", value: "25" }), expect.objectContaining({ label: "Size", value: "768 x 1344" }), ]), ); expect(state.ltxDirector?.segments).toEqual([ expect.objectContaining({ label: "Segment 1", prompt: "camera turns to the other side of the model face", inLabel: "0s", outLabel: "2s", filename: "Krea2_turbo_00020_.png", isVideo: false, isAudio: false, previewCandidates: expect.arrayContaining([ expect.stringContaining("Krea2_turbo_00020_.png"), ]), }), expect.objectContaining({ label: "Segment 2", prompt: "the light starts shifting across her skin", inLabel: "2s", outLabel: "4s", filename: "Krea2_turbo_00021_.png", }), ]); expect(state.modelFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "UNet", value: "ltx-2.3-22b-distilled" }), expect.objectContaining({ label: "CLIP", value: "gemma_3_12B_it" }), ]), ); expect(state.samplingFields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Seed", value: 0 }), expect.objectContaining({ label: "Sampler", value: "euler" }), ]), ); }); it("builds Ideogram 4 prompt builder state from the JSON text encoder input", () => { const promptGraph = { "167": { class_type: "CLIPTextEncode", inputs: { text: ["185", 0], clip: ["168", 0] }, _meta: { title: "CLIP Text Encode (Positive Prompt)" }, }, "185": { class_type: "Ideogram4PromptBuilderKJ", inputs: { width: 768, height: 1376, high_level_description: "luxury skincare campaign with a graceful silhouette", background: "warm sunset gradient background", style: "photo", "style.photo": "DSLR photography", aesthetics: "premium beauty mood", lighting: "sunrise, sunset, warm", medium: "photography", elements_data: JSON.stringify([ { x: 0.29, y: 0.33, w: 0.13, h: 0.12, type: "obj", text: "", desc: "scientific graphic around the cheek", palette: ["#ffbf66"], }, ]), bg_brightness: 25, coord_mode: "normalized", bbox_order: "yx", }, _meta: { title: "Ideogram 4 Prompt Builder KJ" }, }, }; const state = buildGenerationSectionState({ id: 1, kind: "image", metadata_raw: { prompt: JSON.stringify(promptGraph) }, }); expect(state.kind).toBe("full"); expect(state.positivePrompt).toBe(""); expect(state.ideogram?.highLevelDescription).toBe("luxury skincare campaign with a graceful silhouette"); expect(state.ideogram?.background).toBe("warm sunset gradient background"); expect(state.ideogram?.json).toContain('"high_level_description": "luxury skincare campaign with a graceful silhouette"'); expect(state.ideogram?.json).toContain('"elements"'); expect(state.ideogram?.elements).toEqual([ expect.objectContaining({ label: "Element 1", description: "scientific graphic around the cheek", bbox: "0.29, 0.33, 0.13, 0.12", palette: ["#ffbf66"], }), ]); expect(state.ideogram?.fields).toEqual( expect.arrayContaining([ expect.objectContaining({ label: "Style", value: "photo" }), expect.objectContaining({ label: "Size", value: "768 x 1376" }), ]), ); }); });