Files
ComfyUI_frontend/src/components/rightSidePanel/errors/useErrorGroups.ts
jaeone94 74147d7ee2 feat: lift boundary-exposed validation errors to the subgraph host (#13542)
## Summary

Backend validation errors (`node_errors`) are reported against the
**flattened** prompt, so an error whose real fix lives on a subgraph
host node gets attached to an interior node the user may never open —
and, after ADR 0009, often *cannot* meaningfully fix there. This PR
re-surfaces a validation error onto the subgraph host node **when — and
only when — the error's subject (the specific input/widget named by
`extra_info.input_name`) is exposed through the subgraph boundary**.

## Why this is needed

Two concrete situations motivated this, both observed in real workflows:

1. **Broken link at the host.** Root node A should feed subgraph host B,
whose boundary input is linked to interior node C. If the A→B link is
missing, the backend flattens the prompt, sees C with no resolved input,
and raises `required_input_missing` **on C** (`"B:C"`). The actual fix —
connect B's input — is one level up, on a node the error never points
at.
2. **Host-owned widget values.** Per ADR 0009 (subgraph promoted widgets
use linked inputs), a promoted widget's value is owned by the host
`SubgraphNode`; the interior widget only supplies schema/defaults. When
the backend raises `value_not_in_list` (or min/max violations) for that
value, attributing it to the interior node is factually wrong — the
value that failed validation *is the host's value*.

This continues the direction of #13059, which moved **missing-model**
detection identity to `{hostExecutionId, hostWidgetName}` with the
interior path kept as diagnostics. That was possible in the FE pre-scan;
this PR applies the same ownership principle to **backend-received**
errors via a receive-side mapping, since the backend cannot know about
subgraph boundaries in a flattened prompt.

## The rule (design)

> Lift an error from interior node N to host H **iff** N's input slot
named by the error is linked to the containing subgraph's boundary
(`SubgraphInput`). Apply the same test again at H (boundary-by-boundary,
matching ADR 0009's chaining principle) and stop at the first level
where the subject is no longer boundary-linked — that node is where the
user can actually fix it.

The predicate is **structural (boundary exposure), not data-flow**.

### In scope — examples

- `required_input_missing` on interior `"12:5"` whose input is fed by
the boundary → surfaces on host `12`'s input slot (red slot ring on the
host, errors-tab card titled/located at the host, message names the
host's `SubgraphInput.name`).
- `value_not_in_list` / `value_smaller_than_min` /
`value_bigger_than_max` on a promoted interior widget → surfaces on the
host's promoted widget. Nested hosts chain: `"1:2:3"` lifts to `"1:2"`,
and further to `"1"` only if `1:2`'s own slot is boundary-linked too.
- Clearing follows the surface: connecting the highlighted host input or
fixing the host widget clears the underlying interior (raw) error —
range-guarded per target, so a still-out-of-range host value does
**not** clear.

### Out of scope — examples

- **No value-flow ancestry.** All in the root graph: A's widget links to
B, B's to C, and C rejects the value that originated at A → the error
**stays on C**. Following same-graph links to a "root cause" node is
explicitly not this feature.
- Errors without an `input_name` subject, node-level types
(`exception_during_validation`, `dependency_cycle`, image-not-loaded),
and unknown validation types — never lifted. Unknown types stay
node-scoped to match how the error catalog renders them (the shared
`isNodeLevelValidationError` in `executionErrorUtil` encodes this, and
the catalog derives its node-level rules from the same set).
- Runtime execution errors (exceptions during a run) — validation
responses only.
- Interior errors whose input is fed by another interior node — fixable
in place, stay in place.
- Fan-out display dedupe: when one boundary input feeds multiple
interior nodes and the host slot is unconnected, each interior error
lifts to the same host slot as a separate panel line. A single fix
(connecting the host input) clears all of them — the clearing
translation already fans out — so the duplication is cosmetic.
Display-level dedupe is a follow-up; deduping inside the lift would
break the one-source-per-error clearing contract.
- Reactive re-lifting on graph topology edits while errors are displayed
(invariant documented on the computed; follow-up), and deriving the
error catalog's full validation rule table from the shared
classification (follow-up; the node-level type set and the
image-not-loaded predicate are already single-sourced in
`executionErrorUtil` and consumed by both the lift and the catalog).

## Changes

- **What**: New pure module
`core/graph/subgraph/liftNodeErrorsToBoundary.ts` — per-error, fail-open
record transform (unresolvable ids/slots/links leave the error where the
backend put it; raw payload is never mutated). `executionErrorStore`
derives `surfacedNodeErrors` from it and display consumers switch over
(errors tab grouping, canvas node/slot flags, Vue node badges,
app-mode/linear hints); raw `lastNodeErrors` remains the source of truth
for mutation. Host-side clearing translates through the lift's
diagnostics fields (`source_execution_id` / `source_input_name`) with a
per-target range guard.
- **Breaking**: None. No persistence/serialization changes; interior
identity survives as diagnostics metadata only (ADR 0009 language).

## Review Focus

- The lift predicate lives entirely on link topology
(`LLink.originIsIoNode` → `SubgraphInput`) — no
`proxyWidgets`/promotion-store style source authority is reintroduced.
- `clearSlotErrorsWithRangeCheck` now resolves clear targets first and
range-checks each target's raw errors; the lifted-path twin of the
existing range-retention test pins this.
- `useProcessedWidgets` deliberately stays on the raw record: host
promoted widgets already map to interior errors via
`widget.sourceExecutionId`, so an interior widget keeps its local red
hint when the user opens the subgraph (hint layer vs surface layer).
- Unit coverage is carried by the pure module (real litegraph subgraph
fixtures, incl. nested recursion, promoted widgets via
`promoteValueWidgetViaSubgraphInput`, ordering, fail-open/no-mutation);
one e2e pins the user-visible contract (host ring + host slot dot +
interior clean).

## Screenshots

### Before 


https://github.com/user-attachments/assets/81e5c4db-515d-4f1f-8f8a-e07ac490510f

### After



https://github.com/user-attachments/assets/2949da06-a049-41c1-a480-98ee28333bf2
2026-07-14 14:39:20 +00:00

917 lines
27 KiB
TypeScript

import { computed, reactive, ref, toValue, watch } from 'vue'
import type { MaybeRefOrGetter } from 'vue'
import Fuse from 'fuse.js'
import type { IFuseOptions } from 'fuse.js'
import { useMissingModelStore } from '@/platform/missingModel/missingModelStore'
import { useMissingMediaStore } from '@/platform/missingMedia/missingMediaStore'
import { useExecutionErrorStore } from '@/stores/executionErrorStore'
import { useMissingNodesErrorStore } from '@/platform/nodeReplacement/missingNodesErrorStore'
import { useComfyRegistryStore } from '@/stores/comfyRegistryStore'
import { useCanvasStore } from '@/renderer/core/canvas/canvasStore'
import { app } from '@/scripts/app'
import { isCloud } from '@/platform/distribution/types'
import { SubgraphNode } from '@/lib/litegraph/src/litegraph'
import type { LGraphNode } from '@/lib/litegraph/src/litegraph'
import {
getNodeByExecutionId,
getExecutionIdByNode
} from '@/utils/graphTraversalUtil'
import { resolveNodeDisplayName } from '@/utils/nodeTitleUtil'
import { isLGraphNode } from '@/utils/litegraphUtil'
import { st } from '@/i18n'
import type { MissingNodeType } from '@/types/comfy'
import type { ErrorCardData, ErrorGroup, ErrorItem } from './types'
import { shouldRenderExecutionItemList } from './executionItemList'
import { someNodeTypeInSelection } from './selectionEmphasis'
import type { NodeExecutionId } from '@/types/nodeIdentification'
import type { MissingModelGroup } from '@/platform/missingModel/types'
import type { ResolvedCatalogErrorMessage } from '@/platform/errorCatalog/types'
import type { MissingMediaGroup } from '@/platform/missingMedia/types'
import {
countMissingModels,
groupMissingModelCandidates
} from '@/platform/missingModel/missingModelGrouping'
import { groupCandidatesByMediaType } from '@/platform/missingMedia/missingMediaScan'
import { countMissingMediaReferences } from '@/platform/missingMedia/missingMediaGrouping'
import {
resolveMissingErrorMessage,
resolveRunErrorMessage
} from '@/platform/errorCatalog/errorMessageResolver'
import {
compareExecutionId,
tryNormalizeNodeExecutionId
} from '@/types/nodeIdentification'
const PROMPT_CARD_ID = '__prompt__'
/** Sentinel: distinguishes "fetch in-flight" from "fetch done, pack not found (null)". */
const RESOLVING = '__RESOLVING__'
export interface MissingPackGroup {
packId: string | null
nodeTypes: MissingNodeType[]
isResolving: boolean
}
export interface SwapNodeGroup {
type: string
newNodeId: string | undefined
nodeTypes: MissingNodeType[]
}
interface GroupEntry {
type: 'execution'
displayTitle: string
displayMessage?: string
priority: number
cards: Map<string, ErrorCardData>
}
interface ErrorSearchItem {
groupIndex: number
cardIndex: number
searchableNodeId: string
searchableNodeTitle: string
searchableRawMessage: string
searchableRawDetails: string
searchableMessage: string
searchableDetails: string
}
type CataloguedErrorItem = ErrorItem & ResolvedCatalogErrorMessage
/** Resolve display info for a node by its execution ID. */
function resolveNodeInfo(nodeId: NodeExecutionId) {
const graphNode = getNodeByExecutionId(app.rootGraph, nodeId)
return {
title: resolveNodeDisplayName(graphNode, {
emptyLabel: '',
untitledLabel: '',
st
}),
graphNodeId: graphNode ? String(graphNode.id) : undefined
}
}
function getOrCreateGroup(
groupsMap: Map<string, GroupEntry>,
groupKey: string,
displayTitle = groupKey,
priority = 1,
displayMessage?: string
): Map<string, ErrorCardData> {
let entry = groupsMap.get(groupKey)
if (!entry) {
entry = {
type: 'execution',
displayTitle,
displayMessage,
priority,
cards: new Map()
}
groupsMap.set(groupKey, entry)
} else if (!entry.displayMessage && displayMessage) {
entry.displayMessage = displayMessage
}
return entry.cards
}
function createErrorCard(
nodeId: NodeExecutionId,
classType: string,
idPrefix: string
): ErrorCardData {
const nodeInfo = resolveNodeInfo(nodeId)
return {
id: `${idPrefix}-${nodeId}`,
title: classType,
nodeId,
nodeTitle: nodeInfo.title,
graphNodeId: nodeInfo.graphNodeId,
errors: []
}
}
function compareNodeId(a: ErrorCardData, b: ErrorCardData): number {
return compareExecutionId(a.nodeId, b.nodeId)
}
function countExecutionCards(cards: ErrorCardData[]): number {
if (shouldRenderExecutionItemList(cards)) {
return cards.reduce((count, card) => count + card.errors.length, 0)
}
return cards.length
}
function toSortedGroups(groupsMap: Map<string, GroupEntry>): ErrorGroup[] {
return Array.from(groupsMap.entries())
.map(([rawGroupKey, groupData]) => {
const cards = Array.from(groupData.cards.values()).sort(compareNodeId)
return {
type: 'execution' as const,
groupKey: `execution:${rawGroupKey}`,
displayTitle: groupData.displayTitle,
displayMessage: groupData.displayMessage,
count: countExecutionCards(cards),
cards,
priority: groupData.priority
}
})
.sort((a, b) => {
if (a.priority !== b.priority) return a.priority - b.priority
return a.displayTitle.localeCompare(b.displayTitle)
})
}
function searchErrorGroups(groups: ErrorGroup[], query: string) {
if (!query) return groups
const searchableList: ErrorSearchItem[] = []
for (let gi = 0; gi < groups.length; gi++) {
const group = groups[gi]
if (group.type !== 'execution') continue
for (let ci = 0; ci < group.cards.length; ci++) {
const card = group.cards[ci]
searchableList.push({
groupIndex: gi,
cardIndex: ci,
searchableNodeId: card.nodeId ?? '',
searchableNodeTitle: card.nodeTitle ?? '',
searchableRawMessage: card.errors.map((e) => e.message).join(' '),
searchableRawDetails: card.errors.map((e) => e.details).join(' '),
searchableMessage: card.errors
.map((e) =>
[e.displayTitle, e.displayMessage, e.message]
.filter(Boolean)
.join(' ')
)
.join(' '),
searchableDetails: card.errors
.map((e) => [e.displayDetails, e.details].filter(Boolean).join(' '))
.join(' ')
})
}
}
const fuseOptions: IFuseOptions<ErrorSearchItem> = {
keys: [
{ name: 'searchableRawMessage', weight: 0.3 },
{ name: 'searchableNodeId', weight: 0.2 },
{ name: 'searchableNodeTitle', weight: 0.2 },
{ name: 'searchableMessage', weight: 0.2 },
{ name: 'searchableRawDetails', weight: 0.1 },
{ name: 'searchableDetails', weight: 0.1 }
],
threshold: 0.3
}
const fuse = new Fuse(searchableList, fuseOptions)
const results = fuse.search(query)
const matchedCardKeys = new Set(
results.map((r) => `${r.item.groupIndex}:${r.item.cardIndex}`)
)
return groups
.map((group, gi) => {
if (group.type !== 'execution') return group
const cards = group.cards.filter((_: ErrorCardData, ci: number) =>
matchedCardKeys.has(`${gi}:${ci}`)
)
return {
...group,
cards,
count: countExecutionCards(cards)
}
})
.filter((group) => group.type !== 'execution' || group.cards.length > 0)
}
export function useErrorGroups(searchQuery: MaybeRefOrGetter<string>) {
const executionErrorStore = useExecutionErrorStore()
const missingNodesStore = useMissingNodesErrorStore()
const missingModelStore = useMissingModelStore()
const missingMediaStore = useMissingMediaStore()
const canvasStore = useCanvasStore()
const { inferPackFromNodeName } = useComfyRegistryStore()
const collapseState = reactive<Record<string, boolean>>({})
const selectedNodeInfo = computed(() => {
const items = canvasStore.selectedItems
const nodeIds = new Set<string>()
const containerExecutionIds = new Set<NodeExecutionId>()
for (const item of items) {
if (!isLGraphNode(item)) continue
nodeIds.add(String(item.id))
if (item instanceof SubgraphNode && app.rootGraph) {
const execId = getExecutionIdByNode(app.rootGraph, item)
if (execId) containerExecutionIds.add(execId)
}
}
return {
nodeIds: nodeIds.size > 0 ? nodeIds : null,
containerExecutionIds
}
})
const hasSelection = computed(() => selectedNodeInfo.value.nodeIds !== null)
const selectedNodeCount = computed(
() => selectedNodeInfo.value.nodeIds?.size ?? 0
)
const selectedNodeTitle = computed(() => {
if (selectedNodeCount.value !== 1) return null
const node = canvasStore.selectedItems.find(isLGraphNode)
if (!node) return null
return (
resolveNodeDisplayName(node, {
emptyLabel: '',
untitledLabel: '',
st
}) || null
)
})
const errorNodeCache = computed(() => {
const map = new Map<string, LGraphNode>()
for (const execId of executionErrorStore.allErrorExecutionIds) {
const node = getNodeByExecutionId(app.rootGraph, execId)
if (node) map.set(execId, node)
}
return map
})
const missingNodeCache = computed(() => {
const map = new Map<string, LGraphNode>()
const nodeTypes = missingNodesStore.missingNodesError?.nodeTypes ?? []
for (const nodeType of nodeTypes) {
if (typeof nodeType === 'string') continue
if (nodeType.nodeId == null) continue
const nodeId = String(nodeType.nodeId)
const node = getNodeByExecutionId(app.rootGraph, nodeId)
if (node) map.set(nodeId, node)
}
return map
})
function isErrorInSelection(executionNodeId: NodeExecutionId): boolean {
const nodeIds = selectedNodeInfo.value.nodeIds
if (!nodeIds) return true
const graphNode = errorNodeCache.value.get(executionNodeId)
if (graphNode && nodeIds.has(String(graphNode.id))) return true
for (const containerExecId of selectedNodeInfo.value
.containerExecutionIds) {
if (executionNodeId.startsWith(`${containerExecId}:`)) return true
}
return false
}
function addNodeErrorToGroup(
groupsMap: Map<string, GroupEntry>,
nodeId: NodeExecutionId,
classType: string,
idPrefix: string,
error: CataloguedErrorItem,
filterBySelection = false
) {
if (filterBySelection && !isErrorInSelection(nodeId)) return
const cards = getOrCreateGroup(
groupsMap,
error.catalogId,
error.displayTitle ?? classType,
1,
error.displayMessage
)
if (!cards.has(nodeId)) {
cards.set(nodeId, createErrorCard(nodeId, classType, idPrefix))
}
const card = cards.get(nodeId)
if (!card) return
card.errors.push(error)
}
function processPromptError(
groupsMap: Map<string, GroupEntry>,
filterBySelection = false
) {
if (
(filterBySelection && selectedNodeInfo.value.nodeIds) ||
!executionErrorStore.lastPromptError
)
return
const error = executionErrorStore.lastPromptError
const resolvedDisplay = resolveRunErrorMessage({
kind: 'prompt',
error,
isCloud
})
const groupDisplayTitle = resolvedDisplay.displayTitle ?? error.message
const cards = getOrCreateGroup(
groupsMap,
`prompt:${error.type}`,
groupDisplayTitle,
0,
resolvedDisplay.displayMessage
)
// Prompt errors are not tied to a node, so they bypass addNodeErrorToGroup.
cards.set(PROMPT_CARD_ID, {
id: PROMPT_CARD_ID,
title: groupDisplayTitle,
errors: [
{
message: error.message,
...resolvedDisplay
}
]
})
}
function processNodeErrors(
groupsMap: Map<string, GroupEntry>,
filterBySelection = false
) {
if (!executionErrorStore.surfacedNodeErrors) return
for (const [rawNodeId, nodeError] of Object.entries(
executionErrorStore.surfacedNodeErrors
)) {
const nodeId = tryNormalizeNodeExecutionId(rawNodeId)
if (!nodeId) continue
const nodeDisplayName =
resolveNodeInfo(nodeId).title || nodeError.class_type
for (const e of nodeError.errors) {
addNodeErrorToGroup(
groupsMap,
nodeId,
nodeError.class_type,
'node',
{
message: e.message,
details: e.details ?? undefined,
...resolveRunErrorMessage({
kind: 'node_validation',
error: e,
nodeDisplayName
})
},
filterBySelection
)
}
}
}
function processExecutionError(
groupsMap: Map<string, GroupEntry>,
filterBySelection = false
) {
if (!executionErrorStore.lastExecutionError) return
const e = executionErrorStore.lastExecutionError
const nodeId = tryNormalizeNodeExecutionId(e.node_id)
if (!nodeId) return
addNodeErrorToGroup(
groupsMap,
nodeId,
e.node_type,
'exec',
{
message: `${e.exception_type}: ${e.exception_message}`,
details: e.traceback.join('\n'),
isRuntimeError: true,
exceptionType: e.exception_type,
...resolveRunErrorMessage({
kind: 'execution',
error: e,
nodeDisplayName: resolveNodeInfo(nodeId).title || e.node_type
})
},
filterBySelection
)
}
// Async pack-ID resolution for missing node types that lack a cnrId
const asyncResolvedIds = ref<Map<string, string | null>>(new Map())
const pendingTypes = computed(() =>
(missingNodesStore.missingNodesError?.nodeTypes ?? []).filter(
(n): n is Exclude<MissingNodeType, string> =>
typeof n !== 'string' && !n.cnrId
)
)
watch(
pendingTypes,
async (pending, _, onCleanup) => {
const toResolve = pending.filter(
(n) => asyncResolvedIds.value.get(n.type) === undefined
)
if (!toResolve.length) return
const resolvingTypes = toResolve.map((n) => n.type)
let cancelled = false
onCleanup(() => {
cancelled = true
const next = new Map(asyncResolvedIds.value)
for (const type of resolvingTypes) {
if (next.get(type) === RESOLVING) next.delete(type)
}
asyncResolvedIds.value = next
})
const updated = new Map(asyncResolvedIds.value)
for (const type of resolvingTypes) updated.set(type, RESOLVING)
asyncResolvedIds.value = updated
const results = await Promise.allSettled(
toResolve.map(async (n) => ({
type: n.type,
packId: (await inferPackFromNodeName.call(n.type))?.id ?? null
}))
)
if (cancelled) return
const final = new Map(asyncResolvedIds.value)
for (const r of results) {
if (r.status === 'fulfilled') {
final.set(r.value.type, r.value.packId)
} else {
console.warn('Failed to resolve pack ID:', r.reason)
}
}
// Clear any remaining RESOLVING markers for failed lookups
for (const type of resolvingTypes) {
if (final.get(type) === RESOLVING) final.set(type, null)
}
asyncResolvedIds.value = final
},
{ immediate: true }
)
// Evict stale entries when missing nodes are cleared
watch(
() => missingNodesStore.missingNodesError,
(error) => {
if (!error && asyncResolvedIds.value.size > 0) {
asyncResolvedIds.value = new Map()
}
}
)
const missingPackGroups = computed<MissingPackGroup[]>(() => {
const nodeTypes = missingNodesStore.missingNodesError?.nodeTypes ?? []
const map = new Map<
string | null,
{ nodeTypes: MissingNodeType[]; isResolving: boolean }
>()
const resolvingKeys = new Set<string | null>()
for (const nodeType of nodeTypes) {
if (typeof nodeType !== 'string' && nodeType.isReplaceable) continue
let packId: string | null
if (typeof nodeType === 'string') {
packId = null
} else if (nodeType.cnrId) {
packId = nodeType.cnrId
} else {
const resolved = asyncResolvedIds.value.get(nodeType.type)
if (resolved === undefined || resolved === RESOLVING) {
packId = null
resolvingKeys.add(null)
} else {
packId = resolved
}
}
const existing = map.get(packId)
if (existing) {
existing.nodeTypes.push(nodeType)
} else {
map.set(packId, { nodeTypes: [nodeType], isResolving: false })
}
}
for (const key of resolvingKeys) {
const group = map.get(key)
if (group) group.isResolving = true
}
return Array.from(map.entries())
.sort(([packIdA], [packIdB]) => {
// null (Unknown Pack) always goes last
if (packIdA === null) return 1
if (packIdB === null) return -1
return packIdA.localeCompare(packIdB)
})
.map(([packId, { nodeTypes, isResolving }]) => ({
packId,
nodeTypes: [...nodeTypes].sort((a, b) => {
const typeA = typeof a === 'string' ? a : a.type
const typeB = typeof b === 'string' ? b : b.type
const typeCmp = typeA.localeCompare(typeB)
if (typeCmp !== 0) return typeCmp
const idA = typeof a === 'string' ? '' : String(a.nodeId ?? '')
const idB = typeof b === 'string' ? '' : String(b.nodeId ?? '')
return idA.localeCompare(idB, undefined, { numeric: true })
}),
isResolving
}))
})
const swapNodeGroups = computed<SwapNodeGroup[]>(() => {
const nodeTypes = missingNodesStore.missingNodesError?.nodeTypes ?? []
const map = new Map<string, SwapNodeGroup>()
for (const nodeType of nodeTypes) {
if (typeof nodeType === 'string' || !nodeType.isReplaceable) continue
const typeName = nodeType.type
const existing = map.get(typeName)
if (existing) {
existing.nodeTypes.push(nodeType)
} else {
map.set(typeName, {
type: typeName,
newNodeId: nodeType.replacement?.new_node_id,
nodeTypes: [nodeType]
})
}
}
return Array.from(map.values()).sort((a, b) => a.type.localeCompare(b.type))
})
/**
* Builds ErrorGroups from missingNodesError. Returns [] when none present.
* `includeGroup` narrows which swap/pack groups are counted (used to scope
* emphasis to the canvas selection); groups reduced to zero are omitted.
*/
function buildMissingNodeGroups(
includeGroup: (nodeTypes: MissingNodeType[]) => boolean = () => true
): ErrorGroup[] {
const error = missingNodesStore.missingNodesError
if (!error) return []
const groups: ErrorGroup[] = []
const swapCount = swapNodeGroups.value.filter((group) =>
includeGroup(group.nodeTypes)
).length
const packCount = missingPackGroups.value.filter((group) =>
includeGroup(group.nodeTypes)
).length
if (swapCount > 0) {
groups.push({
type: 'swap_nodes' as const,
groupKey: 'swap_nodes',
count: swapCount,
priority: 0,
...resolveMissingErrorMessage({
kind: 'swap_nodes',
nodeTypes: error.nodeTypes,
count: swapCount,
isCloud
})
})
}
if (packCount > 0) {
groups.push({
type: 'missing_node' as const,
groupKey: 'missing_node',
count: packCount,
priority: 1,
...resolveMissingErrorMessage({
kind: 'missing_node',
nodeTypes: error.nodeTypes,
count: packCount,
isCloud
})
})
}
return groups.sort((a, b) => a.priority - b.priority)
}
const missingModelGroups = computed<MissingModelGroup[]>(() => {
return groupMissingModelCandidates(
missingModelStore.missingModelCandidates,
isCloud
)
})
function buildMissingModelGroups(): ErrorGroup[] {
if (!missingModelGroups.value.length) return []
const count = countMissingModels(missingModelGroups.value)
return [
{
type: 'missing_model' as const,
groupKey: 'missing_model',
count,
priority: 2,
...resolveMissingErrorMessage({
kind: 'missing_model',
groups: missingModelGroups.value,
count,
isCloud
})
}
]
}
const missingMediaGroups = computed<MissingMediaGroup[]>(() => {
const candidates = missingMediaStore.missingMediaCandidates
if (!candidates?.length) return []
return groupCandidatesByMediaType(candidates)
})
function buildMissingMediaGroups(): ErrorGroup[] {
if (!missingMediaGroups.value.length) return []
const totalRows = countMissingMediaReferences(missingMediaGroups.value)
return [
{
type: 'missing_media' as const,
groupKey: 'missing_media',
count: totalRows,
priority: 3,
...resolveMissingErrorMessage({
kind: 'missing_media',
groups: missingMediaGroups.value,
count: totalRows,
isCloud
})
}
]
}
function isAssetErrorInSelection(executionNodeId: NodeExecutionId): boolean {
const nodeIds = selectedNodeInfo.value.nodeIds
if (!nodeIds) return true
// Try missing node cache first
const cachedNode = missingNodeCache.value.get(executionNodeId)
if (cachedNode && nodeIds.has(String(cachedNode.id))) return true
// Resolve from graph for model/media candidates
if (app.rootGraph) {
const graphNode = getNodeByExecutionId(app.rootGraph, executionNodeId)
if (graphNode && nodeIds.has(String(graphNode.id))) return true
}
for (const containerExecId of selectedNodeInfo.value
.containerExecutionIds) {
if (executionNodeId.startsWith(`${containerExecId}:`)) return true
}
return false
}
function isAssetCandidateInSelection(nodeId: string | number): boolean {
const executionNodeId = tryNormalizeNodeExecutionId(nodeId)
return executionNodeId ? isAssetErrorInSelection(executionNodeId) : false
}
/** Model groups narrowed to the selection, for emphasis derivation only. */
const missingModelGroupsForSelection = computed(() => {
if (!hasSelection.value) return []
const candidates = missingModelStore.missingModelCandidates
if (!candidates?.length) return []
const matched = candidates.filter(
(c) => c.nodeId != null && isAssetCandidateInSelection(c.nodeId)
)
if (!matched.length) return []
return groupMissingModelCandidates(matched, isCloud)
})
/** Media groups narrowed to the selection, for emphasis derivation only. */
const missingMediaGroupsForSelection = computed(() => {
if (!hasSelection.value) return []
const candidates = missingMediaStore.missingMediaCandidates
if (!candidates?.length) return []
const matched = candidates.filter(
(c) => c.nodeId != null && isAssetCandidateInSelection(c.nodeId)
)
if (!matched.length) return []
return groupCandidatesByMediaType(matched)
})
function buildMissingModelGroupsForSelection(): ErrorGroup[] {
if (!missingModelGroupsForSelection.value.length) return []
const count = countMissingModels(missingModelGroupsForSelection.value)
return [
{
type: 'missing_model' as const,
groupKey: 'missing_model',
count,
priority: 2,
...resolveMissingErrorMessage({
kind: 'missing_model',
groups: missingModelGroupsForSelection.value,
count,
isCloud
})
}
]
}
function buildMissingMediaGroupsForSelection(): ErrorGroup[] {
if (!missingMediaGroupsForSelection.value.length) return []
const totalRows = countMissingMediaReferences(
missingMediaGroupsForSelection.value
)
return [
{
type: 'missing_media' as const,
groupKey: 'missing_media',
count: totalRows,
priority: 3,
...resolveMissingErrorMessage({
kind: 'missing_media',
groups: missingMediaGroupsForSelection.value,
count: totalRows,
isCloud
})
}
]
}
const allErrorGroups = computed<ErrorGroup[]>(() => {
const groupsMap = new Map<string, GroupEntry>()
processPromptError(groupsMap)
processNodeErrors(groupsMap)
processExecutionError(groupsMap)
return [
...buildMissingNodeGroups(),
...buildMissingModelGroups(),
...buildMissingMediaGroups(),
...toSortedGroups(groupsMap)
]
})
/**
* The subset of error groups whose errors belong to the current canvas
* selection. Empty when nothing is selected. Display always shows all
* groups; this subset only drives selection emphasis (auto-expand, card
* highlight, context strip).
*/
const selectionScopedGroups = computed<ErrorGroup[]>(() => {
if (!hasSelection.value) return []
const groupsMap = new Map<string, GroupEntry>()
processPromptError(groupsMap, true)
processNodeErrors(groupsMap, true)
processExecutionError(groupsMap, true)
return [
...buildMissingNodeGroups((nodeTypes) =>
someNodeTypeInSelection(nodeTypes, selectionMatchedAssetNodeIds.value)
),
...buildMissingModelGroupsForSelection(),
...buildMissingMediaGroupsForSelection(),
...toSortedGroups(groupsMap)
]
})
/**
* Execution node ids referenced by any missing-asset candidate (models,
* media, missing node types).
*/
const assetNodeIdsWithError = computed<string[]>(() => {
const candidateIds = [
...(missingModelStore.missingModelCandidates ?? []),
...(missingMediaStore.missingMediaCandidates ?? [])
].map((candidate) => candidate.nodeId)
const missingNodeTypeIds = (
missingNodesStore.missingNodesError?.nodeTypes ?? []
).map((nodeType) =>
typeof nodeType === 'string' ? undefined : nodeType.nodeId
)
return [...candidateIds, ...missingNodeTypeIds]
.filter((nodeId) => nodeId != null)
.map(String)
})
/**
* Asset node ids that belong to the current selection. Drives row-level
* highlighting inside the missing-* cards.
*/
const selectionMatchedAssetNodeIds = computed<Set<string>>(() => {
if (!hasSelection.value) return new Set()
return new Set(
assetNodeIdsWithError.value.filter(isAssetCandidateInSelection)
)
})
const selectionMatchedGroupKeys = computed<Set<string>>(() => {
if (!hasSelection.value) return new Set()
return new Set(selectionScopedGroups.value.map((group) => group.groupKey))
})
const selectionMatchedCardIds = computed<Set<string>>(() => {
if (!hasSelection.value) return new Set()
return new Set(
selectionScopedGroups.value
.flatMap((group) => (group.type === 'execution' ? group.cards : []))
.map((card) => card.id)
)
})
const selectionErrorCount = computed(() => {
if (!hasSelection.value) return 0
return selectionScopedGroups.value.reduce(
(sum, group) => sum + group.count,
0
)
})
/** Distinct nodes affected by any error (workflow-level summary). */
const errorNodeCount = computed(() => {
const executionNodeIds = allErrorGroups.value
.flatMap((group) => (group.type === 'execution' ? group.cards : []))
.map((card) => card.nodeId)
.filter((nodeId) => nodeId != null)
return new Set([...executionNodeIds, ...assetNodeIdsWithError.value]).size
})
const filteredGroups = computed<ErrorGroup[]>(() => {
const query = toValue(searchQuery).trim()
return searchErrorGroups(allErrorGroups.value, query)
})
return {
allErrorGroups,
filteredGroups,
collapseState,
errorNodeCache,
missingNodeCache,
missingPackGroups,
missingModelGroups,
missingMediaGroups,
swapNodeGroups,
hasSelection,
selectedNodeCount,
selectedNodeTitle,
selectionMatchedGroupKeys,
selectionMatchedCardIds,
selectionMatchedAssetNodeIds,
selectionErrorCount,
errorNodeCount
}
}