mirror of
https://gitee.com/mateos/mateclaw.git
synced 2026-09-13 19:23:42 +08:00
2870 lines
109 KiB
Vue
2870 lines
109 KiB
Vue
<template>
|
||
<div class="mc-page-shell chat-console-shell">
|
||
<div class="mc-page-frame chat-console-frame">
|
||
<div class="chat-layout mc-surface-card">
|
||
<!-- 移动端会话面板遮罩 -->
|
||
<Transition name="fade">
|
||
<div v-if="isMobile && convPanelOpen" class="conv-backdrop" @click="convPanelOpen = false"></div>
|
||
</Transition>
|
||
|
||
<!-- 会话侧边栏 -->
|
||
<ConversationSidebar
|
||
:conversations="conversations"
|
||
:current-conversation-id="currentConversationId"
|
||
:agents="agents"
|
||
:selected-agent-id="selectedAgentId"
|
||
:collapsed="convPanelCollapsed"
|
||
:mobile-open="convPanelOpen"
|
||
:is-mobile="isMobile"
|
||
@select="selectConversation"
|
||
@new-chat="newConversation"
|
||
@agent-picked="onAgentPicked"
|
||
@toggle-collapse="toggleConvPanel"
|
||
@refresh="loadConversations"
|
||
@deleted="onConversationsDeleted"
|
||
/>
|
||
|
||
<!-- 主聊天区域 -->
|
||
<div
|
||
class="chat-area"
|
||
@dragenter.prevent="onDragEnter"
|
||
@dragover.prevent
|
||
@dragleave="onDragLeave"
|
||
@drop.prevent="onDrop"
|
||
>
|
||
<!-- 拖拽上传遮罩 -->
|
||
<Transition name="fade">
|
||
<div v-if="isDragging" class="drop-overlay">
|
||
<div class="drop-overlay__content">
|
||
<el-icon><UploadFilled /></el-icon>
|
||
<span>{{ $t('chat.dropToUpload') }}</span>
|
||
</div>
|
||
</div>
|
||
</Transition>
|
||
<!-- 头部 -->
|
||
<div class="chat-header">
|
||
<div class="chat-header-left">
|
||
<button v-if="isMobile" class="conv-toggle-btn" @click="convPanelOpen = !convPanelOpen" :title="$t('chat.conversations')">
|
||
<el-icon><ChatDotRound /></el-icon>
|
||
</button>
|
||
<div class="chat-stage-copy" v-if="currentAgent">
|
||
<div class="chat-stage-kicker">{{ $t('nav.chat') }}</div>
|
||
<!--
|
||
Header reads as "who is this employee" — name + tagline.
|
||
The runtime mode (ReAct / Plan-Execute) is technical jargon
|
||
to end users and lives in the badge tooltip instead, so the
|
||
header doesn't get polluted.
|
||
-->
|
||
<div
|
||
class="agent-badge"
|
||
:title="`${currentAgent.name}${currentAgentRuntimeMode ? ' · ' + currentAgentRuntimeMode : ''}`"
|
||
>
|
||
<span class="agent-badge-icon" :style="{ color: agentIconColor(currentAgent.icon) }"><SkillIcon :value="currentAgent.icon" :size="22" :fallback="'🤖'" /></span>
|
||
<div class="agent-badge-text">
|
||
<span class="agent-badge-name">{{ currentAgent.name }}</span>
|
||
</div>
|
||
<span class="status-dot" :class="connectionStatusClass" :title="connectionStatusLabel"></span>
|
||
</div>
|
||
</div>
|
||
<div v-else class="no-agent-hint">{{ $t('chat.selectAgent') }}</div>
|
||
</div>
|
||
<div class="chat-header-right">
|
||
<!-- Model selector — Issue #81 v2 R3: always pass full providers + show-all-states
|
||
so unhealthy rows render as dimmed entries with status chips and a Fix
|
||
button instead of disappearing entirely. -->
|
||
<ModelSelector
|
||
:providers="providers"
|
||
:active-value="activeModelValue"
|
||
:active-label="activeModelLabel"
|
||
:saving="modelSaving"
|
||
:show-all-states="true"
|
||
:can-configure="canConfigureModels"
|
||
:empty-hint="modelSelectorEmptyHint"
|
||
@select="selectModel"
|
||
@navigate-fix="onModelSelectorFix"
|
||
/>
|
||
<!-- Overflow menu -->
|
||
<div class="header-overflow-wrap">
|
||
<button ref="headerBtnRef" class="header-btn" @click="headerMenuOpen = !headerMenuOpen" :title="$t('common.more')">
|
||
<svg width="16" height="16" viewBox="0 0 24 24" fill="currentColor"><circle cx="12" cy="5" r="1.5"/><circle cx="12" cy="12" r="1.5"/><circle cx="12" cy="19" r="1.5"/></svg>
|
||
</button>
|
||
<DropdownMenu
|
||
:open="headerMenuOpen"
|
||
:anchor="headerBtnRef"
|
||
:items="headerMenuItems"
|
||
@select="onHeaderMenuSelect"
|
||
@close="headerMenuOpen = false"
|
||
>
|
||
<template #item-icon="{ item }">
|
||
<el-icon v-if="item.key === 'config'"><Setting /></el-icon>
|
||
<el-icon v-else-if="item.key === 'clear'"><Delete /></el-icon>
|
||
</template>
|
||
</DropdownMenu>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
|
||
<TeamWorkerBanner
|
||
v-if="workerRunContext"
|
||
:run-id="workerRunContext.runId"
|
||
:task-id="workerRunContext.taskId"
|
||
:team-id="workerRunContext.teamId"
|
||
:lead-conversation-id="workerRunContext.leadConversationId"
|
||
@navigate="router.push($event)"
|
||
/>
|
||
|
||
<!-- 使用组件化的 MessageList -->
|
||
<MessageList
|
||
ref="messageListRef"
|
||
:messages="messages"
|
||
:loading="isGenerating"
|
||
:assistant-icon="currentAgent?.icon || '🤖'"
|
||
:user-icon="userInitial"
|
||
:title="blockingPrompt ? modelPromptText.title : $t('app.title')"
|
||
:subtitle="blockingPrompt ? modelPromptText.desc : $t('chat.subtitle')"
|
||
:suggestions="blockingPrompt ? [] : suggestions"
|
||
:team-runs="teamRuns"
|
||
:expanded-team-run-id="teamRunRouteQuery.teamRunId || null"
|
||
:selected-team-task-id="teamRunRouteQuery.taskId || null"
|
||
:team-runs-has-more="Boolean(teamRunsNextCursor)"
|
||
:team-runs-loading-more="teamRunsLoadingMore"
|
||
:readonly="workerConversationReadOnly"
|
||
@regenerate="handleRegenerate"
|
||
@rewind="handleRewind"
|
||
@suggestion-click="sendSuggestion"
|
||
@toggle-thinking="handleToggleThinking"
|
||
@approve="handleApprove"
|
||
@approve-always="handleApproveAlways"
|
||
@deny="handleDeny"
|
||
@team-run-navigate="router.push($event)"
|
||
@team-runs-load-more="loadMoreTeamRuns"
|
||
>
|
||
<!-- Issue #81 v2 R2: blocking-only popup. Recoverable cases use the
|
||
non-blocking <RecoverableModelBanner> below instead. -->
|
||
<template v-if="blockingPrompt" #empty>
|
||
<div class="model-prompt">
|
||
<div class="model-prompt-title">{{ modelPromptText.title }}</div>
|
||
<div class="model-prompt-desc">{{ modelPromptText.desc }}</div>
|
||
<div class="model-prompt-actions">
|
||
<button class="btn-primary" @click="handlePrimaryAction">
|
||
{{ primaryActionLabel }}
|
||
</button>
|
||
<button
|
||
v-if="bestSwitchTarget"
|
||
class="btn-secondary"
|
||
@click="switchToBestTarget"
|
||
>
|
||
{{ $t('chat.promptAction.switchToModel', { name: bestSwitchTarget.label }) }}
|
||
</button>
|
||
</div>
|
||
</div>
|
||
</template>
|
||
</MessageList>
|
||
|
||
<!-- Issue #81: non-blocking banner — active provider is unhealthy but the
|
||
backend fallback chain has a LIVE provider to take over. -->
|
||
<RecoverableModelBanner
|
||
v-if="recoverablePrompt && activeProvider && bestFallbackName"
|
||
:provider-name="activeProvider.name"
|
||
:fallback-name="bestFallbackName"
|
||
@dismiss="recoverableDismissed = true"
|
||
/>
|
||
|
||
<!-- Cron job in-flight placeholder — visible while T2 hasn't committed
|
||
the assistant message yet. Populated by pollActivity → /cron-jobs/active-runs. -->
|
||
<div v-if="activeCronRuns.length > 0" class="cron-running-bar">
|
||
<div v-for="run in activeCronRuns" :key="run.runId" class="cron-running-item">
|
||
<span class="cron-running-spinner">🌀</span>
|
||
<span class="cron-running-text">
|
||
<strong>{{ run.jobName || $t('chat.cronRunning.fallbackName') }}</strong>
|
||
<span class="cron-running-meta">
|
||
· {{ $t('chat.cronRunning.executing') }}
|
||
<template v-if="run.startedAt"> · {{ elapsedLabel(run.startedAt) }}</template>
|
||
</span>
|
||
</span>
|
||
</div>
|
||
</div>
|
||
|
||
<!-- Terminal-state announcement after a goal completed or exhausted
|
||
in this conversation. Auto-dismisses when the user clicks × or
|
||
starts a new goal. -->
|
||
<GoalSystemLine
|
||
v-if="goalTerminalForCurrent && currentConversationId"
|
||
:variant="goalTerminalForCurrent.status"
|
||
:title="goalSystemLineTitle"
|
||
:detail="goalSystemLineDetail"
|
||
class="goal-system-line-slot"
|
||
@click.stop="onGoalSystemLineDismiss"
|
||
/>
|
||
|
||
<!-- Inline "set a goal?" invitation shown after the first assistant
|
||
reply when the conversation has no active goal and the user
|
||
hasn't dismissed it for this conv. -->
|
||
<GoalSetInlinePrompt
|
||
v-if="showGoalSetPrompt"
|
||
:conversation-id="currentConversationId"
|
||
:agent-id="String(selectedAgentId)"
|
||
:workspace-id="String(currentWorkspaceId || '1')"
|
||
:suggested-title="goalSuggestedTitle"
|
||
class="goal-set-prompt-slot"
|
||
@dismiss="onGoalPromptDismiss"
|
||
/>
|
||
|
||
<!-- 流式处理 Loading 栏（消息和输入框之间） -->
|
||
<StreamLoadingBar
|
||
:is-loading="isGenerating && !blockingPrompt"
|
||
:tool-count="toolCallCount"
|
||
:completion-tokens="currentGeneratingTokens"
|
||
:prompt-tokens="currentPromptTokens"
|
||
:phase="streamPhase"
|
||
:phase-info="phaseInfo"
|
||
:running-tool-name="currentRunningToolName"
|
||
:has-queued="hasQueued"
|
||
:lifecycle-stage="lifecycleStage"
|
||
:compact-status="compactStatus"
|
||
/>
|
||
|
||
<!-- Multimodal routing hint: shown when pending attachments require a
|
||
modality the primary model lacks. -->
|
||
<MultimodalRoutingHint
|
||
:attachments="pendingAttachments"
|
||
:capabilities="agentCapabilities"
|
||
/>
|
||
|
||
<!-- 上下文占用（估算）：点击展开分项面板 -->
|
||
<div v-if="contextUsage" class="ctx-usage-row">
|
||
<ContextUsagePanel :usage="contextUsage" />
|
||
</div>
|
||
|
||
<!-- 使用组件化的 ChatInput -->
|
||
<ChatInput
|
||
ref="chatInputRef"
|
||
v-model="inputText"
|
||
:loading="isGenerating && !hasPendingApproval"
|
||
:disabled="blockingPrompt || !currentAgent || workerConversationReadOnly"
|
||
:skills-enabled="!!currentAgent && !currentAgent.skillsDisabled"
|
||
:placeholder="$t('chat.messagePlaceholder')"
|
||
:hint="currentRuntimeModel"
|
||
:attachments="pendingAttachments"
|
||
:uploading="uploadingAttachment"
|
||
:max-length="10240"
|
||
:pending-approval="activePendingApproval"
|
||
:stream-phase="streamPhase"
|
||
:queued-message="queuedMessage"
|
||
:queue-size="queueSize"
|
||
@submit="handleSendMessage"
|
||
@stop="handleStopStream"
|
||
@cancel-queued="handleCancelQueued"
|
||
@file-select="handleFileSelect"
|
||
@attachment-remove="removeAttachment"
|
||
@approve="handleApprove"
|
||
@approve-always="handleApproveAlways"
|
||
@deny="handleDeny"
|
||
:enable-talk-mode="!!selectedAgentId"
|
||
:thinking-enabled="thinkingEnabled"
|
||
:thinking-supported="currentModelSupportsThinking"
|
||
@toggle-thinking="thinkingEnabled = !thinkingEnabled"
|
||
@talk="showTalkMode = true"
|
||
/>
|
||
</div>
|
||
|
||
<!-- 运行总览侧栏：计划进度 + 子 Agent 实时状态 -->
|
||
<RunOverviewPanel
|
||
:messages="messages"
|
||
:is-generating="isGenerating"
|
||
:agent-type="currentAgent?.agentType"
|
||
/>
|
||
|
||
<!-- Talk Mode 覆盖层 -->
|
||
<TalkMode
|
||
v-if="showTalkMode"
|
||
:visible="showTalkMode"
|
||
:agent-id="selectedAgentId"
|
||
:conversation-id="currentConversationId"
|
||
@close="showTalkMode = false"
|
||
/>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
</template>
|
||
|
||
<script lang="ts">
|
||
let cachedAgents: import('@/types').Agent[] = []
|
||
</script>
|
||
|
||
<script setup lang="ts">
|
||
import { ElMessage } from 'element-plus/es/components/message/index'
|
||
import { ElMessageBox } from 'element-plus/es/components/message-box/index'
|
||
import { ref, computed, onMounted, onBeforeUnmount, onActivated, onDeactivated, watch, nextTick } from 'vue'
|
||
import { useRoute, useRouter } from 'vue-router'
|
||
import { useI18n } from 'vue-i18n'
|
||
import { mcToast } from '@/composables/useMcToast'
|
||
import { ChatDotRound, Delete, Setting, UploadFilled } from '@element-plus/icons-vue'
|
||
import { conversationApi, agentApi, modelApi, chatApi, cronJobApi, approvalApi } from '@/api/index'
|
||
import { copyToClipboard } from '@/utils/clipboard'
|
||
import { useFileDrop } from '@/composables/useFileDrop'
|
||
import { useIsMobile, useMediaQuery, BREAKPOINTS } from '@/composables/useBreakpoint'
|
||
import { useChat } from '@/composables/chat/useChat'
|
||
import { useTeamRuns } from '@/composables/chat/useTeamRuns'
|
||
import { useWorkerConversationGuard } from '@/composables/chat/useWorkerConversationGuard'
|
||
import { parseTeamMessageMetadata } from '@/composables/chat/messageMetadata'
|
||
import RunOverviewPanel from '@/components/chat/RunOverviewPanel.vue'
|
||
import { reconstructErrorInfo } from '@/types/chatError'
|
||
import { reconcileMessages, extractMessages } from '@/utils/messageReconcile'
|
||
import {
|
||
buildChatRouteQuery,
|
||
readLegacyWorkerRouteContext,
|
||
readTeamRunRouteQuery,
|
||
resolveConversationAgentSelection,
|
||
resolveRouteHydrationQuery,
|
||
} from '@/utils/chatRouteHydration'
|
||
import type { Conversation, Agent, ModelConfig, ProviderInfo, ActiveModelsInfo, ChatAttachment, MessageContentPart, Message, ToolCallMeta } from '@/types'
|
||
|
||
// 导入组件化组件
|
||
import MessageList from '@/components/chat/MessageList.vue'
|
||
import TeamWorkerBanner from '@/components/chat/TeamWorkerBanner.vue'
|
||
import RecoverableModelBanner from '@/components/chat/RecoverableModelBanner.vue'
|
||
import SkillIcon from '@/components/common/SkillIcon.vue'
|
||
import ConversationSidebar from '@/components/chat/ConversationSidebar.vue'
|
||
import DropdownMenu, { type DropdownMenuItem } from '@/components/common/DropdownMenu.vue'
|
||
import { agentIconColor } from '@/utils/agentIconColor'
|
||
import ChatInput from '@/components/chat/ChatInput.vue'
|
||
import MultimodalRoutingHint from '@/components/chat/MultimodalRoutingHint.vue'
|
||
import StreamLoadingBar from '@/components/chat/StreamLoadingBar.vue'
|
||
import ContextUsagePanel from '@/components/chat/ContextUsagePanel.vue'
|
||
import TalkMode from '@/components/chat/TalkMode.vue'
|
||
import ModelSelector from '@/components/chat/ModelSelector.vue'
|
||
import { useEChartsRenderer } from '@/composables/useEChartsRenderer'
|
||
import { useKatexRenderer } from '@/composables/useKatexRenderer'
|
||
import { useMermaidRenderer, handleMermaidDownload } from '@/composables/useMermaidRenderer'
|
||
import { useGoalStore } from '@/stores/useGoalStore'
|
||
import { useWorkspaceStore } from '@/stores/useWorkspaceStore'
|
||
import { buildViewerModelProviders } from '@/utils/viewerModelProviders'
|
||
import GoalSetInlinePrompt from '@/components/goal/GoalSetInlinePrompt.vue'
|
||
import GoalSystemLine from '@/components/goal/GoalSystemLine.vue'
|
||
|
||
// ============ Talk Mode ============
|
||
const showTalkMode = ref(false)
|
||
|
||
// ============ 移动端 & 响应式状态 ============
|
||
const convPanelOpen = ref(false)
|
||
const convPanelCollapsed = ref(localStorage.getItem('mc-conv-collapsed') === 'true')
|
||
const userExplicitConvCollapse = ref(localStorage.getItem('mc-conv-collapsed') === 'true')
|
||
|
||
const isMobile = useIsMobile()
|
||
const compactViewport = useMediaQuery(BREAKPOINTS.compact)
|
||
|
||
// Leaving the mobile breakpoint closes the conversation drawer.
|
||
watch(isMobile, (mobile) => {
|
||
if (!mobile) convPanelOpen.value = false
|
||
})
|
||
|
||
// Auto-collapse the conversation panel on narrow desktop unless the user
|
||
// toggled it explicitly.
|
||
watch(compactViewport, (compact) => {
|
||
if (!userExplicitConvCollapse.value) convPanelCollapsed.value = compact
|
||
}, { immediate: true })
|
||
|
||
function toggleConvPanel() {
|
||
convPanelCollapsed.value = !convPanelCollapsed.value
|
||
userExplicitConvCollapse.value = convPanelCollapsed.value
|
||
localStorage.setItem('mc-conv-collapsed', String(convPanelCollapsed.value))
|
||
}
|
||
|
||
// ============ 配置和常量 ============
|
||
const suggestions = computed(() => {
|
||
const agent = currentAgent.value
|
||
// If agent has custom suggestions (stored as newline-separated string in description etc.)
|
||
const agentSuggestions = (agent as any)?.suggestions as string | undefined
|
||
if (agentSuggestions) {
|
||
const parsed = agentSuggestions.split('\n').filter(Boolean).slice(0, 4)
|
||
if (parsed.length) return parsed
|
||
}
|
||
// Agent-type-aware defaults
|
||
if (agent?.agentType === 'plan_execute') {
|
||
return [
|
||
t('chat.suggestionPlan1', '帮我制定一个完整的项目计划'),
|
||
t('chat.suggestionPlan2', '分步骤帮我完成一个复杂任务'),
|
||
t('chat.suggestionIntro'),
|
||
t('chat.suggestionWeather'),
|
||
]
|
||
}
|
||
return [
|
||
t('chat.suggestionIntro'),
|
||
t('chat.suggestionPoem'),
|
||
t('chat.suggestionCode'),
|
||
t('chat.suggestionWeather'),
|
||
]
|
||
})
|
||
|
||
// ============ 状态 ============
|
||
const router = useRouter()
|
||
const route = useRoute()
|
||
const { t } = useI18n()
|
||
|
||
const agents = ref<Agent[]>(cachedAgents.length > 0 ? [...cachedAgents] : [])
|
||
const conversations = ref<Conversation[]>([])
|
||
const selectedAgentId = ref<string | number>('')
|
||
const currentConversationId = ref<string>('')
|
||
const inputText = ref('')
|
||
const modelSaving = ref(false)
|
||
// Monotonic counter for in-flight setModel PUTs. The finally handler
|
||
// only clears modelSaving when its captured seq is still the latest, so a
|
||
// stale-finishing earlier PUT can't unlock the selector while a newer one
|
||
// is still in flight, and (crucially) switching conversations mid-PUT
|
||
// can't permanently lock the selector by leaving modelSaving stuck true.
|
||
let modelSaveSeq = 0
|
||
// Issue #81 v2 R2: split the single showModelPrompt boolean into two flags so
|
||
// the chat surface can either hard-block (blockingPrompt) or warn but let the
|
||
// backend fallback chain take over (recoverablePrompt). Driven by
|
||
// recomputePromptFlags() — see the watcher below.
|
||
const blockingPrompt = ref(false)
|
||
const recoverablePrompt = ref(false)
|
||
const recoverableDismissed = ref(false)
|
||
const defaultModel = ref<ModelConfig | null>(null)
|
||
const providers = ref<ProviderInfo[]>([])
|
||
// True when a viewer uses the credential-free provider projection. Provider
|
||
// config (API keys, base URLs, liveness) is admin-only, so viewers can switch
|
||
// models but prompt flags must still trust the active model's runtime state.
|
||
const providersUnavailable = ref(false)
|
||
// Mirror of /models/enabled (viewer-accessible). It supplies the model half of
|
||
// the safe picker projection and resolves the active label during hydration.
|
||
const enabledModels = ref<ModelConfig[]>([])
|
||
// The model the CURRENT conversation uses. Per-conversation — switching it
|
||
// never leaks into other conversations (see selectModel / applyConversationModel).
|
||
const activeModels = ref<ActiveModelsInfo | null>(null)
|
||
// Global default model — seeds the selector for conversations with no pin yet.
|
||
const globalDefaultModel = ref<{ providerId: string; model: string } | null>(null)
|
||
// True once the user manually picks a model for the CURRENT conversation.
|
||
// Guards the agent-capability re-seed (see the selectedAgentId watcher) from
|
||
// clobbering that pick during the window where the async capability fetch
|
||
// resolves after a fresh conversation was already seeded. Reset whenever we
|
||
// move to a different conversation.
|
||
const userPickedModel = ref(false)
|
||
const pendingAttachments = ref<ChatAttachment[]>([])
|
||
const uploadingAttachment = ref(false)
|
||
|
||
// Per-agent capability snapshot for the multimodal routing hint above the
|
||
// input box. Refetched whenever the active agent changes; cached locally to
|
||
// avoid an extra request per attachment change.
|
||
const agentCapabilities = ref<import('@/types').AgentCapabilities | null>(null)
|
||
|
||
// 思考模式：只有两个状态 — 开或关
|
||
const thinkingEnabled = ref(localStorage.getItem('mateclaw_thinking') !== 'off')
|
||
const thinkingLevel = computed(() => thinkingEnabled.value ? 'high' : 'off')
|
||
watch(thinkingEnabled, (v) => localStorage.setItem('mateclaw_thinking', v ? 'on' : 'off'))
|
||
|
||
// Dropdowns & menus
|
||
const headerMenuOpen = ref(false)
|
||
const headerBtnRef = ref<HTMLElement | null>(null)
|
||
|
||
const headerMenuItems = computed<DropdownMenuItem[]>(() => [
|
||
{ key: 'config', label: t('chat.configModel') },
|
||
{ key: 'sessions', label: t('chat.openSessions') },
|
||
{ divider: true },
|
||
{ key: 'clear', label: t('chat.clearMessages'), danger: true },
|
||
])
|
||
|
||
function onHeaderMenuSelect(item: DropdownMenuItem) {
|
||
if (item.key === 'config') goToModelSettings()
|
||
else if (item.key === 'sessions') router.push('/sessions')
|
||
else if (item.key === 'clear') clearMessages()
|
||
}
|
||
|
||
function onAgentPicked(value: string | number | null) {
|
||
if (value == null) return
|
||
if (String(value) !== String(selectedAgentId.value)) {
|
||
selectedAgentId.value = value
|
||
newConversation()
|
||
}
|
||
}
|
||
|
||
function selectModel(value: string) {
|
||
const [providerId, model] = value.split('::')
|
||
if (!providerId || !model) return
|
||
// Remember the explicit pick so a late-arriving agent-capability fetch
|
||
// doesn't re-seed over it on a not-yet-persisted conversation.
|
||
userPickedModel.value = true
|
||
// Per-conversation model: switching here only affects THIS conversation.
|
||
// We update the selector + the local list entry immediately so the UI is
|
||
// responsive, then persist the pin to the server right away IF the
|
||
// conversation already exists. Without the eager persist, IM channels
|
||
// (Feishu / DingTalk / WeCom …) keep using whatever the conversation row
|
||
// last had — they don't see the /chat/stream payload that the web path
|
||
// pins on send — so the user "switches model in the chat box" but the
|
||
// next IM inbound message still picks the old / default model.
|
||
//
|
||
// Snapshot the previous selection BEFORE the optimistic update so a
|
||
// failed PUT can roll the UI back instead of stranding the user with a
|
||
// model the backend isn't using.
|
||
const prevLlm = activeModels.value?.activeLlm
|
||
const prevActive: ActiveModelsInfo | null = prevLlm?.providerId && prevLlm?.model
|
||
? { activeLlm: { providerId: prevLlm.providerId, model: prevLlm.model } }
|
||
: null
|
||
const conv = conversations.value.find(c => c.conversationId === currentConversationId.value)
|
||
const prevConvProvider = conv?.modelProvider
|
||
const prevConvModel = conv?.modelName
|
||
|
||
activeModels.value = { activeLlm: { providerId, model } }
|
||
if (conv) {
|
||
conv.modelProvider = providerId
|
||
conv.modelName = model
|
||
}
|
||
// Only persist when the conversation is already in the server-side list.
|
||
// A brand-new chat (newConversation() generated a local id that hasn't
|
||
// been sent through /chat/stream yet) has no row to PUT against; that
|
||
// case still relies on the first /chat/stream call writing the pin.
|
||
if (conv && currentConversationId.value) {
|
||
const cid = currentConversationId.value
|
||
const mySeq = ++modelSaveSeq
|
||
modelSaving.value = true
|
||
conversationApi.setModel(cid, providerId, model)
|
||
.catch((e: any) => {
|
||
console.warn('[ChatConsole] Failed to persist model pin:', e)
|
||
mcToast.warning(t('chat.modelSaveFailed'))
|
||
// Roll back the visible selector + the cached conv pin so the UI
|
||
// doesn't keep claiming a model the backend isn't using. Only do
|
||
// it when the user is still on the same conversation AND hasn't
|
||
// picked yet another model — otherwise we'd corrupt the more
|
||
// recent state with this PUT's snapshot.
|
||
const liveConv = conversations.value.find(c => c.conversationId === cid)
|
||
if (liveConv && liveConv.modelProvider === providerId && liveConv.modelName === model) {
|
||
liveConv.modelProvider = prevConvProvider
|
||
liveConv.modelName = prevConvModel
|
||
}
|
||
if (currentConversationId.value !== cid) return
|
||
const stillShowingFailedPick =
|
||
activeModels.value?.activeLlm?.providerId === providerId
|
||
&& activeModels.value?.activeLlm?.model === model
|
||
if (!stillShowingFailedPick) return
|
||
activeModels.value = prevActive
|
||
})
|
||
.finally(() => {
|
||
// Only the LATEST in-flight PUT clears the saving flag. An earlier
|
||
// PUT finishing late must not flip saving to false while a newer
|
||
// one is still pending (the selector would unlock during a live
|
||
// request); and a conversation switch mid-PUT must not strand the
|
||
// flag at true forever (which would lock the selector across
|
||
// every conversation — the original bug this seq counter fixes).
|
||
if (mySeq === modelSaveSeq) {
|
||
modelSaving.value = false
|
||
}
|
||
})
|
||
}
|
||
}
|
||
|
||
/**
|
||
* The selected agent's model override, as a (provider, model) pair — or null
|
||
* when the agent has no usable override.
|
||
*
|
||
* Two sources, in order:
|
||
* 1. agentCapabilities — the backend-resolved pair from /agents/{id}/capabilities.
|
||
* Authoritative: it runs the same resolveModel the runtime uses (honouring
|
||
* disabled→default fallback and provider disambiguation). But it arrives
|
||
* via an async fetch that resolves AFTER the synchronous seeding on an
|
||
* agent switch / deep-link, and is lost entirely if that fetch fails.
|
||
* 2. currentAgent.modelName resolved against the enabled-model list — a
|
||
* SYNCHRONOUS fallback so an agent switch, a capability-fetch failure, or a
|
||
* not-yet-hydrated deep-link still honour the per-agent override instead of
|
||
* silently dropping to the global default (which handleSendMessage would
|
||
* then pin, re-introducing the very clobber this whole change fixes).
|
||
*
|
||
* enabledModels is viewer-accessible (/models/enabled) and loaded at mount, so
|
||
* the fallback works for viewers too.
|
||
*/
|
||
function agentSeedModel(): { providerId: string; model: string } | null {
|
||
const cap = agentCapabilities.value
|
||
if (cap?.providerId && cap?.modelName) {
|
||
return { providerId: cap.providerId, model: cap.modelName }
|
||
}
|
||
const name = currentAgent.value?.modelName
|
||
if (name) {
|
||
const hit = enabledModels.value.find(m => m.modelName === name)
|
||
if (hit?.provider) {
|
||
return { providerId: hit.provider, model: hit.modelName }
|
||
}
|
||
}
|
||
return null
|
||
}
|
||
|
||
/**
|
||
* Point the model selector at the model this conversation should run, honouring
|
||
* the same precedence the backend uses when it resolves the runtime model
|
||
* (see AgentGraphBuilder.resolveRuntimeBaseModel):
|
||
*
|
||
* conversation pin > selected agent's model override > global default
|
||
*
|
||
* The middle tier is the fix for the "I set the agent's model but chat used
|
||
* another one" bug: without it a fresh conversation seeded the global default,
|
||
* and handleSendMessage then PINNED that default onto the conversation row —
|
||
* which outranks the agent override and silently clobbered it.
|
||
*/
|
||
function applyConversationModel(conv?: Conversation | null) {
|
||
if (conv?.modelProvider && conv?.modelName) {
|
||
activeModels.value = { activeLlm: { providerId: conv.modelProvider, model: conv.modelName } }
|
||
return
|
||
}
|
||
const agentModel = agentSeedModel()
|
||
if (agentModel) {
|
||
activeModels.value = { activeLlm: { providerId: agentModel.providerId, model: agentModel.model } }
|
||
return
|
||
}
|
||
if (globalDefaultModel.value) {
|
||
activeModels.value = { activeLlm: { ...globalDefaultModel.value } }
|
||
}
|
||
}
|
||
|
||
/**
|
||
* After a poll refreshes the conversation list, the currently-open
|
||
* conversation may have drifted server-side: an admin may have rebound the
|
||
* channel to a different agent, or pinned a different model via another
|
||
* tab / API call. Pull the new server-side state into the local selector +
|
||
* agent header so the chat surface doesn't keep claiming the old binding.
|
||
*
|
||
* Skipped while a turn is generating — yanking the model / agent mid-stream
|
||
* would orphan the active SSE subscription.
|
||
*/
|
||
function reconcileCurrentConversation() {
|
||
if (!currentConversationId.value) return
|
||
if (isGenerating.value) return
|
||
// Don't fight an in-flight setModel write — the poll cycle may run BEFORE
|
||
// the PUT lands, in which case the server still reports the old pin and
|
||
// we'd flicker the UI back. Wait for the next tick.
|
||
if (modelSaving.value) return
|
||
const fresh = conversations.value.find(c => c.conversationId === currentConversationId.value)
|
||
if (!fresh) return
|
||
if (fresh.agentId != null && String(fresh.agentId) !== String(selectedAgentId.value)) {
|
||
selectedAgentId.value = fresh.agentId
|
||
}
|
||
const pickedProvider = activeModels.value?.activeLlm?.providerId
|
||
const pickedModel = activeModels.value?.activeLlm?.model
|
||
const serverHasPin = !!(fresh.modelProvider && fresh.modelName)
|
||
if (serverHasPin) {
|
||
if (fresh.modelProvider !== pickedProvider || fresh.modelName !== pickedModel) {
|
||
activeModels.value = { activeLlm: { providerId: fresh.modelProvider!, model: fresh.modelName! } }
|
||
}
|
||
} else if (pickedProvider || pickedModel) {
|
||
// Server-side pin was cleared (admin reset, model deleted, …) but the
|
||
// local selector still shows the old pick. Drop back to whatever the
|
||
// global default resolves to — applyConversationModel does the right
|
||
// thing when conv has no pin.
|
||
applyConversationModel(fresh)
|
||
}
|
||
}
|
||
|
||
// 拖拽上传 — useFileDrop owns the hover/counter state; the directory-aware
|
||
// payload handling (electron paths vs web FileSystem entries) stays here.
|
||
const { isDragging, onDragEnter, onDragLeave, onDrop } = useFileDrop(processDroppedItems)
|
||
|
||
async function processDroppedItems(e: DragEvent) {
|
||
if (workerConversationReadOnly.value) return
|
||
const dtFiles = Array.from(e.dataTransfer?.files || [])
|
||
const items = Array.from(e.dataTransfer?.items || [])
|
||
|
||
const electronDirs: File[] = []
|
||
const webDirEntries: FileSystemDirectoryEntry[] = []
|
||
const regularFiles: File[] = []
|
||
|
||
for (let i = 0; i < items.length; i++) {
|
||
const entry = items[i].webkitGetAsEntry?.()
|
||
const file = dtFiles[i]
|
||
if (entry?.isDirectory) {
|
||
if ((file as any)?.path) {
|
||
// Electron: has absolute path
|
||
electronDirs.push(file)
|
||
} else {
|
||
// Web: need to recursively collect files
|
||
webDirEntries.push(entry as FileSystemDirectoryEntry)
|
||
}
|
||
} else if (file) {
|
||
regularFiles.push(file)
|
||
}
|
||
}
|
||
|
||
// Electron directories → record path reference
|
||
if (electronDirs.length) {
|
||
handleDirectoryAttach(electronDirs)
|
||
}
|
||
// Web directories → recursively collect files and upload
|
||
if (webDirEntries.length) {
|
||
const collected = await collectFilesFromEntries(webDirEntries)
|
||
if (collected.length) {
|
||
handleFileSelect(collected)
|
||
}
|
||
}
|
||
// Regular files → normal upload
|
||
if (regularFiles.length) {
|
||
handleFileSelect(regularFiles)
|
||
}
|
||
}
|
||
|
||
function handleDirectoryAttach(dirFiles: File[]) {
|
||
if (workerConversationReadOnly.value) return
|
||
if (!currentConversationId.value) {
|
||
newConversation()
|
||
}
|
||
for (const dir of dirFiles) {
|
||
const dirPath = (dir as any).path as string
|
||
pendingAttachments.value.push({
|
||
name: dirPath.split('/').pop() || dir.name,
|
||
size: 0,
|
||
url: '',
|
||
storedName: '',
|
||
path: dirPath,
|
||
contentType: 'inode/directory',
|
||
})
|
||
}
|
||
}
|
||
|
||
async function collectFilesFromEntries(dirEntries: FileSystemDirectoryEntry[]): Promise<File[]> {
|
||
const files: File[] = []
|
||
|
||
async function readDir(dir: FileSystemDirectoryEntry) {
|
||
const reader = dir.createReader()
|
||
let batch: FileSystemEntry[]
|
||
do {
|
||
batch = await new Promise<FileSystemEntry[]>((resolve, reject) => {
|
||
reader.readEntries(resolve, reject)
|
||
})
|
||
for (const entry of batch) {
|
||
if (entry.isFile) {
|
||
const file = await new Promise<File>((resolve, reject) => {
|
||
(entry as FileSystemFileEntry).file(resolve, reject)
|
||
})
|
||
files.push(file)
|
||
} else if (entry.isDirectory) {
|
||
await readDir(entry as FileSystemDirectoryEntry)
|
||
}
|
||
}
|
||
} while (batch.length > 0) // readEntries returns empty when done
|
||
}
|
||
|
||
for (const dir of dirEntries) {
|
||
await readDir(dir)
|
||
}
|
||
return files
|
||
}
|
||
|
||
const messageListRef = ref<InstanceType<typeof MessageList> | null>(null)
|
||
const chatInputRef = ref<InstanceType<typeof ChatInput> | null>(null)
|
||
|
||
// Post-render augmentations (ECharts, KaTeX, Mermaid) all watch the same
|
||
// MessageList container — placeholders emitted by useMarkdownRenderer get
|
||
// upgraded in place after Vue paints the rendered Markdown HTML.
|
||
const echartsContainerRef = computed(() => messageListRef.value?.$el as HTMLElement | null)
|
||
const { startObserving: startECharts, dispose: disposeECharts } = useEChartsRenderer(echartsContainerRef)
|
||
const { startObserving: startKatex, dispose: disposeKatex } = useKatexRenderer(echartsContainerRef)
|
||
const { startObserving: startMermaid, dispose: disposeMermaid } = useMermaidRenderer(echartsContainerRef)
|
||
|
||
// Last-attempt draft, restored into the input box when the SSE error event
|
||
// arrives async (sendChatMessage resolves on connect, the error fires later,
|
||
// so the catch in handleSendMessage cannot recover input by itself).
|
||
const pendingSendDraft = ref<{ input: string; attachments: any[] } | null>(null)
|
||
|
||
// 使用 useChat composable
|
||
const {
|
||
messages,
|
||
isGenerating,
|
||
streamPhase,
|
||
phaseInfo,
|
||
queuedMessage,
|
||
hasQueued,
|
||
queueSize,
|
||
heartbeat,
|
||
compactStatus,
|
||
contextUsage,
|
||
lifecycleStage,
|
||
sendMessage: sendChatMessage,
|
||
stopGeneration: stopChatGeneration,
|
||
cancelQueued,
|
||
reconnectStream: reconnectChatStream,
|
||
resetForNewConversation,
|
||
} = useChat({
|
||
baseUrl: '',
|
||
thinkingLevel,
|
||
onStreamEnd: async (meta) => {
|
||
// Restore the input/attachments if the turn ended in an error and the
|
||
// user hasn't typed something else in the meantime.
|
||
if (meta.reason === 'error' && pendingSendDraft.value) {
|
||
const draft = pendingSendDraft.value
|
||
if (!inputText.value) inputText.value = draft.input
|
||
if (pendingAttachments.value.length === 0) pendingAttachments.value = draft.attachments
|
||
}
|
||
if (meta.reason !== 'error') {
|
||
pendingSendDraft.value = null
|
||
}
|
||
// 流结束后刷新会话列表（更新 lastActiveTime / 标题等）
|
||
await loadConversations()
|
||
if (meta.conversationId && meta.conversationId === currentConversationId.value) {
|
||
// Skip DB refresh for awaiting_approval / interrupted / error:
|
||
// - awaiting_approval / interrupted: avoids overwriting local-only state
|
||
// or breaking message ordering.
|
||
// - error: the failed turn (e.g. SSE setup failure like "无权操作该会话")
|
||
// was never persisted, so refreshing would wipe the user's just-sent
|
||
// bubble and the failed assistant placeholder, leaving no trace of
|
||
// the attempt in the chat window.
|
||
if (meta.reason !== 'awaiting_approval'
|
||
&& meta.reason !== 'interrupted'
|
||
&& meta.reason !== 'error') {
|
||
await refreshCurrentConversationMessages(meta.conversationId)
|
||
}
|
||
}
|
||
},
|
||
})
|
||
|
||
const teamRunRouteQuery = computed(() => readTeamRunRouteQuery(route.query))
|
||
const metadataWorkerRunId = computed(() => {
|
||
for (let index = messages.value.length - 1; index >= 0; index -= 1) {
|
||
const metadata = parseTeamMessageMetadata(messages.value[index])
|
||
if (metadata.runId && metadata.taskId) return metadata.runId
|
||
}
|
||
return undefined
|
||
})
|
||
const linkedTeamRunId = computed(() => teamRunRouteQuery.value.teamRunId ?? metadataWorkerRunId.value)
|
||
const {
|
||
runs: teamRuns,
|
||
nextCursor: teamRunsNextCursor,
|
||
loadingMore: teamRunsLoadingMore,
|
||
loadMore: loadMoreTeamRuns,
|
||
} = useTeamRuns(currentConversationId, { linkedRunId: linkedTeamRunId })
|
||
const currentConversationKind = computed(() => conversations.value
|
||
.find(conversation => conversation.conversationId === currentConversationId.value)?.conversationKind)
|
||
const workerRouteHint = computed(() => Boolean(
|
||
teamRunRouteQuery.value.teamRunId
|
||
|| teamRunRouteQuery.value.taskId
|
||
|| currentConversationKind.value === 'team_worker'))
|
||
const workerGuard = useWorkerConversationGuard({
|
||
conversationId: currentConversationId,
|
||
workerHint: workerRouteHint,
|
||
load: async (conversationId) => {
|
||
if (isEphemeralConversation(conversationId)) return null
|
||
const query = teamRunRouteQuery.value
|
||
const response = await conversationApi.getTeamWorkerContext(conversationId, {
|
||
runId: query.teamRunId,
|
||
taskId: query.taskId,
|
||
})
|
||
return response.data ?? null
|
||
},
|
||
})
|
||
const workerRunContext = computed(() => workerGuard.context.value
|
||
?? readLegacyWorkerRouteContext(currentConversationId.value, route.query))
|
||
const workerConversationReadOnly = computed(() => workerGuard.readOnly.value)
|
||
|
||
// ============ 连接状态 ============
|
||
const connectionStatusClass = computed(() => {
|
||
if (isGenerating.value) return 'status-streaming'
|
||
if (streamPhase.value === 'failed') return 'status-error'
|
||
return 'status-idle'
|
||
})
|
||
const connectionStatusLabel = computed(() => {
|
||
if (isGenerating.value) return t('chat.status.streaming', 'Generating...')
|
||
if (streamPhase.value === 'failed') return t('chat.status.error', 'Disconnected')
|
||
return t('chat.status.idle', 'Ready')
|
||
})
|
||
|
||
// ============ 计算属性 ============
|
||
const currentAgent = computed(() => agents.value.find(a => String(a.id) === String(selectedAgentId.value)))
|
||
|
||
/** Human label for the agent's runtime mode — surfaces in the badge tooltip
|
||
* only, never in the visible header. */
|
||
const currentAgentRuntimeMode = computed(() => {
|
||
const a = currentAgent.value
|
||
if (!a) return ''
|
||
return a.agentType === 'react' ? t('agents.types.react') : t('agents.types.planExecute')
|
||
})
|
||
|
||
// Per-conversation last-viewed timestamp store (localStorage-backed, MVP).
|
||
// Keyed by conversationId. Updated when the user opens a conversation; the
|
||
// sidebar reads it (ConversationSidebar.hasUnread) to render the accent dot.
|
||
// Will move to a server-side table once we want cross-device read state.
|
||
const VIEWED_KEY_PREFIX = 'mc-conv-viewed:'
|
||
function markConversationViewed(conversationId: string | undefined, lastActiveTime?: string) {
|
||
if (!conversationId) return
|
||
const ts = lastActiveTime ? new Date(lastActiveTime).getTime() : Date.now()
|
||
try {
|
||
localStorage.setItem(VIEWED_KEY_PREFIX + conversationId, String(ts))
|
||
} catch {
|
||
// localStorage full / disabled — degrade silently; dot just stays on.
|
||
}
|
||
}
|
||
|
||
const currentRuntimeModel = computed(() => {
|
||
// Bind to the per-conversation active model — the same source the model
|
||
// selector reads — so the indicator updates the instant the user switches.
|
||
// Reading the global defaultModel here left the indicator frozen on the
|
||
// default while the selector moved.
|
||
const providerId = activeModels.value?.activeLlm?.providerId
|
||
const modelName = activeModels.value?.activeLlm?.model
|
||
if (providerId && modelName) {
|
||
const provider = providers.value.find((p) => p.id === providerId)
|
||
const all = provider ? [...(provider.models || []), ...(provider.extraModels || [])] : []
|
||
const hit = all.find((m) => m.id === modelName || m.name === modelName)
|
||
if (hit) return `${hit.name || hit.id} (${hit.id})`
|
||
// During initial hydration the safe provider projection may not be ready yet.
|
||
const em = enabledModels.value.find(
|
||
(m) => m.provider === providerId && (m.modelName === modelName || m.name === modelName)
|
||
)
|
||
if (em) return em.name ? `${em.name} (${em.modelName})` : em.modelName
|
||
return modelName
|
||
}
|
||
// No active model resolved yet — fall back to the global default, then the agent.
|
||
if (defaultModel.value?.name && defaultModel.value?.modelName) {
|
||
return `${defaultModel.value.name} (${defaultModel.value.modelName})`
|
||
}
|
||
return currentAgent.value?.modelName || 'default'
|
||
})
|
||
|
||
/**
|
||
* RFC-049 PR-1-UI: whether the active runtime model supports <em>any</em> form
|
||
* of deep thinking (OpenAI reasoning_effort / Kimi native / DeepSeek-Reasoner
|
||
* native / Anthropic extended thinking). Drives the enable/disable state of
|
||
* the thinking-depth toggle in ChatInput.
|
||
*
|
||
* Reads the broad capability (`supportsThinking`) from ProviderModelInfo,
|
||
* populated server-side in ModelInfoDTO. The narrow `supportsReasoningEffort`
|
||
* only covers OpenAI gpt-5/o1/o3/o4 and would wrongly gray out Kimi K2.x,
|
||
* DeepSeek-Reasoner, and Claude — all of which legitimately support thinking.
|
||
*/
|
||
const currentModelSupportsThinking = computed<boolean>(() => {
|
||
const providerId = activeModels.value?.activeLlm?.providerId
|
||
const modelName = activeModels.value?.activeLlm?.model
|
||
if (!providerId || !modelName) return false
|
||
const provider = providers.value.find((p) => p.id === providerId)
|
||
if (!provider) return false
|
||
const all = [...(provider.models || []), ...(provider.extraModels || [])]
|
||
const hit = all.find((m) => m.id === modelName || m.name === modelName)
|
||
return Boolean(hit?.supportsThinking)
|
||
})
|
||
|
||
const userInitial = computed(() => (localStorage.getItem('username') || 'U').charAt(0).toUpperCase())
|
||
|
||
const activeModelValue = computed(() => {
|
||
const providerId = activeModels.value?.activeLlm?.providerId
|
||
const model = activeModels.value?.activeLlm?.model
|
||
return providerId && model ? `${providerId}::${model}` : ''
|
||
})
|
||
|
||
const activeModelLabel = computed(() => {
|
||
if (!activeModelValue.value) return ''
|
||
const match = eligibleModels.value.find(m => m.value === activeModelValue.value)
|
||
if (match?.label) return match.label
|
||
// Fall back to the viewer-readable /models/enabled list while the safe
|
||
// provider projection is still hydrating, so the trigger never flashes the
|
||
// "配置模型" placeholder for an already active model.
|
||
const providerId = activeModels.value?.activeLlm?.providerId
|
||
const modelName = activeModels.value?.activeLlm?.model
|
||
if (!providerId || !modelName) return ''
|
||
const hit = enabledModels.value.find(m =>
|
||
m.provider === providerId && (m.modelName === modelName || m.name === modelName))
|
||
if (hit) return hit.name ? `${hit.name} (${hit.modelName})` : hit.modelName
|
||
return `${providerId} / ${modelName}`
|
||
})
|
||
|
||
const activeProvider = computed(() => {
|
||
const providerId = activeModels.value?.activeLlm?.providerId
|
||
return providerId ? providers.value.find((provider) => provider.id === providerId) || null : null
|
||
})
|
||
|
||
// Issue #81: liveness-aware popup state machine. modelPromptKind picks one of
|
||
// six branches; modelPromptText derives title + desc; primaryActionLabel +
|
||
// handlePrimaryAction map to the suggestedAction the backend computed.
|
||
type ModelPromptKind = 'no-active' | 'unconfigured' | 'removed' | 'cooldown' | 'unprobed' | 'no-models'
|
||
|
||
const modelPromptKind = computed<ModelPromptKind>(() => {
|
||
if (!activeModels.value?.activeLlm?.providerId) return 'no-active'
|
||
const p = activeProvider.value
|
||
if (!p) return 'no-active'
|
||
switch (p.liveness) {
|
||
case 'UNCONFIGURED': return 'unconfigured'
|
||
case 'REMOVED': return 'removed'
|
||
case 'COOLDOWN': return 'cooldown'
|
||
case 'UNPROBED': return 'unprobed'
|
||
case 'LIVE': return 'no-models'
|
||
default: return 'no-active'
|
||
}
|
||
})
|
||
|
||
const hintText = computed(() => {
|
||
const p = activeProvider.value
|
||
if (!p?.suggestedActionHintKey) return ''
|
||
return t(p.suggestedActionHintKey, (p.suggestedActionHintArgs || {}) as Record<string, unknown>)
|
||
})
|
||
|
||
const modelPromptText = computed<{ title: string; desc: string }>(() => {
|
||
const p = activeProvider.value
|
||
switch (modelPromptKind.value) {
|
||
case 'no-active':
|
||
return { title: t('chat.prompt.noActive.title'), desc: t('chat.prompt.noActive.desc') }
|
||
case 'unconfigured':
|
||
return {
|
||
title: t('chat.prompt.unconfigured.title', { name: p?.name || '' }),
|
||
desc: t('chat.prompt.unconfigured.desc', { fields: p?.missingFields || '', hint: hintText.value }),
|
||
}
|
||
case 'removed':
|
||
return {
|
||
title: t('chat.prompt.removed.title', { name: p?.name || '' }),
|
||
desc: p?.unavailableReason || t('chat.prompt.removed.descFallback'),
|
||
}
|
||
case 'cooldown':
|
||
return {
|
||
title: t('chat.prompt.cooldown.title', { name: p?.name || '' }),
|
||
desc: t('chat.prompt.cooldown.desc', {
|
||
seconds: Math.max(1, Math.ceil((p?.cooldownRemainingMs || 0) / 1000)),
|
||
}),
|
||
}
|
||
case 'unprobed':
|
||
return { title: t('chat.prompt.unprobed.title'), desc: t('chat.prompt.unprobed.desc') }
|
||
case 'no-models':
|
||
default:
|
||
return {
|
||
title: t('chat.prompt.noModels.title', { name: p?.name || '' }),
|
||
desc: t('chat.prompt.noModels.desc'),
|
||
}
|
||
}
|
||
})
|
||
|
||
const primaryActionLabel = computed(() => {
|
||
const action = activeProvider.value?.suggestedAction || 'configure_required_fields'
|
||
switch (action) {
|
||
case 'fill_base_url': return t('chat.promptAction.fillBaseUrl')
|
||
case 'fill_api_key': return t('chat.promptAction.fillApiKey')
|
||
case 'start_oauth': return t('chat.promptAction.startOAuth')
|
||
case 'test_connection': return t('chat.promptAction.testConnection')
|
||
case 'pull_model': return t('chat.promptAction.pullModel')
|
||
case 'wait_cooldown': return t('chat.promptAction.waitCooldown')
|
||
case 'reprobe': return t('chat.promptAction.reprobe')
|
||
case 'configure_required_fields':
|
||
default: return t('chat.goToModelSettings')
|
||
}
|
||
})
|
||
|
||
function handlePrimaryAction() {
|
||
goToModelSettings(activeProvider.value?.id)
|
||
}
|
||
|
||
/** First eligible model that is NOT the active one — what the secondary button switches to. */
|
||
const bestSwitchTarget = computed<{ value: string; label: string } | null>(() => {
|
||
for (const m of eligibleModels.value) {
|
||
if (m.value !== activeModelValue.value) return m
|
||
}
|
||
return null
|
||
})
|
||
|
||
/** First LIVE provider name — used by RecoverableModelBanner. */
|
||
const bestFallbackName = computed<string>(() => {
|
||
const target = bestSwitchTarget.value
|
||
if (!target) return ''
|
||
const [providerId] = target.value.split('::')
|
||
return providers.value.find(p => p.id === providerId)?.name || ''
|
||
})
|
||
|
||
function switchToBestTarget() {
|
||
const t = bestSwitchTarget.value
|
||
if (t) selectModel(t.value)
|
||
}
|
||
|
||
function onModelSelectorFix(provider: { id: string }) {
|
||
goToModelSettings(provider.id)
|
||
}
|
||
|
||
const availableProviders = computed(() =>
|
||
providers.value.filter((p) => p.available && [...(p.models || []), ...(p.extraModels || [])].length > 0)
|
||
)
|
||
|
||
const eligibleModels = computed(() => {
|
||
return availableProviders.value.flatMap((provider) => {
|
||
const allModels = [...(provider.models || []), ...(provider.extraModels || [])]
|
||
return allModels.map((model) => ({
|
||
value: `${provider.id}::${model.id}`,
|
||
label: `${provider.name} / ${model.name || model.id}`,
|
||
}))
|
||
})
|
||
})
|
||
|
||
// ============ 生命周期 ============
|
||
// Global shortcuts (Ctrl+K agents, Ctrl+N new chat) live in MainLayout so they
|
||
// work from any page; this view reacts to the dispatched event when mounted.
|
||
function handleChatShortcut(e: Event) {
|
||
const action = (e as CustomEvent).detail as 'newChat' | undefined
|
||
if (action === 'newChat') {
|
||
newConversation()
|
||
nextTick(() => chatInputRef.value?.focus?.())
|
||
}
|
||
}
|
||
|
||
// Cross-page hand-off from MainLayout's global shortcuts: read once on mount
|
||
// (before loadAgents triggers syncRouteState, which would wipe the action key)
|
||
// and apply after agents are loaded so the dropdown actually has something to show.
|
||
let pendingRouteAction: 'newChat' | '' = ''
|
||
|
||
function captureRouteAction() {
|
||
const action = route.query.action
|
||
if (action === 'newChat') {
|
||
pendingRouteAction = action
|
||
}
|
||
}
|
||
|
||
function applyPendingRouteAction() {
|
||
const action = pendingRouteAction
|
||
pendingRouteAction = ''
|
||
if (action === 'newChat') {
|
||
newConversation()
|
||
nextTick(() => chatInputRef.value?.focus?.())
|
||
}
|
||
}
|
||
|
||
// 轮询定时器：让 ChatConsole 能实时感知外部渠道（WeChat/DingTalk/…）推进来的新消息，
|
||
// 无需 F5 即可看到侧栏列表更新和选中会话的消息/流状态。
|
||
let activityPollTimer: number | null = null
|
||
// Reentrancy guard: setInterval fires every ACTIVITY_POLL_MS regardless of
|
||
// whether the previous async pollActivity has finished. If one cycle runs long
|
||
// (slow reconnectStream / sluggish backend), unguarded ticks stack up and run
|
||
// concurrently, multiplying in-flight requests. This flag keeps one cycle at a
|
||
// time — late ticks become no-ops until the running cycle returns.
|
||
let activityPolling = false
|
||
const ACTIVITY_POLL_MS = 4000
|
||
|
||
// Cron progress placeholder: when a cron job is mid-run on the currently
|
||
// visible conversation (tasks_<wsId> / cron_<id>) the assistant bubble only
|
||
// appears after T2 commits, which can be 1–5 minutes for tool-heavy ReAct
|
||
// loops. activeCronRuns is filled by the same pollActivity tick so the user
|
||
// sees a "executing…" placeholder instead of staring at a blank screen.
|
||
interface ActiveCronRun {
|
||
runId: number | string
|
||
jobId: number | string
|
||
jobName?: string
|
||
triggerType?: string
|
||
conversationId?: string
|
||
startedAt?: string
|
||
}
|
||
const activeCronRuns = ref<ActiveCronRun[]>([])
|
||
function isCronConversation(cid: string | null | undefined): boolean {
|
||
return !!cid && (cid.startsWith('tasks_') || cid.startsWith('cron_'))
|
||
}
|
||
// True when a conversation id was minted client-side for a brand-new chat
|
||
// the user hasn't sent anything in yet, so the backend has never persisted
|
||
// (and therefore can't "own") it. Every per-conversation endpoint called with
|
||
// such an id is rejected with 403 "无权访问该会话" — the status/messages poller
|
||
// and the goal loader both skip the request entirely for these ids instead of
|
||
// spamming the console with ownership errors.
|
||
//
|
||
// Conservative by construction: an id counts as ephemeral only when it is
|
||
// absent from BOTH the loaded conversation list AND the local message buffer,
|
||
// so a freshly-persisted conversation is never mistaken for one. At worst a
|
||
// single stray 403 slips through during the persist race, which the leaf
|
||
// callers swallow.
|
||
function isEphemeralConversation(cid: string | null | undefined): boolean {
|
||
if (!cid) return true
|
||
// Server-managed virtual conversations (cron / scheduled tasks) always exist.
|
||
if (cid.startsWith('tasks_') || cid.startsWith('cron_')) return false
|
||
if (conversations.value.some((c) => c.conversationId === cid)) return false
|
||
return !messages.value.some((m: any) => m.conversationId === cid)
|
||
}
|
||
async function refreshActiveCronRuns(cid: string) {
|
||
if (!isCronConversation(cid)) {
|
||
activeCronRuns.value = []
|
||
return
|
||
}
|
||
try {
|
||
const res: any = await cronJobApi.activeRuns(cid)
|
||
if (currentConversationId.value !== cid) return
|
||
const next: ActiveCronRun[] = res?.data ?? []
|
||
const wasRunning = activeCronRuns.value.length > 0
|
||
activeCronRuns.value = next
|
||
// Transition from "had runs" to "no runs" → assistant bubble was just
|
||
// persisted by T2; fetch messages so it shows up without waiting for the
|
||
// next pollActivity tick to align.
|
||
if (wasRunning && next.length === 0) {
|
||
await refreshCurrentConversationMessages(cid)
|
||
}
|
||
} catch {
|
||
// Network blip — keep the previous state, next tick will retry.
|
||
}
|
||
}
|
||
// Reactive ticker so the elapsed label updates without depending on a poll.
|
||
const elapsedNow = ref(Date.now())
|
||
let elapsedTickTimer: number | null = null
|
||
function elapsedLabel(startedAt?: string): string {
|
||
if (!startedAt) return ''
|
||
const ms = elapsedNow.value - new Date(startedAt).getTime()
|
||
if (ms < 0 || !Number.isFinite(ms)) return ''
|
||
const sec = Math.floor(ms / 1000)
|
||
if (sec < 60) return `${sec}s`
|
||
const min = Math.floor(sec / 60)
|
||
const rem = sec % 60
|
||
return `${min}m${rem > 0 ? rem + 's' : ''}`
|
||
}
|
||
|
||
/**
|
||
* 判断当前消息列表的末尾是不是一条"本地仅有的失败气泡"。
|
||
* 典型场景：SSE setup 阶段就抛错（如"无权操作该会话"），
|
||
* 这次 turn 的 user / assistant 消息从未持久化进 DB。
|
||
* 数据库快照不知道它们存在，pollActivity 的对齐会把它们冲掉。
|
||
*
|
||
* 识别条件：末尾 assistant 状态为 failed、带 errorInfo、且 id 不是 DB 数值 id（client uuid）。
|
||
*/
|
||
function hasLocalOnlyFailedTail(): boolean {
|
||
const last = messages.value[messages.value.length - 1] as any
|
||
if (!last || last.role !== 'assistant') return false
|
||
if (last.status !== 'failed') return false
|
||
if (!last.errorInfo) return false
|
||
return !/^\d+$/.test(String(last.id))
|
||
}
|
||
|
||
async function pollActivity() {
|
||
// 页面不可见时不轮询，避免切到别的标签还在空耗
|
||
if (typeof document !== 'undefined' && document.hidden) return
|
||
// Skip when the previous cycle is still running so slow polls can't stack.
|
||
if (activityPolling) return
|
||
activityPolling = true
|
||
try {
|
||
try {
|
||
await loadConversations()
|
||
reconcileCurrentConversation()
|
||
} catch {
|
||
// 静默失败，下一轮再试
|
||
}
|
||
// 自己没在生成时才刷新当前选中会话的消息 + 探测是否该接入流。
|
||
// 跳过尚未落库的临时会话(新建空会话、还没发消息):后端查不到会话行,
|
||
// getStatus / listMessages 一律 403,4s 轮询会把 console 刷爆(见 issue #408)。
|
||
if (currentConversationId.value
|
||
&& !isGenerating.value
|
||
&& streamPhase.value !== 'awaiting_approval'
|
||
&& !isEphemeralConversation(currentConversationId.value)) {
|
||
const cid = currentConversationId.value
|
||
try {
|
||
const statusRes: any = await conversationApi.getStatus(cid)
|
||
if (currentConversationId.value !== cid) return
|
||
const running = statusRes?.data?.streamStatus === 'running'
|
||
if (running) {
|
||
// 外部渠道正在跑：
|
||
// 1. 先从 DB 拉消息，把刚插入的 user 消息（"你在干什么"之类）带进来，
|
||
// 否则只接入流的话前端只能看到 assistant content_delta，看不到用户问题。
|
||
// 2. 再接入流，让后续 content_delta 实时累积到 assistant 气泡。
|
||
await refreshCurrentConversationMessages(cid)
|
||
if (currentConversationId.value !== cid || isGenerating.value) return
|
||
// Match selectConversation's behavior: a stale scroll-escape from an
|
||
// earlier upward scroll must not suppress follow-scroll on reconnect.
|
||
messageListRef.value?.resetScrollLock()
|
||
await reconnectStream(cid)
|
||
} else if (!hasLocalOnlyFailedTail()) {
|
||
// 不在跑：从 DB 对齐消息（新 user 消息 / 刚落库 assistant 会合并进来）。
|
||
// 但若末尾是本地失败气泡（SSE setup 失败一类，后端从未持久化过），
|
||
// 就跳过对齐 —— 不然这次的 user/失败 assistant 会被 DB 快照覆盖掉，
|
||
// 用户除了上面的 toast 看不到任何痕迹。
|
||
await refreshCurrentConversationMessages(cid)
|
||
}
|
||
} catch {
|
||
// 忽略探测失败
|
||
}
|
||
// Cron progress placeholder — independent of streamStatus because cron
|
||
// runs use the non-streaming chat() path, so streamStatus stays idle.
|
||
await refreshActiveCronRuns(cid)
|
||
}
|
||
} finally {
|
||
activityPolling = false
|
||
}
|
||
}
|
||
|
||
onMounted(async () => {
|
||
captureRouteAction()
|
||
window.addEventListener('mc:chat-shortcut', handleChatShortcut)
|
||
document.addEventListener('click', handleCodeCopy)
|
||
startECharts()
|
||
startKatex()
|
||
startMermaid()
|
||
await Promise.all([loadAgents(), loadModelState(), loadConversations()])
|
||
await hydrateStateFromRoute()
|
||
applyPendingRouteAction()
|
||
activityPollTimer = window.setInterval(pollActivity, ACTIVITY_POLL_MS)
|
||
elapsedTickTimer = window.setInterval(() => {
|
||
if (activeCronRuns.value.length > 0) elapsedNow.value = Date.now()
|
||
}, 1000)
|
||
})
|
||
|
||
onBeforeUnmount(() => {
|
||
window.removeEventListener('mc:chat-shortcut', handleChatShortcut)
|
||
document.removeEventListener('click', handleCodeCopy)
|
||
disposeECharts()
|
||
disposeKatex()
|
||
disposeMermaid()
|
||
if (activityPollTimer !== null) {
|
||
clearInterval(activityPollTimer)
|
||
activityPollTimer = null
|
||
}
|
||
if (elapsedTickTimer !== null) {
|
||
clearInterval(elapsedTickTimer)
|
||
elapsedTickTimer = null
|
||
}
|
||
// Switching tabs / route changes / mouse-detach unmount this component, but the
|
||
// backend agent should keep running so the user can reconnect later. Use
|
||
// resetForNewConversation (front-end SSE disconnect only) instead of
|
||
// stopChatGeneration which would POST /stop and abort the in-flight turn.
|
||
resetForNewConversation()
|
||
// 释放所有附件的 ObjectURL，防止内存泄漏
|
||
revokeAllPreviewUrls()
|
||
})
|
||
|
||
onDeactivated(() => {
|
||
if (activityPollTimer !== null) {
|
||
clearInterval(activityPollTimer)
|
||
activityPollTimer = null
|
||
}
|
||
if (elapsedTickTimer !== null) {
|
||
clearInterval(elapsedTickTimer)
|
||
elapsedTickTimer = null
|
||
}
|
||
document.removeEventListener('click', handleCodeCopy)
|
||
disposeECharts()
|
||
disposeKatex()
|
||
disposeMermaid()
|
||
revokeAllPreviewUrls()
|
||
resetForNewConversation()
|
||
})
|
||
|
||
onActivated(async () => {
|
||
document.addEventListener('click', handleCodeCopy)
|
||
startECharts()
|
||
startKatex()
|
||
startMermaid()
|
||
// Issue #538: ChatConsole is kept alive by the router (meta.keepAlive), so
|
||
// navigating away and back only fires onActivated, not onMounted — an
|
||
// agent created/edited/deleted elsewhere (e.g. the Employees page) never
|
||
// reached this component's own `agents` list otherwise, and stayed
|
||
// invisible in the picker until a full page reload forced a fresh mount.
|
||
await loadAgents()
|
||
activityPollTimer = window.setInterval(pollActivity, ACTIVITY_POLL_MS)
|
||
elapsedTickTimer = window.setInterval(() => {
|
||
if (activeCronRuns.value.length > 0) elapsedNow.value = Date.now()
|
||
}, 1000)
|
||
// 登出已改为 window.location.href（刷新页面），所以切回时不会有跨用户残留
|
||
if (currentConversationId.value && !isEphemeralConversation(currentConversationId.value)) {
|
||
try {
|
||
const statusRes: any = await conversationApi.getStatus(currentConversationId.value)
|
||
if (currentConversationId.value && statusRes.data?.streamStatus === 'running') {
|
||
await reconnectStream(currentConversationId.value)
|
||
}
|
||
} catch {
|
||
// 忽略
|
||
}
|
||
}
|
||
})
|
||
|
||
watch(() => route.query, () => {
|
||
// If a fresh action arrives (e.g. user re-fires Ctrl+K via the URL while
|
||
// the view is already alive), pick it up immediately.
|
||
captureRouteAction()
|
||
if (pendingRouteAction) applyPendingRouteAction()
|
||
void hydrateStateFromRoute()
|
||
})
|
||
|
||
watch([selectedAgentId, currentConversationId], () => {
|
||
syncRouteState()
|
||
})
|
||
|
||
// Load the active goal whenever the user switches conversation. The
|
||
// avatar ring listens on goalStore.activeGoalByConv[cid]; without this
|
||
// fetch the ring would only appear after an SSE event mutated the store.
|
||
const goalStore = useGoalStore()
|
||
const workspaceStore = useWorkspaceStore()
|
||
const currentWorkspaceId = computed(() => workspaceStore.currentWorkspaceId ?? '1')
|
||
const canConfigureModels = computed(() => workspaceStore.isGlobalAdmin)
|
||
const modelSelectorEmptyHint = computed(() => canConfigureModels.value
|
||
? undefined
|
||
: t('chat.noModelsAvailableContactAdmin'))
|
||
watch(currentConversationId, async (cid) => {
|
||
// Skip un-persisted conversations: a brand-new empty chat has no goal yet
|
||
// and the lookup would only 403 (Not the owner). The ring is hydrated by the
|
||
// goal_created SSE event once the first turn lands.
|
||
if (cid && !isEphemeralConversation(cid)) {
|
||
await goalStore.loadActiveForConversation(cid)
|
||
}
|
||
}, { immediate: true })
|
||
|
||
// Re-fetch the active goal when a turn finishes. A goal can be created or
|
||
// mutated mid-conversation — e.g. auto-derived server-side from a Plan-Execute
|
||
// plan, or completed by the agent — without the conversation id changing and
|
||
// without a goal_* SSE event reaching this client (skipped creation, missed
|
||
// event, reconnect). The per-conversation watch above only fires on switch, so
|
||
// this transition-to-idle refresh is what keeps the goal ring honest after
|
||
// every turn against the persisted truth.
|
||
watch(isGenerating, async (generating, wasGenerating) => {
|
||
if (wasGenerating && !generating
|
||
&& currentConversationId.value
|
||
&& !isEphemeralConversation(currentConversationId.value)) {
|
||
await goalStore.loadActiveForConversation(currentConversationId.value)
|
||
}
|
||
})
|
||
|
||
// Derive props for the inline prompt + system-line slots that sit
|
||
// between MessageList and ChatInput. The prompt shows only when:
|
||
// 1) there's a current conversation, agent, and at least one assistant
|
||
// reply (otherwise the prompt is premature);
|
||
// 2) there's no active goal (the ring already covers active state);
|
||
// 3) the user hasn't dismissed the prompt on this conv;
|
||
// 4) we're not mid-stream (don't pop suggestions while the agent
|
||
// is still typing).
|
||
const goalTerminalForCurrent = computed(() =>
|
||
currentConversationId.value
|
||
? goalStore.recentTerminal(currentConversationId.value)
|
||
: null,
|
||
)
|
||
const goalSystemLineTitle = computed(() => {
|
||
const t = goalTerminalForCurrent.value
|
||
if (!t) return ''
|
||
// The leading icon is owned by GoalSystemLine (✦ / ⚠) so we don't
|
||
// prepend one here — doing so produced "✦ 🎉 …" double-glyph titles.
|
||
return t.status === 'completed' ? `目标达成 · ${t.title}` : `这次的预算用完了 · ${t.title}`
|
||
})
|
||
const goalSystemLineDetail = computed(() => {
|
||
const t = goalTerminalForCurrent.value
|
||
if (!t) return ''
|
||
if (t.status === 'completed') {
|
||
return t.score != null
|
||
? `已完成 · final score ${t.score.toFixed(2)}`
|
||
: '已完成 · 总结已存入长期记忆'
|
||
}
|
||
// exhausted
|
||
if (t.reason === 'turn_budget') return '预算轮数用完。'
|
||
if (t.reason === 'llm_call_budget') return 'LLM 调用预算用完。'
|
||
return '预算耗尽。'
|
||
})
|
||
|
||
// True when the current conversation's message stream already contains a
|
||
// setGoal tool call — authoritative even before the goal_created SSE event
|
||
// updates the goal store.
|
||
const goalSetInStream = computed(() => {
|
||
return messages.value.some((m) => {
|
||
if (m.role !== 'assistant') return false
|
||
const tcs: any = (m as any).metadata?.toolCalls
|
||
return Array.isArray(tcs) && tcs.some((tc: any) => tc?.name === 'setGoal')
|
||
})
|
||
})
|
||
|
||
const showGoalSetPrompt = computed(() => {
|
||
if (!currentConversationId.value || !selectedAgentId.value) return false
|
||
if (isGenerating.value) return false
|
||
// Active goal? The ring covers that — no need for a prompt.
|
||
if (goalStore.activeGoal(currentConversationId.value)) return false
|
||
// A goal was already set this conversation via the setGoal tool — even if
|
||
// the goal_created/goal_evaluated SSE event hasn't updated the store yet
|
||
// (the turn can end a render frame before that event lands). Reading the
|
||
// stream directly closes that flash window: never offer to set a goal when
|
||
// the agent already set one here.
|
||
if (goalSetInStream.value) return false
|
||
// Recent terminal still showing? Let the user dismiss that first.
|
||
if (goalTerminalForCurrent.value) return false
|
||
if (goalStore.isPromptDismissed(currentConversationId.value)) return false
|
||
// Need at least one user → assistant exchange so the prompt has
|
||
// context to derive a suggested title from.
|
||
const hasAssistantReply = messages.value.some(m => m.role === 'assistant')
|
||
if (!hasAssistantReply) return false
|
||
// Heuristic: don't claim "this looks multi-turn" without evidence. Fire
|
||
// the prompt only when at least one signal of a real ongoing task is
|
||
// present. Without these the prompt fires after one-shot Q&A like
|
||
// "三句话告诉我 X" and the copy lies to the user.
|
||
const userTurns = messages.value.filter(m => m.role === 'user').length
|
||
if (userTurns >= 2) return true // multiple user messages = ongoing thread
|
||
// Single-turn case: only suggest if the agent did non-trivial work.
|
||
return messages.value.some(m => {
|
||
if (m.role !== 'assistant') return false
|
||
const md: any = (m as any).metadata
|
||
if (!md) return false
|
||
// Tool calls — strongest signal of real work.
|
||
if (Array.isArray(md.toolCalls) && md.toolCalls.length > 0) return true
|
||
// Plan steps — Plan-Execute agent ran a multi-step plan.
|
||
if (md.plan && Array.isArray(md.plan.steps) && md.plan.steps.length > 1) return true
|
||
// Multiple ReAct iterations / segments — agent looped.
|
||
if (Array.isArray(md.segments) && md.segments.length > 3) return true
|
||
return false
|
||
})
|
||
})
|
||
|
||
// Build a sensible default title from the conversation's first user
|
||
// message. The user can always edit later via the goal page.
|
||
const goalSuggestedTitle = computed(() => {
|
||
const firstUser = messages.value.find(m => m.role === 'user')
|
||
const raw = (firstUser?.content || '').trim()
|
||
if (!raw) return '新目标'
|
||
// 80 char clip mirrors GoalController.create validation.
|
||
return raw.length > 80 ? raw.slice(0, 77) + '...' : raw
|
||
})
|
||
|
||
function onGoalPromptDismiss() {
|
||
if (currentConversationId.value) {
|
||
goalStore.dismissPrompt(currentConversationId.value)
|
||
}
|
||
}
|
||
|
||
function onGoalSystemLineDismiss() {
|
||
if (currentConversationId.value) {
|
||
goalStore.clearRecentTerminal(currentConversationId.value)
|
||
}
|
||
}
|
||
|
||
// Refetch agent capabilities (modalities + sidecar config) on agent change so
|
||
// the multimodal routing hint above the input box can react synchronously when
|
||
// the user attaches an image / video.
|
||
watch(selectedAgentId, async (id) => {
|
||
if (!id) { agentCapabilities.value = null; return }
|
||
try {
|
||
const res: any = await agentApi.getCapabilities(id)
|
||
agentCapabilities.value = res.data || null
|
||
// The capability fetch is async and typically resolves AFTER the
|
||
// synchronous newConversation()/applyConversationModel() that ran on this
|
||
// same agent switch (which fell back to the global default because caps
|
||
// weren't loaded yet). Re-seed now that we know the agent's resolved model
|
||
// — but only for a conversation with no server-side pin and where the user
|
||
// hasn't manually picked a model, so we never clobber an explicit choice.
|
||
const conv = conversations.value.find(c => c.conversationId === currentConversationId.value)
|
||
const hasServerPin = !!(conv?.modelProvider && conv?.modelName)
|
||
if (!hasServerPin && !userPickedModel.value) {
|
||
applyConversationModel(conv)
|
||
}
|
||
} catch {
|
||
agentCapabilities.value = null
|
||
}
|
||
}, { immediate: true })
|
||
|
||
// ============ 方法 ============
|
||
async function loadAgents() {
|
||
try {
|
||
// Hide disabled agents from the picker — they cannot be chatted with
|
||
// (the chat endpoints reject disabled agents), so showing them invites
|
||
// a confusing failure path. The admin Agents view passes no filter.
|
||
const res: any = await agentApi.list({ enabled: true })
|
||
agents.value = res.data || []
|
||
if (agents.value.length > 0) cachedAgents = [...agents.value]
|
||
// 只有在 URL 没有指定 agentId 且当前无选中时，才默认选第一个
|
||
if (agents.value.length > 0 && !selectedAgentId.value && !route.query.agentId) {
|
||
selectedAgentId.value = agents.value[0].id
|
||
}
|
||
} catch (e) {
|
||
mcToast.error(t('chat.loadAgentsFailed'))
|
||
}
|
||
}
|
||
|
||
async function loadModelState() {
|
||
// /default + /active + /enabled are viewer-accessible and required to chat.
|
||
// /models (provider list) is admin-only because it returns API keys + base
|
||
// URLs. Viewers join /models/options with /models/enabled for the selector,
|
||
// but still skip the liveness banner because neither safe response exposes
|
||
// runtime diagnostics.
|
||
try {
|
||
const providerOptionsRequest = workspaceStore.isGlobalAdmin
|
||
? Promise.resolve({ data: [] })
|
||
: modelApi.listProviderOptions()
|
||
const [defaultRes, activeRes, enabledRes, providerOptionsRes]: any = await Promise.all([
|
||
modelApi.getDefault(),
|
||
modelApi.getActive(),
|
||
modelApi.listEnabled(),
|
||
providerOptionsRequest,
|
||
])
|
||
defaultModel.value = defaultRes.data || null
|
||
const ga = activeRes.data?.activeLlm
|
||
globalDefaultModel.value = ga?.providerId && ga?.model
|
||
? { providerId: ga.providerId, model: ga.model }
|
||
: null
|
||
// Seed the selector when no conversation has set it yet (fresh chat, or
|
||
// before a conversation is selected). A conversation that already has a
|
||
// model keeps it — selectConversation/applyConversationModel own that.
|
||
if (!activeModels.value && globalDefaultModel.value) {
|
||
activeModels.value = { activeLlm: { ...globalDefaultModel.value } }
|
||
}
|
||
enabledModels.value = enabledRes.data || []
|
||
if (!workspaceStore.isGlobalAdmin) {
|
||
providers.value = buildViewerModelProviders(
|
||
providerOptionsRes.data || [],
|
||
enabledModels.value,
|
||
)
|
||
// The safe projection intentionally has no liveness diagnostics. Trust
|
||
// the active model and let the runtime report a real call failure.
|
||
providersUnavailable.value = true
|
||
recomputePromptFlags()
|
||
return
|
||
}
|
||
} catch (e) {
|
||
mcToast.error(t('chat.loadModelFailed'))
|
||
blockingPrompt.value = true
|
||
recoverablePrompt.value = false
|
||
return
|
||
}
|
||
try {
|
||
const providersRes: any = await modelApi.listProviders()
|
||
providers.value = providersRes.data || []
|
||
providersUnavailable.value = false
|
||
} catch (e: any) {
|
||
if (e?.response?.status === 403) {
|
||
providers.value = []
|
||
providersUnavailable.value = true
|
||
} else {
|
||
// Non-403 failure is still a real problem worth surfacing.
|
||
mcToast.error(t('chat.loadModelFailed'))
|
||
}
|
||
}
|
||
recomputePromptFlags()
|
||
}
|
||
|
||
/**
|
||
* Issue #81 v2 R2: derive blocking / recoverable prompt flags from the current
|
||
* providers + active model snapshot. Called from loadModelState after every
|
||
* /providers refresh, and from a watcher when the user switches model. The
|
||
* runtime fallback chain in NodeStreamingChatHelper picks the first LIVE
|
||
* provider regardless of which one is "active", so as long as ANY provider is
|
||
* LIVE we should NOT block — we just hint with a banner.
|
||
*/
|
||
function recomputePromptFlags() {
|
||
const active = activeModels.value?.activeLlm
|
||
if (!active?.providerId || !active?.model) {
|
||
blockingPrompt.value = true
|
||
recoverablePrompt.value = false
|
||
recoverableDismissed.value = false
|
||
return
|
||
}
|
||
// Viewer-level users cannot read the provider list (admin-only because it
|
||
// returns API keys), so we can't compute liveness. Trust the active model
|
||
// and let the agent runtime surface any per-call failure instead of
|
||
// blocking the entire chat surface.
|
||
if (providersUnavailable.value) {
|
||
blockingPrompt.value = false
|
||
recoverablePrompt.value = false
|
||
recoverableDismissed.value = false
|
||
return
|
||
}
|
||
const ap = providers.value.find(p => p.id === active.providerId) || null
|
||
const apHasModels = ap
|
||
? ((ap.models?.length || 0) + (ap.extraModels?.length || 0)) > 0
|
||
: false
|
||
const activeUsable = ap?.liveness === 'LIVE' && apHasModels
|
||
if (activeUsable) {
|
||
blockingPrompt.value = false
|
||
recoverablePrompt.value = false
|
||
recoverableDismissed.value = false
|
||
return
|
||
}
|
||
const anyUsable = providers.value.some(p =>
|
||
p.liveness === 'LIVE'
|
||
&& ((p.models?.length || 0) + (p.extraModels?.length || 0)) > 0)
|
||
blockingPrompt.value = !anyUsable
|
||
recoverablePrompt.value = anyUsable && !recoverableDismissed.value
|
||
}
|
||
|
||
// Issue #81 v2 R2: keep blocking/recoverable in sync with the providers list
|
||
// and the active model selection without forcing every mutation site to call
|
||
// recomputePromptFlags() manually.
|
||
watch([providers, activeModels], recomputePromptFlags, { deep: true })
|
||
|
||
async function loadConversations() {
|
||
try {
|
||
const res: any = await conversationApi.list()
|
||
conversations.value = res.data || []
|
||
} catch (e) {
|
||
mcToast.error(t('chat.loadConversationsFailed'))
|
||
}
|
||
}
|
||
|
||
async function refreshCurrentConversationMessages(conversationId: string) {
|
||
if (!conversationId) return
|
||
if (isGenerating.value) return
|
||
if (streamPhase.value === 'awaiting_approval') return
|
||
try {
|
||
const res: any = await conversationApi.listMessages(conversationId)
|
||
// Stale guard：await 返回后确认仍是当前会话
|
||
if (currentConversationId.value !== conversationId) return
|
||
// 二次 isGenerating 检查：如果 await 期间用户已发新消息，不覆盖本地状态
|
||
if (isGenerating.value) return
|
||
const fetched = extractMessages(res).messages.map((msg: Message) => normalizeMessage(msg))
|
||
// 严格过滤：只保留 conversationId 完全匹配的本地消息，orphan（空 conversationId）直接丢弃
|
||
const currentMessages = messages.value.filter(
|
||
(m: any) => m.conversationId === conversationId
|
||
)
|
||
messages.value = reconcileMessages(currentMessages, fetched)
|
||
} catch (e) {
|
||
console.warn('[ChatConsole] Failed to refresh current conversation messages:', e)
|
||
}
|
||
}
|
||
|
||
async function hydrateStateFromRoute() {
|
||
const { agentId, conversationId } = resolveRouteHydrationQuery({
|
||
routeAgentId: route.query.agentId ? String(route.query.agentId) : '',
|
||
routeConversationId: String(route.query.conversationId || ''),
|
||
agents: agents.value,
|
||
conversations: conversations.value,
|
||
})
|
||
|
||
if (agentId && agentId !== String(selectedAgentId.value)) {
|
||
selectedAgentId.value = agentId
|
||
}
|
||
|
||
if (conversationId && conversationId !== currentConversationId.value) {
|
||
const matchedConversation = conversations.value.find(conv => conv.conversationId === conversationId)
|
||
if (matchedConversation) {
|
||
await selectConversation(matchedConversation, agentId)
|
||
} else {
|
||
// 会话不在已加载列表中（可能来自 Sessions 页面跳转），尝试加载消息
|
||
currentConversationId.value = conversationId
|
||
messages.value = []
|
||
try {
|
||
const res: any = await conversationApi.listMessages(conversationId)
|
||
if (currentConversationId.value !== conversationId) return
|
||
messages.value = extractMessages(res).messages.map((msg: Message) => normalizeMessage(msg, true))
|
||
} catch {
|
||
// 消息加载失败，保持空
|
||
}
|
||
try {
|
||
if (currentConversationId.value !== conversationId) return
|
||
const statusRes: any = await conversationApi.getStatus(conversationId)
|
||
if (currentConversationId.value === conversationId && statusRes.data?.streamStatus === 'running') {
|
||
await reconnectStream(conversationId)
|
||
}
|
||
} catch {
|
||
// 忽略
|
||
}
|
||
}
|
||
}
|
||
|
||
// 如果仍然没选中 agent，默认选第一个
|
||
if (!selectedAgentId.value && agents.value.length > 0) {
|
||
selectedAgentId.value = agents.value[0].id
|
||
}
|
||
}
|
||
|
||
function syncRouteState() {
|
||
const query = buildChatRouteQuery({
|
||
currentQuery: route.query,
|
||
agentId: selectedAgentId.value ? String(selectedAgentId.value) : undefined,
|
||
conversationId: currentConversationId.value || undefined,
|
||
})
|
||
router.replace({ path: '/chat', query })
|
||
}
|
||
|
||
async function selectConversation(conv: Conversation, routeAgentId = '') {
|
||
if (isMobile.value) convPanelOpen.value = false
|
||
// 切换到不同会话：只清理本地 UI/SSE（resetForNewConversation 会 stream.disconnect + 清变量），
|
||
// 但不 POST /chat/{A}/stop —— 让 A 的后台 agent run 跑到完成。
|
||
// 用户之后回到 A：pollActivity / selectConversation 的 /status 探测会自动 reconnect 接回实时流；
|
||
// 若 A 已完成，refreshCurrentConversationMessages 会从 DB 拉完整结果。
|
||
// 点同一个会话则完全不 reset，避免打断正在观察的流。
|
||
const switchingAway = currentConversationId.value !== conv.conversationId
|
||
if (switchingAway) {
|
||
resetForNewConversation()
|
||
messageListRef.value?.resetScrollLock()
|
||
}
|
||
currentConversationId.value = conv.conversationId
|
||
selectedAgentId.value = resolveConversationAgentSelection({
|
||
routeAgentId,
|
||
conversationAgentId: conv.agentId,
|
||
currentAgentId: selectedAgentId.value,
|
||
})
|
||
// Opening another conversation: its pin (or the agent/global fallback) is
|
||
// authoritative, so clear the previous conversation's manual-pick guard.
|
||
userPickedModel.value = false
|
||
// Restore this conversation's pinned model into the selector.
|
||
applyConversationModel(conv)
|
||
// Reset cron placeholder state up front; the immediate fetch below repopulates
|
||
// it for cron conversations so the user doesn't wait up to 4s for the next tick.
|
||
activeCronRuns.value = []
|
||
if (isCronConversation(conv.conversationId)) {
|
||
void refreshActiveCronRuns(conv.conversationId)
|
||
}
|
||
// Mark as read when opened — clears the unread dot on tasks_<wsId> after
|
||
// the user actually visits the cron output. localStorage-only for MVP;
|
||
// server-side last-viewed table is a future enhancement.
|
||
markConversationViewed(conv.conversationId, conv.lastActiveTime)
|
||
const requestedConvId = conv.conversationId
|
||
try {
|
||
const res: any = await conversationApi.listMessages(requestedConvId)
|
||
// Stale guard：await 返回后确认仍是当前会话，否则丢弃
|
||
if (currentConversationId.value !== requestedConvId) return
|
||
// 点同一个会话时，若已有 SSE 在跑就不要覆盖本地消息状态
|
||
if (switchingAway || !isGenerating.value) {
|
||
const convRunning = conv.streamStatus === 'running'
|
||
messages.value = extractMessages(res).messages.map((msg: Message) => normalizeMessage(msg, convRunning))
|
||
}
|
||
|
||
// Hydrate pending approvals：恢复刷新后丢失的审批卡片（RFC-067 §4.9）
|
||
//
|
||
// Two-way reconciliation between the server's pending list and each
|
||
// message's metadata.pendingApproval:
|
||
// 1. Forward — server pending → align onto the message that already
|
||
// carries the same pendingId (so multi-pending convs don't have
|
||
// every banner overwrite the same row); fallback to last assistant
|
||
// only when no message has that id yet.
|
||
// 2. Reverse — local message metadata still says pending_approval but
|
||
// the server no longer lists that pendingId → flip to 'expired'
|
||
// locally. This closes the GC/timeout loop without requiring an
|
||
// extra server-side broadcast: the next refresh sees a clean state.
|
||
try {
|
||
const approvalRes: any = await chatApi.getPendingApprovals(requestedConvId)
|
||
if (currentConversationId.value !== requestedConvId) return
|
||
const pendingApprovals: any[] = approvalRes.data || []
|
||
|
||
// Index existing messages by their embedded pendingId (assistant only).
|
||
const indexById = new Map<string, Message>()
|
||
for (const m of messages.value) {
|
||
if (m.role !== 'assistant') continue
|
||
const pid = (m as any).metadata?.pendingApproval?.pendingId
|
||
if (typeof pid === 'string' && pid) indexById.set(pid, m)
|
||
}
|
||
|
||
// Forward direction: align server-known pending onto its owning message.
|
||
for (const pa of pendingApprovals) {
|
||
const enriched = {
|
||
pendingId: pa.pendingId,
|
||
toolName: pa.toolName,
|
||
arguments: pa.toolArguments,
|
||
reason: pa.reason,
|
||
status: 'pending_approval' as const,
|
||
findings: pa.findingsJson ? JSON.parse(pa.findingsJson) : undefined,
|
||
maxSeverity: pa.maxSeverity || undefined,
|
||
summary: pa.summary || undefined,
|
||
}
|
||
const target = indexById.get(pa.pendingId)
|
||
if (target) {
|
||
(target as any).metadata = {
|
||
...(target as any).metadata,
|
||
currentPhase: 'awaiting_approval',
|
||
pendingApproval: enriched,
|
||
}
|
||
} else {
|
||
// Fallback: no message in the loaded history claims this pendingId
|
||
// (typical when the assistant message hasn't been persisted yet —
|
||
// e.g., approval fired before doOnComplete). Append to the last
|
||
// assistant; same as pre-RFC behavior, but logged so a regression
|
||
// where multiple unmatched pendings collide is observable.
|
||
const assistantMessages = messages.value.filter(m => m.role === 'assistant')
|
||
const lastAssistant = assistantMessages[assistantMessages.length - 1]
|
||
if (lastAssistant) {
|
||
console.warn('[hydrate] pendingId %s has no owning message — falling back to last assistant', pa.pendingId)
|
||
;(lastAssistant as any).metadata = {
|
||
...(lastAssistant as any).metadata,
|
||
currentPhase: 'awaiting_approval',
|
||
pendingApproval: enriched,
|
||
}
|
||
}
|
||
}
|
||
}
|
||
|
||
// Reverse direction: any local pending_approval whose pendingId is not
|
||
// in the server's list got resolved (timeout / consume) without a UI
|
||
// event — flip to expired so MessageBubble hides the banner.
|
||
const serverIds = new Set<string>(pendingApprovals.map((p: any) => p.pendingId))
|
||
for (const m of messages.value) {
|
||
if (m.role !== 'assistant') continue
|
||
const meta = (m as any).metadata
|
||
const local = meta?.pendingApproval
|
||
if (local?.status === 'pending_approval'
|
||
&& local.pendingId
|
||
&& !serverIds.has(local.pendingId)) {
|
||
(m as any).metadata = {
|
||
...meta,
|
||
pendingApproval: { ...local, status: 'expired' },
|
||
}
|
||
}
|
||
}
|
||
} catch {
|
||
// hydration 失败不影响正常使用
|
||
}
|
||
|
||
// 决定是否重连 SSE：
|
||
// - 快照 streamStatus==='running' → 直接重连
|
||
// - 否则探测实时状态（兜底处理：渠道消息进入后侧栏快照未刷新时，仍能接入运行中的流）
|
||
let shouldReconnect = conv.streamStatus === 'running'
|
||
if (!shouldReconnect) {
|
||
try {
|
||
const statusRes: any = await conversationApi.getStatus(requestedConvId)
|
||
if (currentConversationId.value !== requestedConvId) return
|
||
shouldReconnect = statusRes?.data?.streamStatus === 'running'
|
||
} catch {
|
||
// 探测失败不阻断主流程
|
||
}
|
||
}
|
||
if (currentConversationId.value === requestedConvId && shouldReconnect) {
|
||
await reconnectStream(requestedConvId)
|
||
}
|
||
} catch (e) {
|
||
mcToast.error(t('chat.loadMessagesFailed'))
|
||
}
|
||
}
|
||
|
||
function newConversation() {
|
||
// Creating a new chat is just local navigation. Keep any previous backend
|
||
// run alive so the user can return and reconnect to it later.
|
||
resetForNewConversation()
|
||
currentConversationId.value = `conv_${Date.now()}_${Math.random().toString(36).slice(2, 8)}`
|
||
messages.value = []
|
||
// A fresh conversation defers to the selected agent's model (then the global
|
||
// default) until the user explicitly picks one.
|
||
userPickedModel.value = false
|
||
applyConversationModel()
|
||
}
|
||
|
||
// The sidebar performs the delete API call(s) and emits the removed ids.
|
||
// Drop them from the local list and reset the chat area if the conversation
|
||
// currently open was among those deleted.
|
||
function onConversationsDeleted(ids: string[]) {
|
||
conversations.value = conversations.value.filter(c => !ids.includes(c.conversationId))
|
||
if (ids.includes(currentConversationId.value)) {
|
||
resetStreamingState()
|
||
messages.value = []
|
||
currentConversationId.value = ''
|
||
// Deleting the open conversation drops us back to a blank slate — clear the
|
||
// manual-pick guard so the next message (which mints a fresh conversation)
|
||
// seeds from the agent override again instead of inheriting the deleted
|
||
// conversation's pick. Mirrors newConversation()/selectConversation().
|
||
userPickedModel.value = false
|
||
}
|
||
}
|
||
|
||
async function clearMessages() {
|
||
if (!currentConversationId.value) return
|
||
try {
|
||
resetStreamingState()
|
||
await conversationApi.clearMessages(currentConversationId.value)
|
||
messages.value = []
|
||
} catch {
|
||
messages.value = []
|
||
}
|
||
}
|
||
|
||
// onModelChange removed — replaced by selectModel()
|
||
|
||
function goToModelSettings(providerId?: string) {
|
||
// Issue #81: when called with a providerId (e.g. from the unhealthy popup or
|
||
// ModelSelector's Fix button), pass it as a query param so a follow-up PR can
|
||
// scroll/focus the right card on the settings page. Today the consumer just
|
||
// ignores it; harmless meanwhile.
|
||
router.push({ path: '/settings/models', query: providerId ? { focus: providerId } : {} })
|
||
}
|
||
|
||
// ============ 计算属性：是否有待审批 ============
|
||
const hasPendingApproval = computed(() => {
|
||
return messages.value.some(
|
||
m => m.role === 'assistant' && (m as any).metadata?.pendingApproval?.status === 'pending_approval'
|
||
)
|
||
})
|
||
|
||
// 当前待审批的那条数据（传给 ChatInput 用于渲染审批栏）
|
||
const activePendingApproval = computed(() => {
|
||
const msg = messages.value.findLast(
|
||
m => m.role === 'assistant' && (m as any).metadata?.pendingApproval?.status === 'pending_approval'
|
||
)
|
||
return (msg as any)?.metadata?.pendingApproval ?? null
|
||
})
|
||
|
||
// 当前工具调用数
|
||
const toolCallCount = computed(() => {
|
||
const lastMsg = messages.value.findLast(m => m.role === 'assistant')
|
||
return lastMsg?.metadata?.toolCalls?.length ?? 0
|
||
})
|
||
|
||
// 当前正在执行的工具名称
|
||
const currentRunningToolName = computed(() => {
|
||
if (!isGenerating.value) return ''
|
||
const lastMsg = messages.value.findLast(m => m.role === 'assistant')
|
||
const metadata = lastMsg?.metadata
|
||
if (metadata?.runningToolName) return metadata.runningToolName
|
||
const runningTool = metadata?.toolCalls?.findLast((tc: any) => tc.status === 'running')
|
||
return runningTool?.name || heartbeat.value?.runningToolName || ''
|
||
})
|
||
|
||
// 当前正在生成的消息的 token 数据
|
||
const currentGeneratingTokens = computed(() => {
|
||
if (!isGenerating.value) return 0
|
||
// 找到最后一条 assistant 消息（可能仍在生成）
|
||
const lastMsg = messages.value.findLast(m => m.role === 'assistant')
|
||
// 返回 completionTokens（从服务器响应中获取）
|
||
return (lastMsg as any)?.completionTokens ?? 0
|
||
})
|
||
|
||
const currentPromptTokens = computed(() => {
|
||
if (!isGenerating.value) return 0
|
||
const lastMsg = messages.value.findLast(m => m.role === 'assistant')
|
||
return (lastMsg as any)?.promptTokens ?? 0
|
||
})
|
||
|
||
// ============ 消息发送和处理 ============
|
||
async function handleSendMessage(content: string) {
|
||
// 允许在等待审批时发送审批命令
|
||
const isApprovalCommand = /^\/(approve|deny)$/i.test(content.trim())
|
||
|
||
if ((!content && pendingAttachments.value.length === 0)
|
||
|| !selectedAgentId.value
|
||
|| blockingPrompt.value
|
||
|| workerConversationReadOnly.value) return
|
||
// 不再阻止运行中发送 — useChat 会自动走 interrupt/queue 路径
|
||
|
||
// 拦截 /approve 和 /deny 命令 —— 通过 SSE 流发送（和普通消息相同通道）
|
||
const trimmed = content.trim().toLowerCase()
|
||
if (trimmed === '/approve' || trimmed === '/deny') {
|
||
if (!currentConversationId.value) {
|
||
mcToast.warning('No active conversation')
|
||
inputText.value = ''
|
||
chatInputRef.value?.clear?.()
|
||
return
|
||
}
|
||
|
||
// 检查是否有 pending approval
|
||
const pendingMsg = messages.value.findLast(
|
||
m => m.role === 'assistant' && (m as any).metadata?.pendingApproval?.status === 'pending_approval'
|
||
)
|
||
if (!pendingMsg) {
|
||
mcToast.warning('No pending approval to process')
|
||
inputText.value = ''
|
||
chatInputRef.value?.clear?.()
|
||
return
|
||
}
|
||
|
||
// 乐观更新审批状态
|
||
const decision = trimmed === '/approve' ? 'approved' : 'denied'
|
||
;(pendingMsg as any).metadata.pendingApproval.status = decision
|
||
|
||
inputText.value = ''
|
||
chatInputRef.value?.clear?.()
|
||
|
||
// 通过正常 SSE 流发送（复用聊天通道，replay 结果实时流式推送）
|
||
try {
|
||
await sendChatMessage(trimmed, {
|
||
conversationId: currentConversationId.value,
|
||
agentId: selectedAgentId.value,
|
||
contentParts: [],
|
||
})
|
||
} catch (e: any) {
|
||
console.error('Approval stream failed:', e)
|
||
// 回滚乐观更新
|
||
;(pendingMsg as any).metadata.pendingApproval.status = 'pending_approval'
|
||
mcToast.error(e?.message || 'Approval failed')
|
||
}
|
||
return
|
||
}
|
||
|
||
if (!currentConversationId.value) {
|
||
currentConversationId.value = `conv_${Date.now()}_${Math.random().toString(36).slice(2, 8)}`
|
||
}
|
||
|
||
// Issue #81 v2 R2: only abort when there is genuinely no usable provider.
|
||
// If the active provider is unhealthy but another is LIVE, let the request
|
||
// through — NodeStreamingChatHelper's fallback walker will pick it up and
|
||
// emit a "warning" SSE delta which the input handler surfaces as a toast.
|
||
if (!activeModels.value?.activeLlm?.providerId || !activeModels.value?.activeLlm?.model) {
|
||
blockingPrompt.value = true
|
||
return
|
||
}
|
||
if (blockingPrompt.value) {
|
||
return
|
||
}
|
||
|
||
const outgoingAttachments = pendingAttachments.value.map((attachment) => ({ ...attachment }))
|
||
const contentParts = buildOutgoingParts(content, outgoingAttachments)
|
||
|
||
// 先暂存，发送成功后再清空（失败时恢复）
|
||
const savedInput = inputText.value
|
||
const savedAttachments = [...pendingAttachments.value]
|
||
// Stash for async-error recovery in onStreamEnd (sync catch can't reach this).
|
||
pendingSendDraft.value = { input: savedInput, attachments: savedAttachments }
|
||
inputText.value = ''
|
||
chatInputRef.value?.clear?.()
|
||
pendingAttachments.value = []
|
||
|
||
try {
|
||
await sendChatMessage(content, {
|
||
conversationId: currentConversationId.value,
|
||
agentId: selectedAgentId.value,
|
||
contentParts,
|
||
thinkingLevel: thinkingLevel.value,
|
||
modelProvider: activeModels.value?.activeLlm?.providerId,
|
||
modelName: activeModels.value?.activeLlm?.model,
|
||
attachments: outgoingAttachments.map(a => ({
|
||
type: 'file' as const,
|
||
fileUrl: a.url,
|
||
fileName: a.name,
|
||
storedName: a.storedName,
|
||
contentType: a.contentType,
|
||
fileSize: a.size,
|
||
path: a.path,
|
||
})),
|
||
})
|
||
// 发送成功后释放 ObjectURL
|
||
revokeAllPreviewUrls()
|
||
} catch (e) {
|
||
console.error('Send message failed:', e)
|
||
// 发送失败：恢复输入和附件，用户不丢失已上传的文件
|
||
if (!inputText.value) inputText.value = savedInput
|
||
if (pendingAttachments.value.length === 0) pendingAttachments.value = savedAttachments
|
||
}
|
||
}
|
||
|
||
function handleStopStream() {
|
||
stopChatGeneration()
|
||
}
|
||
|
||
async function handleRegenerate(message: Message) {
|
||
if (workerConversationReadOnly.value
|
||
|| isGenerating.value || !currentConversationId.value || !selectedAgentId.value) return
|
||
const idx = messages.value.indexOf(message)
|
||
if (idx >= 0) {
|
||
// The server drops the trailing assistant block and reuses the persisted
|
||
// seed user message (no duplicate user row) — mirror the truncation locally.
|
||
messages.value.splice(idx)
|
||
}
|
||
try {
|
||
await sendChatMessage('', {
|
||
conversationId: currentConversationId.value,
|
||
agentId: selectedAgentId.value,
|
||
contentParts: [],
|
||
thinkingLevel: thinkingLevel.value,
|
||
modelProvider: activeModels.value?.activeLlm?.providerId,
|
||
modelName: activeModels.value?.activeLlm?.model,
|
||
regenerate: true,
|
||
})
|
||
} catch (e: any) {
|
||
console.error('Regenerate failed:', e)
|
||
mcToast.error(e?.message || t('chat.regenerateFailed'))
|
||
}
|
||
}
|
||
|
||
async function handleRewind(message: Message) {
|
||
if (workerConversationReadOnly.value || isGenerating.value || !currentConversationId.value) return
|
||
const idx = messages.value.indexOf(message)
|
||
if (idx < 0) return
|
||
const count = messages.value.length - idx
|
||
try {
|
||
await ElMessageBox.confirm(
|
||
t('chat.rewindConfirm', { count }),
|
||
t('chat.rewindHere'),
|
||
{ type: 'warning' }
|
||
)
|
||
} catch {
|
||
return // user cancelled
|
||
}
|
||
try {
|
||
const res = await conversationApi.rewindMessage(
|
||
currentConversationId.value,
|
||
String(message.id)
|
||
)
|
||
messages.value.splice(idx)
|
||
// Sync the sidebar entry from the server-recomputed aggregates.
|
||
const data = res.data as { deletedCount: number; messageCount: number; lastMessage: string | null } | undefined
|
||
const conv = conversations.value.find(c => c.conversationId === currentConversationId.value)
|
||
if (conv && data) {
|
||
conv.lastMessage = data.lastMessage ?? ''
|
||
conv.messageCount = data.messageCount
|
||
}
|
||
} catch (e: any) {
|
||
mcToast.error(e?.message || t('chat.rewindFailed'))
|
||
}
|
||
}
|
||
|
||
function sendSuggestion(text: string) {
|
||
inputText.value = text
|
||
handleSendMessage(text)
|
||
}
|
||
|
||
function handleToggleThinking(message: import('@/types').Message, expanded: boolean) {
|
||
message.thinkingExpanded = expanded
|
||
}
|
||
|
||
// ============ 审批处理 ============
|
||
async function handleApprove(pendingId: string) {
|
||
if (!currentConversationId.value) return
|
||
await handleSendMessage('/approve')
|
||
}
|
||
|
||
async function handleDeny(pendingId: string) {
|
||
if (!currentConversationId.value) return
|
||
await handleSendMessage('/deny')
|
||
}
|
||
|
||
// Always-approve: create the matching grant first, then send /approve as usual.
|
||
// Failure to create the grant doesn't block the approval — we still forward
|
||
// /approve so the user's click isn't lost, just toast the error.
|
||
async function handleApproveAlways(
|
||
payload: { pendingId: string; scope: 'CONVERSATION' | 'AGENT' | 'USER' },
|
||
) {
|
||
if (!currentConversationId.value) return
|
||
const pa = activePendingApproval.value
|
||
if (!pa) return
|
||
|
||
// Resolve scope_id from the scope dimension.
|
||
let scopeId = ''
|
||
if (payload.scope === 'CONVERSATION') {
|
||
scopeId = currentConversationId.value
|
||
} else if (payload.scope === 'AGENT') {
|
||
scopeId = String(currentAgent.value?.id ?? '')
|
||
} else if (payload.scope === 'USER') {
|
||
// Login persists the id under 'userId' (see Login.vue); the old 'mc-user-id'
|
||
// key was never written, so USER-scope always-approve silently failed to
|
||
// resolve a scopeId and fell back to a one-shot /approve without a grant.
|
||
const me = localStorage.getItem('userId')
|
||
if (me) scopeId = me
|
||
}
|
||
if (!scopeId) {
|
||
ElMessage.error('Cannot resolve scope id for always-approve')
|
||
await handleSendMessage('/approve')
|
||
return
|
||
}
|
||
|
||
try {
|
||
const sev = pa.maxSeverity ?? 'LOW'
|
||
// Severity ceiling = at-or-above the current finding's severity. CRITICAL
|
||
// never enters this path (the backend rejects it), so HIGH covers the rest.
|
||
const ceiling = sev === 'HIGH' || sev === 'CRITICAL' ? 'HIGH'
|
||
: sev === 'MEDIUM' ? 'MEDIUM' : 'LOW'
|
||
const ruleId = pa.findings?.find((f: { ruleId?: string }) => !!f.ruleId)?.ruleId ?? null
|
||
await approvalApi.createGrant({
|
||
scopeType: payload.scope,
|
||
scopeId,
|
||
toolName: pa.toolName,
|
||
ruleId,
|
||
maxSeverity: ceiling,
|
||
grantKind: payload.scope === 'CONVERSATION' ? 'UNTIL_CONVERSATION_END' : 'ALWAYS',
|
||
note: `created from approval banner (${pa.toolName})`,
|
||
})
|
||
ElMessage.success(
|
||
t('chat.approveAlwaysCreated', { tool: pa.toolName }) as string,
|
||
)
|
||
} catch (e: any) {
|
||
ElMessage.error(e?.message || 'Failed to create auto-approve rule')
|
||
}
|
||
await handleSendMessage('/approve')
|
||
}
|
||
|
||
// 重连到运行中的流
|
||
async function reconnectStream(conversationId: string) {
|
||
try {
|
||
await reconnectChatStream(conversationId)
|
||
} catch (e) {
|
||
console.error('[ChatConsole] Reconnect failed:', e)
|
||
mcToast.warning(t('chat.reconnectFailed') || 'Stream reconnection failed')
|
||
}
|
||
}
|
||
|
||
function handleCancelQueued() {
|
||
cancelQueued()
|
||
}
|
||
|
||
// 简化版重置函数
|
||
function resetStreamingState() {
|
||
// 先通知后端停止旧流（fire-and-forget），再彻底清理前端状态
|
||
stopChatGeneration()
|
||
resetForNewConversation()
|
||
}
|
||
|
||
// ============ 附件处理 ============
|
||
async function handleFileSelect(files: File[]) {
|
||
if (workerConversationReadOnly.value) return
|
||
if (!currentConversationId.value) {
|
||
newConversation()
|
||
}
|
||
|
||
uploadingAttachment.value = true
|
||
try {
|
||
for (const file of files) {
|
||
const res: any = await chatApi.uploadFile(currentConversationId.value, file)
|
||
const data = res.data || {}
|
||
// 图片/视频使用本地 ObjectURL 预览（避免 /api/v1/chat/files/ 需要 JWT 认证导致加载失败）
|
||
const ct = data.contentType || file.type || ''
|
||
const isPreviewable = ct.startsWith('image/') || ct.startsWith('video/')
|
||
const previewUrl = isPreviewable ? URL.createObjectURL(file) : data.url
|
||
pendingAttachments.value.push({
|
||
name: data.fileName || file.name,
|
||
size: data.size || file.size,
|
||
url: data.url,
|
||
storedName: data.storedName,
|
||
path: data.path,
|
||
contentType: data.contentType || file.type,
|
||
previewUrl,
|
||
})
|
||
}
|
||
} catch (e) {
|
||
mcToast.error(t('chat.uploadFailed'))
|
||
} finally {
|
||
uploadingAttachment.value = false
|
||
}
|
||
}
|
||
|
||
function removeAttachment(key: string) {
|
||
// revoke 被移除附件的 ObjectURL，防止内存泄漏
|
||
const removed = pendingAttachments.value.find(a => a.storedName === key || a.path === key)
|
||
if (removed?.previewUrl?.startsWith('blob:')) {
|
||
URL.revokeObjectURL(removed.previewUrl)
|
||
}
|
||
pendingAttachments.value = pendingAttachments.value.filter(
|
||
a => a.storedName !== key && a.path !== key
|
||
)
|
||
}
|
||
|
||
/** 释放所有 pending 附件的 ObjectURL */
|
||
function revokeAllPreviewUrls() {
|
||
for (const a of pendingAttachments.value) {
|
||
if (a.previewUrl?.startsWith('blob:')) {
|
||
URL.revokeObjectURL(a.previewUrl)
|
||
}
|
||
}
|
||
}
|
||
|
||
function buildOutgoingParts(text: string, attachments: ChatAttachment[]): MessageContentPart[] {
|
||
const parts: MessageContentPart[] = []
|
||
if (text) parts.push({ type: 'text', text })
|
||
for (const attachment of attachments) {
|
||
const ct = attachment.contentType || ''
|
||
const partType: MessageContentPart['type'] = ct.startsWith('video/') ? 'video'
|
||
: ct.startsWith('image/') ? 'image'
|
||
: 'file'
|
||
parts.push({
|
||
type: partType,
|
||
fileUrl: attachment.url,
|
||
fileName: attachment.name,
|
||
storedName: attachment.storedName,
|
||
contentType: attachment.contentType,
|
||
fileSize: attachment.size,
|
||
path: attachment.path,
|
||
})
|
||
}
|
||
return parts
|
||
}
|
||
|
||
// ============ 工具函数 ============
|
||
function normalizeMessage(raw: Message, preserveGeneratingStatus?: boolean): Message {
|
||
const msg: Message = { ...raw, contentParts: raw.contentParts ? [...raw.contentParts] : [] }
|
||
|
||
// 统一解析 metadata：确保是对象而非 JSON 字符串
|
||
// 注意：后端 metadata 在 DB 中是 JSON 字符串，Jackson 序列化时可能双重编码
|
||
if (typeof msg.metadata === 'string') {
|
||
try {
|
||
let parsed = JSON.parse(msg.metadata)
|
||
// 处理双重编码：parse 后仍然是字符串的情况
|
||
if (typeof parsed === 'string') {
|
||
try { parsed = JSON.parse(parsed) } catch { /* ignore */ }
|
||
}
|
||
msg.metadata = parsed
|
||
} catch { msg.metadata = {} as any }
|
||
}
|
||
|
||
// 保留后端返回的 token 字段（MessageVO 新增）
|
||
if ((raw as any).promptTokens) msg.promptTokens = (raw as any).promptTokens
|
||
if ((raw as any).completionTokens) msg.completionTokens = (raw as any).completionTokens
|
||
if ((raw as any).cacheReadTokens) msg.cacheReadTokens = (raw as any).cacheReadTokens
|
||
if ((raw as any).cacheWriteTokens) msg.cacheWriteTokens = (raw as any).cacheWriteTokens
|
||
if ((raw as any).reasoningTokens) msg.reasoningTokens = (raw as any).reasoningTokens
|
||
if ((raw as any).runtimeModel) msg.runtimeModel = (raw as any).runtimeModel
|
||
if ((raw as any).runtimeProvider) msg.runtimeProvider = (raw as any).runtimeProvider
|
||
|
||
if (msg.contentParts.length === 0 && msg.content) {
|
||
if (msg.role === 'assistant') {
|
||
const parsed = parseThinkingContent(msg.content)
|
||
// thinkingLevel=off 时不展示 thinking 内容，直接剥离 <think> 标签
|
||
if (parsed.thinking && thinkingLevel.value !== 'off') {
|
||
msg.contentParts.push({ type: 'thinking', text: parsed.thinking })
|
||
}
|
||
if (parsed.content) msg.contentParts.push({ type: 'text', text: parsed.content })
|
||
msg.content = parsed.content
|
||
} else {
|
||
msg.contentParts.push({ type: 'text', text: msg.content })
|
||
}
|
||
}
|
||
|
||
// 从 tool_call contentParts 还原 metadata.toolCalls
|
||
const toolCallParts = msg.contentParts.filter(p => p.type === 'tool_call')
|
||
if (toolCallParts.length > 0) {
|
||
const toolCalls: ToolCallMeta[] = []
|
||
for (const part of toolCallParts) {
|
||
try {
|
||
const parsed = JSON.parse(part.text || '{}')
|
||
toolCalls.push({
|
||
name: parsed.name || '',
|
||
arguments: parsed.arguments,
|
||
result: parsed.result,
|
||
success: parsed.success,
|
||
// 历史消息中不应有 running 状态的工具调用（流已结束）
|
||
status: 'completed',
|
||
})
|
||
} catch {
|
||
// skip malformed tool_call parts
|
||
}
|
||
}
|
||
if (toolCalls.length > 0) {
|
||
msg.metadata = { ...msg.metadata, toolCalls }
|
||
}
|
||
msg.contentParts = msg.contentParts.filter(p => p.type !== 'tool_call')
|
||
}
|
||
|
||
// 历史消息的 metadata.toolCalls 也需要清理 running 状态
|
||
if (msg.metadata?.toolCalls) {
|
||
const cleaned = msg.metadata.toolCalls.map((tc: ToolCallMeta) => ({
|
||
...tc,
|
||
status: tc.status === 'running' ? 'completed' as const : tc.status,
|
||
}))
|
||
msg.metadata = { ...msg.metadata, toolCalls: cleaned }
|
||
}
|
||
|
||
if (!preserveGeneratingStatus && msg.status === 'generating') msg.status = 'failed'
|
||
// interrupted 是合法的历史状态（interrupt-with-followup），不映射为 stopped
|
||
if (!msg.status) msg.status = 'completed'
|
||
|
||
// 从 file/image/video contentParts 恢复 attachments（历史消息 API 不返回单独的 attachments 字段）
|
||
const fileParts = msg.contentParts.filter(p => (p.type === 'file' || p.type === 'image' || p.type === 'video') && p.fileUrl)
|
||
if (fileParts.length > 0 && (!msg.attachments || msg.attachments.length === 0)) {
|
||
msg.attachments = fileParts.map(p => ({
|
||
name: p.fileName || 'unknown',
|
||
size: typeof p.fileSize === 'number' ? p.fileSize : Number(p.fileSize) || 0,
|
||
url: p.fileUrl!,
|
||
storedName: p.storedName || '',
|
||
path: p.path || '',
|
||
contentType: p.contentType,
|
||
}))
|
||
}
|
||
|
||
// 从持久化的 [错误] 文本重建 errorInfo，使刷新后也能显示错误卡片
|
||
if (msg.role === 'assistant' && !msg.errorInfo) {
|
||
const text = msg.content || msg.contentParts?.find(p => p.type === 'text')?.text || ''
|
||
const rebuilt = reconstructErrorInfo(text)
|
||
if (rebuilt) {
|
||
msg.errorInfo = rebuilt
|
||
msg.status = 'failed'
|
||
}
|
||
}
|
||
|
||
return msg
|
||
}
|
||
|
||
function parseThinkingContent(raw: string): { content: string; thinking: string; hasThinking: boolean } {
|
||
if (!raw) return { content: '', thinking: '', hasThinking: false }
|
||
|
||
const normalized = raw.replace(/<thinking>/gi, '<think>').replace(/<\/thinking>/gi, '</think>')
|
||
const thinkingParts: string[] = []
|
||
const content = normalized.replace(/<think>([\s\S]*?)<\/think>/gi, (_, thinkingText: string) => {
|
||
const cleanText = thinkingText.trim()
|
||
if (cleanText) thinkingParts.push(cleanText)
|
||
return ''
|
||
}).trim()
|
||
|
||
return {
|
||
content,
|
||
thinking: thinkingParts.join('\n\n').trim(),
|
||
hasThinking: thinkingParts.length > 0,
|
||
}
|
||
}
|
||
|
||
function handleCodeCopy(e: MouseEvent) {
|
||
// Mermaid download button shares the same global click delegation. Handle
|
||
// it first so the SVG export beats the copy-button selector below if the
|
||
// user happens to click in an area where both ancestors are reachable.
|
||
if (handleMermaidDownload(e)) return
|
||
const btn = (e.target as HTMLElement).closest('.code-block__copy') as HTMLElement | null
|
||
if (!btn) return
|
||
// The copy button now sits inside <details><summary> for collapsible code
|
||
// blocks. Without preventDefault the click would also toggle the details
|
||
// open state — a regression introduced when we wrapped long blocks in
|
||
// <details>. stopPropagation guards against any future ancestor handlers.
|
||
e.preventDefault()
|
||
e.stopPropagation()
|
||
const encoded = btn.getAttribute('data-code')
|
||
if (!encoded) return
|
||
const code = decodeURIComponent(encoded)
|
||
copyToClipboard(code).then(() => {
|
||
btn.classList.add('copied')
|
||
const textEl = btn.querySelector('.code-block__copy-text')
|
||
if (textEl) textEl.textContent = t('chat.copied')
|
||
setTimeout(() => {
|
||
btn.classList.remove('copied')
|
||
if (textEl) textEl.textContent = t('chat.copy')
|
||
}, 1500)
|
||
}).catch(() => {
|
||
mcToast.error(t('chat.copyFailed'))
|
||
})
|
||
}
|
||
</script>
|
||
|
||
<style scoped>
|
||
.ctx-usage-row {
|
||
display: flex;
|
||
justify-content: flex-end;
|
||
padding: 0 4px 4px;
|
||
}
|
||
.cron-running-bar {
|
||
display: flex;
|
||
flex-direction: column;
|
||
gap: 6px;
|
||
padding: 10px 16px;
|
||
margin: 0 12px;
|
||
background: var(--mc-warning-bg, rgba(255, 159, 67, 0.08));
|
||
border: 1px solid var(--mc-warning, rgba(255, 159, 67, 0.35));
|
||
border-radius: 10px;
|
||
color: var(--mc-text-primary);
|
||
font-size: 13px;
|
||
}
|
||
.cron-running-item {
|
||
display: flex;
|
||
align-items: center;
|
||
gap: 10px;
|
||
}
|
||
.cron-running-spinner {
|
||
display: inline-block;
|
||
font-size: 16px;
|
||
animation: cron-spin 1.6s linear infinite;
|
||
}
|
||
@keyframes cron-spin {
|
||
from { transform: rotate(0deg); }
|
||
to { transform: rotate(360deg); }
|
||
}
|
||
.cron-running-text { line-height: 1.4; }
|
||
.cron-running-meta { color: var(--mc-text-secondary); margin-left: 4px; }
|
||
|
||
.chat-console-shell {
|
||
background: transparent;
|
||
min-height: 0;
|
||
height: 100%;
|
||
overflow: hidden;
|
||
}
|
||
|
||
.chat-console-frame {
|
||
height: min(calc(100vh - 28px), 100%);
|
||
min-height: 0;
|
||
overflow: hidden;
|
||
}
|
||
|
||
.chat-layout {
|
||
display: flex;
|
||
height: 100%;
|
||
overflow: hidden;
|
||
min-height: 0;
|
||
}
|
||
|
||
.chat-area {
|
||
flex: 1;
|
||
display: flex;
|
||
flex-direction: column;
|
||
overflow: hidden;
|
||
background: linear-gradient(180deg, var(--mc-chat-header-bg), var(--mc-chat-bg));
|
||
position: relative;
|
||
min-height: 0;
|
||
}
|
||
|
||
/* 拖拽上传遮罩 */
|
||
.drop-overlay {
|
||
position: absolute;
|
||
inset: 0;
|
||
z-index: 100;
|
||
display: flex;
|
||
align-items: center;
|
||
justify-content: center;
|
||
background: rgba(217, 119, 87, 0.06);
|
||
backdrop-filter: blur(2px);
|
||
}
|
||
|
||
.drop-overlay__content {
|
||
display: flex;
|
||
flex-direction: column;
|
||
align-items: center;
|
||
gap: 12px;
|
||
padding: 40px 60px;
|
||
border: 2px dashed var(--mc-primary, #D97757);
|
||
border-radius: 16px;
|
||
background: var(--mc-bg-elevated, #f8fafc);
|
||
color: var(--mc-primary, #D97757);
|
||
font-size: 16px;
|
||
font-weight: 500;
|
||
}
|
||
|
||
.fade-enter-active,
|
||
.fade-leave-active {
|
||
transition: opacity 0.2s ease;
|
||
}
|
||
|
||
.fade-enter-from,
|
||
.fade-leave-to {
|
||
opacity: 0;
|
||
}
|
||
|
||
.chat-header {
|
||
display: flex;
|
||
align-items: center;
|
||
justify-content: space-between;
|
||
padding: 10px 16px;
|
||
background: linear-gradient(180deg, var(--mc-panel-raised), var(--mc-surface-overlay));
|
||
border-bottom: 1px solid var(--mc-border);
|
||
min-height: 52px;
|
||
backdrop-filter: blur(12px);
|
||
gap: 10px;
|
||
}
|
||
|
||
.chat-header-left {
|
||
display: flex;
|
||
align-items: center;
|
||
min-width: 0;
|
||
gap: 10px;
|
||
}
|
||
|
||
.chat-header-right {
|
||
display: flex;
|
||
align-items: center;
|
||
gap: 8px;
|
||
flex-shrink: 0;
|
||
}
|
||
|
||
.chat-stage-copy {
|
||
display: flex;
|
||
flex-direction: column;
|
||
gap: 4px;
|
||
min-width: 0;
|
||
}
|
||
|
||
.chat-stage-kicker {
|
||
font-size: 10px;
|
||
font-weight: 700;
|
||
letter-spacing: 0.12em;
|
||
text-transform: uppercase;
|
||
color: var(--mc-accent);
|
||
}
|
||
|
||
.agent-badge {
|
||
display: flex;
|
||
align-items: center;
|
||
gap: 6px;
|
||
padding: 5px 10px;
|
||
background: var(--mc-primary-bg);
|
||
border-radius: 999px;
|
||
max-width: 100%;
|
||
}
|
||
|
||
.agent-badge-icon {
|
||
display: flex;
|
||
align-items: center;
|
||
font-size: 14px;
|
||
}
|
||
|
||
.agent-badge-text {
|
||
display: flex;
|
||
flex-direction: column;
|
||
gap: 1px;
|
||
min-width: 0;
|
||
}
|
||
|
||
.agent-badge-name {
|
||
font-size: 13px;
|
||
font-weight: 600;
|
||
color: var(--mc-primary);
|
||
white-space: nowrap;
|
||
overflow: hidden;
|
||
text-overflow: ellipsis;
|
||
line-height: 1.2;
|
||
}
|
||
|
||
|
||
.status-dot {
|
||
width: 8px; height: 8px; border-radius: 50%; flex-shrink: 0; margin-left: 2px;
|
||
transition: background 0.3s;
|
||
}
|
||
.status-idle { background: #34d399; box-shadow: 0 0 4px rgba(52, 211, 153, 0.5); }
|
||
.status-streaming { background: #fbbf24; box-shadow: 0 0 4px rgba(251, 191, 36, 0.5); animation: pulse-dot 1.2s infinite; }
|
||
.status-error { background: #f87171; box-shadow: 0 0 4px rgba(248, 113, 113, 0.5); }
|
||
@keyframes pulse-dot { 0%, 100% { opacity: 1; } 50% { opacity: 0.4; } }
|
||
|
||
.no-agent-hint {
|
||
font-size: 13px;
|
||
color: var(--mc-text-tertiary);
|
||
}
|
||
|
||
/* Model selector */
|
||
/* Model selector styles moved to ModelSelector.vue */
|
||
|
||
/* Header overflow menu */
|
||
.header-overflow-wrap {
|
||
position: relative;
|
||
}
|
||
|
||
.header-btn {
|
||
width: 30px;
|
||
height: 30px;
|
||
border: 1px solid var(--mc-border);
|
||
background: var(--mc-panel-raised);
|
||
border-radius: 10px;
|
||
cursor: pointer;
|
||
display: flex;
|
||
align-items: center;
|
||
justify-content: center;
|
||
color: var(--mc-text-secondary);
|
||
transition: all 0.15s;
|
||
}
|
||
|
||
.header-btn:hover {
|
||
border-color: var(--mc-danger);
|
||
color: var(--mc-danger);
|
||
}
|
||
|
||
.model-prompt {
|
||
margin: 24px auto 0;
|
||
max-width: 540px;
|
||
padding: 20px;
|
||
background: var(--mc-bg-elevated);
|
||
border: 1px solid var(--mc-border);
|
||
border-radius: 16px;
|
||
text-align: center;
|
||
box-shadow: 0 8px 24px rgba(124, 63, 30, 0.06);
|
||
}
|
||
|
||
.model-prompt-title {
|
||
font-size: 18px;
|
||
font-weight: 700;
|
||
color: var(--mc-text-primary);
|
||
margin-bottom: 8px;
|
||
}
|
||
|
||
.model-prompt-desc {
|
||
font-size: 14px;
|
||
color: var(--mc-text-secondary);
|
||
line-height: 1.6;
|
||
margin-bottom: 16px;
|
||
}
|
||
|
||
.btn-primary {
|
||
padding: 8px 16px;
|
||
background: linear-gradient(135deg, var(--mc-primary), var(--mc-primary-hover));
|
||
color: white;
|
||
border: none;
|
||
border-radius: 12px;
|
||
font-size: 14px;
|
||
cursor: pointer;
|
||
transition: background 0.15s;
|
||
}
|
||
|
||
.btn-primary:hover {
|
||
background: var(--mc-primary-hover);
|
||
}
|
||
|
||
/* Issue #81: side-by-side primary + secondary actions in the model prompt. */
|
||
.model-prompt-actions {
|
||
display: inline-flex;
|
||
flex-wrap: wrap;
|
||
gap: 8px;
|
||
justify-content: center;
|
||
}
|
||
|
||
.btn-secondary {
|
||
padding: 8px 14px;
|
||
background: transparent;
|
||
color: var(--mc-text-primary);
|
||
border: 1px solid var(--mc-border);
|
||
border-radius: 12px;
|
||
font-size: 14px;
|
||
cursor: pointer;
|
||
transition: background 0.15s, color 0.15s, border-color 0.15s;
|
||
}
|
||
|
||
.btn-secondary:hover {
|
||
background: var(--mc-panel-raised);
|
||
border-color: var(--mc-primary);
|
||
}
|
||
|
||
/* ===== 移动端元素（桌面端隐藏） ===== */
|
||
.conv-backdrop {
|
||
display: none;
|
||
}
|
||
|
||
.conv-toggle-btn {
|
||
display: none;
|
||
}
|
||
|
||
/* ===== 移动端适配 ===== */
|
||
@media (max-width: 768px) {
|
||
.chat-console-shell {
|
||
padding: 0 !important;
|
||
height: 100dvh !important;
|
||
height: 100vh !important;
|
||
overflow: hidden !important;
|
||
min-height: 0 !important;
|
||
}
|
||
|
||
.chat-console-frame {
|
||
height: 100% !important;
|
||
min-height: 0 !important;
|
||
overflow: hidden !important;
|
||
border-radius: 0;
|
||
border: none;
|
||
}
|
||
|
||
.conv-backdrop {
|
||
display: block;
|
||
position: fixed;
|
||
inset: 0;
|
||
z-index: 99;
|
||
background: rgba(0, 0, 0, 0.3);
|
||
}
|
||
|
||
.conv-toggle-btn {
|
||
display: flex;
|
||
align-items: center;
|
||
justify-content: center;
|
||
width: 32px;
|
||
height: 32px;
|
||
border: 1px solid var(--mc-border);
|
||
background: var(--mc-bg-elevated);
|
||
border-radius: 6px;
|
||
cursor: pointer;
|
||
color: var(--mc-text-secondary);
|
||
flex-shrink: 0;
|
||
transition: all 0.15s;
|
||
}
|
||
|
||
.conv-toggle-btn:hover {
|
||
border-color: var(--mc-primary);
|
||
color: var(--mc-primary);
|
||
}
|
||
|
||
.chat-header {
|
||
padding: 9px 12px;
|
||
gap: 8px;
|
||
}
|
||
|
||
.agent-badge {
|
||
padding: 4px 8px;
|
||
}
|
||
|
||
.chat-stage-kicker {
|
||
display: none;
|
||
}
|
||
|
||
.agent-badge-text {
|
||
display: none;
|
||
}
|
||
|
||
.model-select-trigger {
|
||
max-width: 160px;
|
||
}
|
||
|
||
.model-dropdown {
|
||
min-width: 200px;
|
||
}
|
||
|
||
.drop-overlay__content {
|
||
padding: 24px 32px;
|
||
font-size: 14px;
|
||
}
|
||
}
|
||
|
||
@media (max-width: 480px) {
|
||
.chat-header {
|
||
padding: 6px 8px;
|
||
min-height: 44px;
|
||
}
|
||
|
||
.chat-header-right {
|
||
gap: 4px;
|
||
}
|
||
|
||
.model-select-trigger {
|
||
max-width: 120px;
|
||
height: 30px;
|
||
padding: 0 8px;
|
||
font-size: 12px;
|
||
}
|
||
|
||
.header-btn {
|
||
width: 28px;
|
||
height: 28px;
|
||
}
|
||
}
|
||
|
||
</style>
|