Files
ComfyUI-Lora-Manager/vue-widgets/src/main.ts
T
Will Miao da71985c3e fix(autocomplete): strip lastAccepted boundary from exported workflows (#1093)
The hidden __lm_autocomplete_meta_* widget persisted lastAccepted
(insertedText/textSnapshot) into exported workflow JSON, leaking old
prompt text even after the user deleted it.

Patch app.graphToPrompt (shared by workflow export, Export API and
queueing) to strip lastAccepted from the serialized result's
widgets_values / widgets_values_named / output inputs. Only the
exported artifact is touched; live node state, undo snapshots,
copy/paste and local saves keep the boundary intact.
2026-09-03 08:29:55 +08:00

996 lines
30 KiB
TypeScript

import { createApp, type App as VueApp } from 'vue'
import PrimeVue from 'primevue/config'
import LoraPoolWidget from '@/components/LoraPoolWidget.vue'
import LoraRandomizerWidget from '@/components/LoraRandomizerWidget.vue'
import LoraCyclerWidget from '@/components/LoraCyclerWidget.vue'
import JsonDisplayWidget from '@/components/JsonDisplayWidget.vue'
import AutocompleteTextWidget from '@/components/AutocompleteTextWidget.vue'
import LoraInfoWidget from '@/components/LoraInfoWidget.vue'
import { createVueWidgetCleanup } from './vue-widget-cleanup'
import type { LoraPoolConfig, RandomizerConfig, CyclerConfig } from './composables/types'
import {
setupModeChangeHandler,
createModeChangeCallback,
LORA_CHAIN_NODE_TYPES
} from './mode-change-handler'
const LORA_POOL_WIDGET_MIN_WIDTH = 500
const LORA_POOL_WIDGET_MIN_HEIGHT = 520
const LORA_RANDOMIZER_WIDGET_MIN_WIDTH = 500
const LORA_RANDOMIZER_WIDGET_MIN_HEIGHT = 448
const LORA_RANDOMIZER_WIDGET_MAX_HEIGHT = LORA_RANDOMIZER_WIDGET_MIN_HEIGHT
const LORA_CYCLER_WIDGET_MIN_WIDTH = 380
const LORA_CYCLER_WIDGET_MIN_HEIGHT = 408
const LORA_CYCLER_WIDGET_MAX_HEIGHT = LORA_CYCLER_WIDGET_MIN_HEIGHT
const JSON_DISPLAY_WIDGET_MIN_WIDTH = 300
const JSON_DISPLAY_WIDGET_MIN_HEIGHT = 200
const LORA_INFO_WIDGET_MIN_WIDTH = 300
const LORA_INFO_WIDGET_MIN_HEIGHT = 200
const AUTOCOMPLETE_TEXT_WIDGET_MIN_HEIGHT = 60
const AUTOCOMPLETE_TEXT_WIDGET_MAX_HEIGHT = 100
// Per-modelType min size hints for node initial sizing.
// These are returned from the factory so ComfyUI's _initialMinSize mechanism
// gives the node a sensible default width (and height for prompt/embeddings).
const AUTOCOMPLETE_TEXT_MIN_WIDTH_DEFAULT = 400
const AUTOCOMPLETE_TEXT_MIN_HEIGHT_DEFAULT = 300
const AUTOCOMPLETE_METADATA_VERSION = 1
const LORA_MANAGER_WIDGET_IDS_PROPERTY = '__lm_widget_ids'
// Access LiteGraph global for Vue DOM mode detection (matches AutocompleteTextWidget.vue)
declare const LiteGraph: { vueNodesMode?: boolean } | undefined
// @ts-ignore - ComfyUI external module
import { app } from '../../../scripts/app.js'
// @ts-ignore - ComfyUI external module
import { api } from '../../../scripts/api.js'
// @ts-ignore
import { getPoolConfigFromConnectedNode, getActiveLorasFromNode, updateConnectedTriggerWords, updateDownstreamLoaders } from '../../web/comfyui/utils.js'
// @ts-ignore
import { stripAutocompleteMetadataFromPromptResult } from '../../web/comfyui/autocomplete.js'
// Strip the autocomplete lastAccepted boundary from exported workflows.
// lastAccepted carries old prompt text (insertedText/textSnapshot) and is
// session-only state; it must not leak into exported JSON (#1093).
// graphToPrompt is shared by workflow export, Export API and queueing.
// Local saves go through the change-tracker snapshot (no graphToPrompt) and
// are intentionally left untouched so cross-session caret continuity is kept.
// Post-processing the resolved result avoids any window/race with
// change-tracker or copy/paste serialization of live state.
const originalGraphToPrompt = app.graphToPrompt.bind(app)
app.graphToPrompt = async (...args: unknown[]) => {
const result = await originalGraphToPrompt(...args)
stripAutocompleteMetadataFromPromptResult(result)
return result
}
function forwardMiddleMouseToCanvas(container: HTMLElement) {
if (!container) return
container.addEventListener('pointerdown', (event) => {
if (event.button === 1) {
const canvas = app.canvas
if (canvas && typeof canvas.processMouseDown === 'function') {
canvas.processMouseDown(event)
}
}
})
container.addEventListener('pointermove', (event) => {
if ((event.buttons & 4) === 4) {
const canvas = app.canvas
if (canvas && typeof canvas.processMouseMove === 'function') {
canvas.processMouseMove(event)
}
}
})
container.addEventListener('pointerup', (event) => {
if (event.button === 1) {
const canvas = app.canvas
if (canvas && typeof canvas.processMouseUp === 'function') {
canvas.processMouseUp(event)
}
}
})
}
const vueApps = new Map<number | string, VueApp>()
let autocompleteTextWidgetInstanceId = 0
export function createAutocompleteTextWidgetInstanceId() {
autocompleteTextWidgetInstanceId += 1
return autocompleteTextWidgetInstanceId
}
// @ts-ignore
function createLoraPoolWidget(node) {
const container = document.createElement('div')
container.id = `lora-pool-widget-${node.id}`
container.style.width = '100%'
container.style.height = '100%'
container.style.display = 'flex'
container.style.flexDirection = 'column'
container.style.overflow = 'hidden'
forwardMiddleMouseToCanvas(container)
let internalValue: LoraPoolConfig | undefined
const widget = node.addDOMWidget(
'pool_config',
'LORA_POOL_CONFIG',
container,
{
getValue() {
return internalValue
},
setValue(v: LoraPoolConfig) {
internalValue = v
// ComfyUI automatically calls widget.callback after setValue
// No need for custom onSetValue mechanism
},
serialize: true,
// Per dev guide: providing getMinHeight via options allows the system to
// skip expensive DOM measurements during rendering loop, improving performance
getMinHeight() {
return LORA_POOL_WIDGET_MIN_HEIGHT
}
}
)
const vueApp = createApp(LoraPoolWidget, {
widget,
node
})
vueApp.use(PrimeVue, {
unstyled: true,
ripple: false
})
vueApp.mount(container)
vueApps.set(node.id, vueApp)
widget.computeLayoutSize = () => {
const minWidth = LORA_POOL_WIDGET_MIN_WIDTH
const minHeight = LORA_POOL_WIDGET_MIN_HEIGHT
return { minHeight, minWidth }
}
widget.onRemove = () => {
const vueApp = vueApps.get(node.id)
if (vueApp) {
vueApp.unmount()
vueApps.delete(node.id)
}
}
return { widget }
}
// @ts-ignore
function createLoraRandomizerWidget(node) {
const container = document.createElement('div')
container.id = `lora-randomizer-widget-${node.id}`
container.style.width = '100%'
container.style.height = '100%'
container.style.display = 'flex'
container.style.flexDirection = 'column'
container.style.overflow = 'hidden'
forwardMiddleMouseToCanvas(container)
// Initialize with default config to avoid sending undefined/empty string to backend
const defaultConfig: RandomizerConfig = {
count_mode: 'range',
count_fixed: 3,
count_min: 2,
count_max: 5,
model_strength_min: 0.0,
model_strength_max: 1.0,
use_same_clip_strength: true,
clip_strength_min: 0.0,
clip_strength_max: 1.0,
roll_mode: 'fixed',
use_recommended_strength: false,
recommended_strength_scale_min: 0.5,
recommended_strength_scale_max: 1.0,
}
let internalValue: RandomizerConfig = defaultConfig
const widget = node.addDOMWidget(
'randomizer_config',
'RANDOMIZER_CONFIG',
container,
{
getValue() {
return internalValue
},
setValue(v: RandomizerConfig) {
internalValue = v
// ComfyUI automatically calls widget.callback after setValue
// No need for custom onSetValue mechanism
},
serialize: true,
getMinHeight() {
return LORA_RANDOMIZER_WIDGET_MIN_HEIGHT
}
}
)
// Add method to get pool config from connected node
node.getPoolConfig = () => getPoolConfigFromConnectedNode(node)
// Handle roll event from Vue component
widget.onRoll = (randomLoras: any[]) => {
// Find the loras widget on this node and update it
const lorasWidget = node.widgets.find((w: any) => w.name === 'loras')
if (lorasWidget) {
lorasWidget.value = randomLoras
}
}
const vueApp = createApp(LoraRandomizerWidget, {
widget,
node,
api
})
vueApp.use(PrimeVue, {
unstyled: true,
ripple: false
})
vueApp.mount(container)
vueApps.set(node.id + 10000, vueApp) // Offset to avoid collision with pool widget
widget.computeLayoutSize = () => {
const minWidth = LORA_RANDOMIZER_WIDGET_MIN_WIDTH
const minHeight = LORA_RANDOMIZER_WIDGET_MIN_HEIGHT
const maxHeight = LORA_RANDOMIZER_WIDGET_MAX_HEIGHT
return { minHeight, minWidth, maxHeight }
}
widget.onRemove = () => {
const vueApp = vueApps.get(node.id + 10000)
if (vueApp) {
vueApp.unmount()
vueApps.delete(node.id + 10000)
}
}
return { widget }
}
// @ts-ignore
function createLoraCyclerWidget(node) {
const container = document.createElement('div')
container.id = `lora-cycler-widget-${node.id}`
container.style.width = '100%'
container.style.height = '100%'
container.style.display = 'flex'
container.style.flexDirection = 'column'
container.style.overflow = 'hidden'
forwardMiddleMouseToCanvas(container)
const defaultConfig: CyclerConfig = {
current_index: 1,
total_count: 0,
pool_config_hash: '',
model_strength: 1.0,
clip_strength: 1.0,
use_same_clip_strength: true,
use_preset_strength: false,
preset_strength_scale: 1.0,
sort_by: 'filename',
current_lora_name: '',
current_lora_filename: '',
repeat_count: 1,
repeat_used: 0,
is_paused: false,
include_no_lora: false,
}
let internalValue: CyclerConfig | undefined = defaultConfig
const widget = node.addDOMWidget(
'cycler_config',
'CYCLER_CONFIG',
container,
{
getValue() {
return internalValue
},
setValue(v: CyclerConfig) {
const oldFilename = internalValue?.current_lora_filename
internalValue = v
// ComfyUI automatically calls widget.callback after setValue
// No need for custom onSetValue mechanism
// Update downstream loaders when the active LoRA filename changes
if (oldFilename !== v?.current_lora_filename) {
updateDownstreamLoaders(node)
}
},
serialize: true,
getMinHeight() {
return LORA_CYCLER_WIDGET_MIN_HEIGHT
}
}
)
// Add method to get pool config from connected node
node.getPoolConfig = () => getPoolConfigFromConnectedNode(node)
const vueApp = createApp(LoraCyclerWidget, {
widget,
node,
api
})
vueApp.use(PrimeVue, {
unstyled: true,
ripple: false
})
vueApp.mount(container)
vueApps.set(node.id + 30000, vueApp) // Offset to avoid collision with other widgets
widget.computeLayoutSize = () => {
const minWidth = LORA_CYCLER_WIDGET_MIN_WIDTH
const minHeight = LORA_CYCLER_WIDGET_MIN_HEIGHT
const maxHeight = LORA_CYCLER_WIDGET_MAX_HEIGHT
return { minHeight, minWidth, maxHeight }
}
widget.onRemove = () => {
const vueApp = vueApps.get(node.id + 30000)
if (vueApp) {
vueApp.unmount()
vueApps.delete(node.id + 30000)
}
}
return { widget }
}
// @ts-ignore
function createJsonDisplayWidget(node) {
const container = document.createElement('div')
container.id = `json-display-widget-${node.id}`
container.style.width = '100%'
container.style.height = '100%'
container.style.display = 'flex'
container.style.flexDirection = 'column'
container.style.overflow = 'hidden'
forwardMiddleMouseToCanvas(container)
let internalValue: Record<string, unknown> | undefined
const widget = node.addDOMWidget(
'metadata',
'JSON_DISPLAY',
container,
{
getValue() {
return internalValue
},
setValue(v: Record<string, unknown>) {
internalValue = v
if (typeof widget.onSetValue === 'function') {
widget.onSetValue(v)
}
},
serialize: false, // Display-only widget - don't save metadata in workflows
getMinHeight() {
return JSON_DISPLAY_WIDGET_MIN_HEIGHT
}
}
)
const vueApp = createApp(JsonDisplayWidget, {
widget,
node
})
vueApp.use(PrimeVue, {
unstyled: true,
ripple: false
})
vueApp.mount(container)
vueApps.set(node.id + 20000, vueApp) // Offset to avoid collision with other widgets
widget.computeLayoutSize = () => {
const minWidth = JSON_DISPLAY_WIDGET_MIN_WIDTH
const minHeight = JSON_DISPLAY_WIDGET_MIN_HEIGHT
return { minHeight, minWidth }
}
widget.onRemove = () => {
const vueApp = vueApps.get(node.id + 20000)
if (vueApp) {
vueApp.unmount()
vueApps.delete(node.id + 20000)
}
}
return { widget }
}
const widgetInputOptions: Map<string, { placeholder?: string }> = new Map()
function getSerializableWidgetNames(node: any): string[] {
return (node.widgets || [])
.filter((widget: any) => widget && widget.serialize !== false)
.map((widget: any) => widget.name)
}
function createAutocompleteMetadataValue(textWidgetName = 'text') {
return {
version: AUTOCOMPLETE_METADATA_VERSION,
textWidgetName
}
}
function shouldBypassAutocompleteWidgetMigration(
node: any,
widgetValues: unknown[]
): boolean {
const inputDefs = node?.constructor?.nodeData?.inputs
if (!inputDefs || !Array.isArray(widgetValues)) {
return false
}
const widgetNames = new Set((node.widgets || []).map((widget: any) => widget?.name))
const hasAutocompleteMetadataWidget = Array.from(widgetNames).some((name) =>
typeof name === 'string' && name.startsWith('__lm_autocomplete_meta_')
)
if (!hasAutocompleteMetadataWidget) {
return false
}
const originalWidgetsInputs = Object.values(inputDefs).filter((input: any) =>
widgetNames.has(input.name) || input.forceInput
)
const widgetIndexHasForceInput = originalWidgetsInputs.flatMap((input: any) =>
input.control_after_generate
? [!!input.forceInput, false]
: [!!input.forceInput]
)
const result = (
widgetIndexHasForceInput.some(Boolean) &&
widgetIndexHasForceInput.length === widgetValues.length
)
return result
}
function remapWidgetValuesByName(
widgetValues: unknown[],
savedWidgetNames: string[],
currentWidgetNames: string[]
): unknown[] {
const valueByName = new Map<string, unknown>()
savedWidgetNames.forEach((name, index) => {
if (index < widgetValues.length) {
valueByName.set(name, widgetValues[index])
}
})
const currentWidgetNameSet = new Set(currentWidgetNames)
const remappedValues: unknown[] = []
for (const name of currentWidgetNames) {
if (valueByName.has(name)) {
remappedValues.push(valueByName.get(name))
}
}
// Append values for saved widget names that are NOT in the current widget
// list (e.g. forceInput widgets like "seed" that haven't been converted
// back to DOM widgets yet at configure time). Without these, the
// resulting array may accidentally match the length of ComfyUI's
// widgetIndexHasForceInput array, causing migrateWidgetsValues to
// incorrectly filter out the wrong values and drop real widget content.
for (const name of savedWidgetNames) {
if (!currentWidgetNameSet.has(name) && valueByName.has(name)) {
remappedValues.push(valueByName.get(name))
}
}
return remappedValues
}
function injectDefaultAutocompleteMetadataValues(
widgetValues: unknown[],
currentWidgetNames: string[]
): unknown[] {
const repairedValues: unknown[] = []
let legacyValueIndex = 0
for (const widgetName of currentWidgetNames) {
if (widgetName.startsWith('__lm_autocomplete_meta_')) {
const textWidgetName = widgetName.replace('__lm_autocomplete_meta_', '') || 'text'
repairedValues.push(createAutocompleteMetadataValue(textWidgetName))
continue
}
if (legacyValueIndex < widgetValues.length) {
repairedValues.push(widgetValues[legacyValueIndex])
legacyValueIndex++
}
}
return repairedValues
}
function normalizeAutocompleteWidgetValues(node: any, info: any) {
if (!info || !Array.isArray(info.widgets_values)) {
return
}
const currentWidgetNames = getSerializableWidgetNames(node)
if (currentWidgetNames.length === 0) {
return
}
const savedWidgetNames = info.properties?.[LORA_MANAGER_WIDGET_IDS_PROPERTY]
if (Array.isArray(savedWidgetNames) && savedWidgetNames.length > 0) {
const remappedValues = remapWidgetValuesByName(
info.widgets_values,
savedWidgetNames,
currentWidgetNames
)
info.widgets_values = remappedValues
return
}
const metadataWidgetCount = currentWidgetNames.filter((name) =>
name.startsWith('__lm_autocomplete_meta_')
).length
if (
metadataWidgetCount > 0 &&
info.widgets_values.length === currentWidgetNames.length - metadataWidgetCount
) {
const repairedValues = injectDefaultAutocompleteMetadataValues(
info.widgets_values,
currentWidgetNames
)
info.widgets_values = repairedValues
}
}
function applyAutocompleteTextLayoutFix(
widget: any,
_container: HTMLElement | undefined,
isVueMode: boolean
): void {
// In Vue rendering mode the WidgetDOM wrapper handles sizing, so we
// only provide a computeSize hint and leave the container unconstrained.
// In canvas mode we clear all custom sizing so LiteGraph's default
// widget-area layout takes over. Neither path sets a hard max-height;
// the textarea can grow freely (e.g. in app mode where
// [&_textarea]:resize-y applies).
if (isVueMode) {
;(widget as any).computeLayoutSize = undefined
widget.computeSize = (width?: number) =>
[width ?? 200, AUTOCOMPLETE_TEXT_WIDGET_MAX_HEIGHT - 4]
} else {
delete (widget as any).computeLayoutSize
delete (widget as any).computeSize
}
}
// Listen for Vue DOM mode setting changes and dispatch custom event
const initVueDomModeListener = () => {
if (app.ui?.settings?.addEventListener) {
app.ui.settings.addEventListener('Comfy.VueNodes.Enabled.change', () => {
// Use requestAnimationFrame to ensure the setting value has been updated
// before we read it (the event may fire before internal state updates)
requestAnimationFrame(() => {
const isVueDomMode = app.ui?.settings?.getSettingValue?.('Comfy.VueNodes.Enabled') ?? false
if (app.graph?.nodes) {
for (const node of app.graph.nodes) {
const textWidget = node.widgets?.find(
(w: any) => w.type === 'AUTOCOMPLETE_TEXT_LORAS'
)
if (!textWidget) continue
const container = (textWidget as any).element as HTMLElement | undefined
applyAutocompleteTextLayoutFix(textWidget, container, isVueDomMode)
}
}
requestAnimationFrame(() => {
for (const nodeEl of document.querySelectorAll('[data-node-id]')) {
const grid = nodeEl.querySelector('[data-testid="node-widgets"]') as HTMLElement | null
if (!grid) continue
const nodeId = nodeEl.getAttribute('data-node-id')
const node = app.graph?.getNodeById(nodeId as any)
if (!node) continue
const rows: string[] = []
let needsFix = false
for (const w of node.widgets ?? []) {
if (w.type === 'LORA_MANAGER_AUTOCOMPLETE_METADATA') {
rows.push('min-content')
} else if (w.name === 'loras') {
rows.push('auto')
} else if (w.name === 'text' && w.type === 'AUTOCOMPLETE_TEXT_LORAS') {
rows.push(isVueDomMode ? 'min-content' : 'auto')
needsFix = true
} else {
rows.push('auto')
}
}
if (needsFix) {
grid.style.gridTemplateRows = rows.join(' ')
}
}
})
app.canvas?.setDirty(true, true)
document.dispatchEvent(new CustomEvent('lora-manager:vue-mode-change', {
detail: { isVueDomMode }
}))
})
})
}
}
// Initialize listener when app is ready
if (app.ui?.settings) {
initVueDomModeListener()
} else {
// Defer until app is ready
const checkAppReady = setInterval(() => {
if (app.ui?.settings) {
initVueDomModeListener()
clearInterval(checkAppReady)
}
}, 100)
}
// @ts-ignore
function createLoraInfoWidget(node: any) {
const container = document.createElement('div')
container.id = `lora-info-widget-${node.id}`
container.style.width = '100%'
container.style.height = '100%'
container.style.display = 'flex'
container.style.flexDirection = 'column'
container.style.overflow = 'hidden'
forwardMiddleMouseToCanvas(container)
let internalValue: { name?: string; notes?: string; filePath?: string; activeTab?: string } | undefined
const widget = node.addDOMWidget(
'lora_info_display',
'LORA_INFO_DISPLAY',
container,
{
getValue() {
return internalValue
},
setValue(v: { name?: string; notes?: string; filePath?: string; activeTab?: string }) {
internalValue = v
if (typeof widget.onSetValue === 'function') {
widget.onSetValue(v)
}
},
serialize: true,
getMinHeight() {
return LORA_INFO_WIDGET_MIN_HEIGHT
}
}
)
const vueApp = createApp(LoraInfoWidget, {
widget,
node,
api,
app,
isVueMode: typeof LiteGraph !== 'undefined' && LiteGraph.vueNodesMode,
})
vueApp.use(PrimeVue, {
unstyled: true,
ripple: false
})
vueApp.mount(container)
vueApps.set(node.id + 40000, vueApp) // Offset to avoid collision
widget.computeLayoutSize = () => {
const minWidth = LORA_INFO_WIDGET_MIN_WIDTH
const minHeight = LORA_INFO_WIDGET_MIN_HEIGHT
return { minHeight, minWidth }
}
widget.onRemove = () => {
const vueApp = vueApps.get(node.id + 40000)
if (vueApp) {
vueApp.unmount()
vueApps.delete(node.id + 40000)
}
}
return { widget }
}
// Factory function for creating autocomplete text widgets
// @ts-ignore
function createAutocompleteTextWidgetFactory(
node: any,
widgetName: string,
modelType: 'loras' | 'prompt',
inputOptions: { placeholder?: string } = {}
) {
const metadataWidgetName = `__lm_autocomplete_meta_${widgetName}`
let container: HTMLElement | null = null
const existingContainers = document.querySelectorAll<HTMLElement>(
'[id^="autocomplete-text-widget-"]'
)
for (const el of existingContainers) {
if (el.children.length === 0) {
container = el
break
}
}
if (!container) {
const instanceId = String(createAutocompleteTextWidgetInstanceId())
container = document.createElement('div')
container.id = `autocomplete-text-widget-${instanceId}`
container.style.width = '100%'
container.style.height = '100%'
container.style.display = 'flex'
container.style.flexDirection = 'column'
container.style.overflow = 'hidden'
forwardMiddleMouseToCanvas(container)
}
// Store textarea reference on the container element so cloned widgets can access it
// This is necessary because when widgets are promoted to subgraph nodes,
// the cloned widget shares the same element but needs access to inputEl
const widgetElementRef = { inputEl: undefined as HTMLTextAreaElement | undefined }
;(container as any).__widgetInputEl = widgetElementRef
const metadataWidget = node.addWidget('text', metadataWidgetName, {
version: AUTOCOMPLETE_METADATA_VERSION,
textWidgetName: widgetName
})
metadataWidget.value = createAutocompleteMetadataValue(widgetName)
metadataWidget.type = 'LORA_MANAGER_AUTOCOMPLETE_METADATA'
metadataWidget.hidden = true
metadataWidget.computeSize = () => [0, -4]
metadataWidget.serializeValue = () => metadataWidget.value
const widget = node.addDOMWidget(
widgetName,
`AUTOCOMPLETE_TEXT_${modelType.toUpperCase()}`,
container,
{
getValue() {
// Access inputEl from widget or from the shared element reference
const inputEl = widget.inputEl ?? (container as any).__widgetInputEl?.inputEl
return inputEl?.value ?? ''
},
setValue(v: string) {
// Access inputEl from widget or from the shared element reference
const inputEl = widget.inputEl ?? (container as any).__widgetInputEl?.inputEl
if (inputEl) {
inputEl.value = v ?? ''
// Notify Vue component of value change via custom event
inputEl.dispatchEvent(new CustomEvent('lora-manager:autocomplete-value-changed', {
detail: { value: v ?? '' }
}))
} else {
;(widget as any)._pendingValue = v ?? ''
}
// Also call onSetValue if defined (for Vue component integration)
if (typeof widget.onSetValue === 'function') {
widget.onSetValue(v ?? '')
}
},
serialize: true,
getMinHeight() {
return AUTOCOMPLETE_TEXT_WIDGET_MIN_HEIGHT
},
...(modelType === 'loras' && {
getMaxHeight() {
return AUTOCOMPLETE_TEXT_WIDGET_MAX_HEIGHT
}
})
}
)
widget.metadataWidget = metadataWidget
// Get spellcheck setting from ComfyUI settings (default: false)
const spellcheck = app.ui?.settings?.getSettingValue?.('Comfy.TextareaWidget.Spellcheck') ?? false
const maxHeight = modelType === 'loras' ? AUTOCOMPLETE_TEXT_WIDGET_MAX_HEIGHT : undefined
const vueApp = createApp(AutocompleteTextWidget, {
widget,
node,
modelType,
placeholder: inputOptions.placeholder || widgetName,
showPreview: true,
spellcheck,
maxHeight
})
vueApp.use(PrimeVue, {
unstyled: true,
ripple: false
})
vueApp.mount(container)
const appKey = container.id
vueApps.set(appKey, vueApp)
if (maxHeight) {
container.style.minHeight = `${AUTOCOMPLETE_TEXT_WIDGET_MIN_HEIGHT}px`
}
if (modelType === 'loras') {
applyAutocompleteTextLayoutFix(
widget,
container,
typeof LiteGraph !== 'undefined' && LiteGraph.vueNodesMode === true
)
}
const vueCleanup = createVueWidgetCleanup(vueApp, () => {
vueApps.delete(appKey)
})
widget.onRemove = () => {
vueCleanup()
}
// Return minWidth/minHeight hints so ComfyUI's _initialMinSize mechanism
// sets a sensible initial node width (and height for prompt/embeddings).
// loras modelType retains its existing height constraints (getMaxHeight: 100).
const minWidth = AUTOCOMPLETE_TEXT_MIN_WIDTH_DEFAULT
const minHeight = modelType === 'loras' ? undefined : AUTOCOMPLETE_TEXT_MIN_HEIGHT_DEFAULT
return { widget, minWidth, minHeight }
}
app.registerExtension({
name: 'LoraManager.VueWidgets',
getCustomWidgets() {
return {
// @ts-ignore
LORA_POOL_CONFIG(node) {
return createLoraPoolWidget(node)
},
// @ts-ignore
RANDOMIZER_CONFIG(node) {
return createLoraRandomizerWidget(node)
},
// @ts-ignore
CYCLER_CONFIG(node) {
return createLoraCyclerWidget(node)
},
// Autocomplete text widget for LoRAs (used by Lora Loader, Lora Stacker, WanVideo Lora Select)
// @ts-ignore
AUTOCOMPLETE_TEXT_LORAS(node) {
const options = widgetInputOptions.get(`${node.comfyClass}:text`) || {}
return createAutocompleteTextWidgetFactory(node, 'text', 'loras', options)
},
// Autocomplete text widget for prompt (used by Prompt and Text nodes)
// @ts-ignore
AUTOCOMPLETE_TEXT_PROMPT(node) {
const options = widgetInputOptions.get(`${node.comfyClass}:text`) || {}
return createAutocompleteTextWidgetFactory(node, 'text', 'prompt', options)
},
}
},
// Add display-only widget to Debug Metadata node
// Register mode change handlers for LoRA provider nodes
// Extract and store input options for autocomplete widgets
// @ts-ignore
async beforeRegisterNodeDef(nodeType, nodeData) {
const comfyClass = nodeType.comfyClass
const inputs = { ...nodeData.input?.required, ...nodeData.input?.optional }
let hasAutocompleteWidget = false
// Extract and store input options for autocomplete widgets
for (const [inputName, inputDef] of Object.entries(inputs)) {
// @ts-ignore
if (Array.isArray(inputDef) && typeof inputDef[0] === 'string' && inputDef[0].startsWith('AUTOCOMPLETE_TEXT_')) {
// @ts-ignore
const options = inputDef[1] || {}
widgetInputOptions.set(`${nodeData.name}:${inputName}`, options)
hasAutocompleteWidget = true
}
}
if (hasAutocompleteWidget) {
const originalOnSerialize = nodeType.prototype.onSerialize
const originalConfigure = nodeType.prototype.configure
nodeType.prototype.onSerialize = function (serialized: any) {
originalOnSerialize?.apply(this, arguments)
serialized.properties = serialized.properties || {}
const widgetIds = getSerializableWidgetNames(this)
serialized.properties[LORA_MANAGER_WIDGET_IDS_PROPERTY] = widgetIds
}
nodeType.prototype.configure = function (info: any) {
normalizeAutocompleteWidgetValues(this, info)
const bypassResult = shouldBypassAutocompleteWidgetMigration(this, info?.widgets_values ?? [])
if (bypassResult) {
info.widgets_values = [...(info.widgets_values ?? []), null]
}
return originalConfigure?.apply(this, arguments)
}
}
// Register mode change handlers for LORA_STACK chain nodes
if (LORA_CHAIN_NODE_TYPES.includes(comfyClass)) {
const originalOnNodeCreated = nodeType.prototype.onNodeCreated
nodeType.prototype.onNodeCreated = function () {
originalOnNodeCreated?.apply(this, arguments)
// Create node-specific callback for Lora Stacker (updates direct trigger toggles)
const nodeSpecificCallback = comfyClass === "Lora Stacker (LoraManager)"
? (activeLoraNames: Set<string>) => updateConnectedTriggerWords(this, activeLoraNames)
: undefined
// Create and set up the mode change handler
const onModeChange = createModeChangeCallback(this, updateDownstreamLoaders, nodeSpecificCallback)
setupModeChangeHandler(this, onModeChange)
}
}
// Add the JSON display widget to Debug Metadata node
if (nodeData.name === 'Debug Metadata (LoraManager)') {
const onNodeCreated = nodeType.prototype.onNodeCreated
nodeType.prototype.onNodeCreated = function () {
onNodeCreated?.apply(this, [])
// Add the JSON display widget
createJsonDisplayWidget(this)
}
}
// Add the Lora Info display widget
if (nodeData.name === 'Lora Info (LoraManager)') {
const onNodeCreated = nodeType.prototype.onNodeCreated
nodeType.prototype.onNodeCreated = function () {
onNodeCreated?.apply(this, [])
// Create the lora info display widget
createLoraInfoWidget(this)
}
}
}
})