chore: add jotai store (#30432)

Signed-off-by: yyh <yuanyouhuilyz@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: yyh <yuanyouhuilyz@gmail.com>

.claude/settings.json

@@ -0,0 +1,8 @@
+{
+  "enabledPlugins": {
+    "feature-dev@claude-plugins-official": true,
+    "context7@claude-plugins-official": true,
+    "typescript-lsp@claude-plugins-official": true,
+    "pyright-lsp@claude-plugins-official": true
+  }
+}

.claude/settings.json.example

@@ -1,19 +0,0 @@
-{
-    "permissions": {
-      "allow": [],
-      "deny": []
-    },
-    "env": {
-      "__comment": "Environment variables for MCP servers. Override in .claude/settings.local.json with actual values.",
-      "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
-    },
-    "enabledMcpjsonServers": [
-      "context7",
-      "sequential-thinking",
-      "github",
-      "fetch",
-      "playwright",
-      "ide"
-    ],
-    "enableAllProjectMcpServers": true
-  }

.claude/skills/component-refactoring/SKILL.md

@@ -0,0 +1,483 @@
+---
+name: component-refactoring
+description: Refactor high-complexity React components in Dify frontend. Use when `pnpm analyze-component --json` shows complexity > 50 or lineCount > 300, when the user asks for code splitting, hook extraction, or complexity reduction, or when `pnpm analyze-component` warns to refactor before testing; avoid for simple/well-structured components, third-party wrappers, or when the user explicitly wants testing without refactoring.
+---
+
+# Dify Component Refactoring Skill
+
+Refactor high-complexity React components in the Dify frontend codebase with the patterns and workflow below.
+
+> **Complexity Threshold**: Components with complexity > 50 (measured by `pnpm analyze-component`) should be refactored before testing.
+
+## Quick Reference
+
+### Commands (run from `web/`)
+
+Use paths relative to `web/` (e.g., `app/components/...`).
+Use `refactor-component` for refactoring prompts and `analyze-component` for testing prompts and metrics.
+
+```bash
+cd web
+
+# Generate refactoring prompt
+pnpm refactor-component <path>
+
+# Output refactoring analysis as JSON
+pnpm refactor-component <path> --json
+
+# Generate testing prompt (after refactoring)
+pnpm analyze-component <path>
+
+# Output testing analysis as JSON
+pnpm analyze-component <path> --json
+```
+
+### Complexity Analysis
+
+```bash
+# Analyze component complexity
+pnpm analyze-component <path> --json
+
+# Key metrics to check:
+# - complexity: normalized score 0-100 (target < 50)
+# - maxComplexity: highest single function complexity
+# - lineCount: total lines (target < 300)
+```
+
+### Complexity Score Interpretation
+
+| Score | Level | Action |
+|-------|-------|--------|
+| 0-25 | 🟢 Simple | Ready for testing |
+| 26-50 | 🟡 Medium | Consider minor refactoring |
+| 51-75 | 🟠 Complex | **Refactor before testing** |
+| 76-100 | 🔴 Very Complex | **Must refactor** |
+
+## Core Refactoring Patterns
+
+### Pattern 1: Extract Custom Hooks
+
+**When**: Component has complex state management, multiple `useState`/`useEffect`, or business logic mixed with UI.
+
+**Dify Convention**: Place hooks in a `hooks/` subdirectory or alongside the component as `use-<feature>.ts`.
+
+```typescript
+// ❌ Before: Complex state logic in component
+const Configuration: FC = () => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+  const [datasetConfigs, setDatasetConfigs] = useState<DatasetConfigs>(...)
+  const [completionParams, setCompletionParams] = useState<FormValue>({})
+  
+  // 50+ lines of state management logic...
+  
+  return <div>...</div>
+}
+
+// ✅ After: Extract to custom hook
+// hooks/use-model-config.ts
+export const useModelConfig = (appId: string) => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+  const [completionParams, setCompletionParams] = useState<FormValue>({})
+  
+  // Related state management logic here
+  
+  return { modelConfig, setModelConfig, completionParams, setCompletionParams }
+}
+
+// Component becomes cleaner
+const Configuration: FC = () => {
+  const { modelConfig, setModelConfig } = useModelConfig(appId)
+  return <div>...</div>
+}
+```
+
+**Dify Examples**:
+- `web/app/components/app/configuration/hooks/use-advanced-prompt-config.ts`
+- `web/app/components/app/configuration/debug/hooks.tsx`
+- `web/app/components/workflow/hooks/use-workflow.ts`
+
+### Pattern 2: Extract Sub-Components
+
+**When**: Single component has multiple UI sections, conditional rendering blocks, or repeated patterns.
+
+**Dify Convention**: Place sub-components in subdirectories or as separate files in the same directory.
+
+```typescript
+// ❌ Before: Monolithic JSX with multiple sections
+const AppInfo = () => {
+  return (
+    <div>
+      {/* 100 lines of header UI */}
+      {/* 100 lines of operations UI */}
+      {/* 100 lines of modals */}
+    </div>
+  )
+}
+
+// ✅ After: Split into focused components
+// app-info/
+//   ├── index.tsx           (orchestration only)
+//   ├── app-header.tsx      (header UI)
+//   ├── app-operations.tsx  (operations UI)
+//   └── app-modals.tsx      (modal management)
+
+const AppInfo = () => {
+  const { showModal, setShowModal } = useAppInfoModals()
+  
+  return (
+    <div>
+      <AppHeader appDetail={appDetail} />
+      <AppOperations onAction={handleAction} />
+      <AppModals show={showModal} onClose={() => setShowModal(null)} />
+    </div>
+  )
+}
+```
+
+**Dify Examples**:
+- `web/app/components/app/configuration/` directory structure
+- `web/app/components/workflow/nodes/` per-node organization
+
+### Pattern 3: Simplify Conditional Logic
+
+**When**: Deep nesting (> 3 levels), complex ternaries, or multiple `if/else` chains.
+
+```typescript
+// ❌ Before: Deeply nested conditionals
+const Template = useMemo(() => {
+  if (appDetail?.mode === AppModeEnum.CHAT) {
+    switch (locale) {
+      case LanguagesSupported[1]:
+        return <TemplateChatZh />
+      case LanguagesSupported[7]:
+        return <TemplateChatJa />
+      default:
+        return <TemplateChatEn />
+    }
+  }
+  if (appDetail?.mode === AppModeEnum.ADVANCED_CHAT) {
+    // Another 15 lines...
+  }
+  // More conditions...
+}, [appDetail, locale])
+
+// ✅ After: Use lookup tables + early returns
+const TEMPLATE_MAP = {
+  [AppModeEnum.CHAT]: {
+    [LanguagesSupported[1]]: TemplateChatZh,
+    [LanguagesSupported[7]]: TemplateChatJa,
+    default: TemplateChatEn,
+  },
+  [AppModeEnum.ADVANCED_CHAT]: {
+    [LanguagesSupported[1]]: TemplateAdvancedChatZh,
+    // ...
+  },
+}
+
+const Template = useMemo(() => {
+  const modeTemplates = TEMPLATE_MAP[appDetail?.mode]
+  if (!modeTemplates) return null
+  
+  const TemplateComponent = modeTemplates[locale] || modeTemplates.default
+  return <TemplateComponent appDetail={appDetail} />
+}, [appDetail, locale])
+```
+
+### Pattern 4: Extract API/Data Logic
+
+**When**: Component directly handles API calls, data transformation, or complex async operations.
+
+**Dify Convention**: Use `@tanstack/react-query` hooks from `web/service/use-*.ts` or create custom data hooks.
+
+```typescript
+// ❌ Before: API logic in component
+const MCPServiceCard = () => {
+  const [basicAppConfig, setBasicAppConfig] = useState({})
+  
+  useEffect(() => {
+    if (isBasicApp && appId) {
+      (async () => {
+        const res = await fetchAppDetail({ url: '/apps', id: appId })
+        setBasicAppConfig(res?.model_config || {})
+      })()
+    }
+  }, [appId, isBasicApp])
+  
+  // More API-related logic...
+}
+
+// ✅ After: Extract to data hook using React Query
+// use-app-config.ts
+import { useQuery } from '@tanstack/react-query'
+import { get } from '@/service/base'
+
+const NAME_SPACE = 'appConfig'
+
+export const useAppConfig = (appId: string, isBasicApp: boolean) => {
+  return useQuery({
+    enabled: isBasicApp && !!appId,
+    queryKey: [NAME_SPACE, 'detail', appId],
+    queryFn: () => get<AppDetailResponse>(`/apps/${appId}`),
+    select: data => data?.model_config || {},
+  })
+}
+
+// Component becomes cleaner
+const MCPServiceCard = () => {
+  const { data: config, isLoading } = useAppConfig(appId, isBasicApp)
+  // UI only
+}
+```
+
+**React Query Best Practices in Dify**:
+- Define `NAME_SPACE` for query key organization
+- Use `enabled` option for conditional fetching
+- Use `select` for data transformation
+- Export invalidation hooks: `useInvalidXxx`
+
+**Dify Examples**:
+- `web/service/use-workflow.ts`
+- `web/service/use-common.ts`
+- `web/service/knowledge/use-dataset.ts`
+- `web/service/knowledge/use-document.ts`
+
+### Pattern 5: Extract Modal/Dialog Management
+
+**When**: Component manages multiple modals with complex open/close states.
+
+**Dify Convention**: Modals should be extracted with their state management.
+
+```typescript
+// ❌ Before: Multiple modal states in component
+const AppInfo = () => {
+  const [showEditModal, setShowEditModal] = useState(false)
+  const [showDuplicateModal, setShowDuplicateModal] = useState(false)
+  const [showConfirmDelete, setShowConfirmDelete] = useState(false)
+  const [showSwitchModal, setShowSwitchModal] = useState(false)
+  const [showImportDSLModal, setShowImportDSLModal] = useState(false)
+  // 5+ more modal states...
+}
+
+// ✅ After: Extract to modal management hook
+type ModalType = 'edit' | 'duplicate' | 'delete' | 'switch' | 'import' | null
+
+const useAppInfoModals = () => {
+  const [activeModal, setActiveModal] = useState<ModalType>(null)
+  
+  const openModal = useCallback((type: ModalType) => setActiveModal(type), [])
+  const closeModal = useCallback(() => setActiveModal(null), [])
+  
+  return {
+    activeModal,
+    openModal,
+    closeModal,
+    isOpen: (type: ModalType) => activeModal === type,
+  }
+}
+```
+
+### Pattern 6: Extract Form Logic
+
+**When**: Complex form validation, submission handling, or field transformation.
+
+**Dify Convention**: Use `@tanstack/react-form` patterns from `web/app/components/base/form/`.
+
+```typescript
+// ✅ Use existing form infrastructure
+import { useAppForm } from '@/app/components/base/form'
+
+const ConfigForm = () => {
+  const form = useAppForm({
+    defaultValues: { name: '', description: '' },
+    onSubmit: handleSubmit,
+  })
+  
+  return <form.Provider>...</form.Provider>
+}
+```
+
+## Dify-Specific Refactoring Guidelines
+
+### 1. Context Provider Extraction
+
+**When**: Component provides complex context values with multiple states.
+
+```typescript
+// ❌ Before: Large context value object
+const value = {
+  appId, isAPIKeySet, isTrailFinished, mode, modelModeType,
+  promptMode, isAdvancedMode, isAgent, isOpenAI, isFunctionCall,
+  // 50+ more properties...
+}
+return <ConfigContext.Provider value={value}>...</ConfigContext.Provider>
+
+// ✅ After: Split into domain-specific contexts
+<ModelConfigProvider value={modelConfigValue}>
+  <DatasetConfigProvider value={datasetConfigValue}>
+    <UIConfigProvider value={uiConfigValue}>
+      {children}
+    </UIConfigProvider>
+  </DatasetConfigProvider>
+</ModelConfigProvider>
+```
+
+**Dify Reference**: `web/context/` directory structure
+
+### 2. Workflow Node Components
+
+**When**: Refactoring workflow node components (`web/app/components/workflow/nodes/`).
+
+**Conventions**:
+- Keep node logic in `use-interactions.ts`
+- Extract panel UI to separate files
+- Use `_base` components for common patterns
+
+```
+nodes/<node-type>/
+  ├── index.tsx              # Node registration
+  ├── node.tsx               # Node visual component
+  ├── panel.tsx              # Configuration panel
+  ├── use-interactions.ts    # Node-specific hooks
+  └── types.ts               # Type definitions
+```
+
+### 3. Configuration Components
+
+**When**: Refactoring app configuration components.
+
+**Conventions**:
+- Separate config sections into subdirectories
+- Use existing patterns from `web/app/components/app/configuration/`
+- Keep feature toggles in dedicated components
+
+### 4. Tool/Plugin Components
+
+**When**: Refactoring tool-related components (`web/app/components/tools/`).
+
+**Conventions**:
+- Follow existing modal patterns
+- Use service hooks from `web/service/use-tools.ts`
+- Keep provider-specific logic isolated
+
+## Refactoring Workflow
+
+### Step 1: Generate Refactoring Prompt
+
+```bash
+pnpm refactor-component <path>
+```
+
+This command will:
+- Analyze component complexity and features
+- Identify specific refactoring actions needed
+- Generate a prompt for AI assistant (auto-copied to clipboard on macOS)
+- Provide detailed requirements based on detected patterns
+
+### Step 2: Analyze Details
+
+```bash
+pnpm analyze-component <path> --json
+```
+
+Identify:
+- Total complexity score
+- Max function complexity
+- Line count
+- Features detected (state, effects, API, etc.)
+
+### Step 3: Plan
+
+Create a refactoring plan based on detected features:
+
+| Detected Feature | Refactoring Action |
+|------------------|-------------------|
+| `hasState: true` + `hasEffects: true` | Extract custom hook |
+| `hasAPI: true` | Extract data/service hook |
+| `hasEvents: true` (many) | Extract event handlers |
+| `lineCount > 300` | Split into sub-components |
+| `maxComplexity > 50` | Simplify conditional logic |
+
+### Step 4: Execute Incrementally
+
+1. **Extract one piece at a time**
+2. **Run lint, type-check, and tests after each extraction**
+3. **Verify functionality before next step**
+
+```
+For each extraction:
+  ┌────────────────────────────────────────┐
+  │ 1. Extract code                        │
+  │ 2. Run: pnpm lint:fix                  │
+  │ 3. Run: pnpm type-check:tsgo           │
+  │ 4. Run: pnpm test                      │
+  │ 5. Test functionality manually         │
+  │ 6. PASS? → Next extraction             │
+  │    FAIL? → Fix before continuing       │
+  └────────────────────────────────────────┘
+```
+
+### Step 5: Verify
+
+After refactoring:
+
+```bash
+# Re-run refactor command to verify improvements
+pnpm refactor-component <path>
+
+# If complexity < 25 and lines < 200, you'll see:
+# ✅ COMPONENT IS WELL-STRUCTURED
+
+# For detailed metrics:
+pnpm analyze-component <path> --json
+
+# Target metrics:
+# - complexity < 50
+# - lineCount < 300
+# - maxComplexity < 30
+```
+
+## Common Mistakes to Avoid
+
+### ❌ Over-Engineering
+
+```typescript
+// ❌ Too many tiny hooks
+const useButtonText = () => useState('Click')
+const useButtonDisabled = () => useState(false)
+const useButtonLoading = () => useState(false)
+
+// ✅ Cohesive hook with related state
+const useButtonState = () => {
+  const [text, setText] = useState('Click')
+  const [disabled, setDisabled] = useState(false)
+  const [loading, setLoading] = useState(false)
+  return { text, setText, disabled, setDisabled, loading, setLoading }
+}
+```
+
+### ❌ Breaking Existing Patterns
+
+- Follow existing directory structures
+- Maintain naming conventions
+- Preserve export patterns for compatibility
+
+### ❌ Premature Abstraction
+
+- Only extract when there's clear complexity benefit
+- Don't create abstractions for single-use code
+- Keep refactored code in the same domain area
+
+## References
+
+### Dify Codebase Examples
+
+- **Hook extraction**: `web/app/components/app/configuration/hooks/`
+- **Component splitting**: `web/app/components/app/configuration/`
+- **Service hooks**: `web/service/use-*.ts`
+- **Workflow patterns**: `web/app/components/workflow/hooks/`
+- **Form patterns**: `web/app/components/base/form/`
+
+### Related Skills
+
+- `frontend-testing` - For testing refactored components
+- `web/testing/testing.md` - Testing specification

.claude/skills/component-refactoring/references/complexity-patterns.md

@@ -0,0 +1,493 @@
+# Complexity Reduction Patterns
+
+This document provides patterns for reducing cognitive complexity in Dify React components.
+
+## Understanding Complexity
+
+### SonarJS Cognitive Complexity
+
+The `pnpm analyze-component` tool uses SonarJS cognitive complexity metrics:
+
+- **Total Complexity**: Sum of all functions' complexity in the file
+- **Max Complexity**: Highest single function complexity
+
+### What Increases Complexity
+
+| Pattern | Complexity Impact |
+|---------|-------------------|
+| `if/else` | +1 per branch |
+| Nested conditions | +1 per nesting level |
+| `switch/case` | +1 per case |
+| `for/while/do` | +1 per loop |
+| `&&`/`||` chains | +1 per operator |
+| Nested callbacks | +1 per nesting level |
+| `try/catch` | +1 per catch |
+| Ternary expressions | +1 per nesting |
+
+## Pattern 1: Replace Conditionals with Lookup Tables
+
+**Before** (complexity: ~15):
+
+```typescript
+const Template = useMemo(() => {
+  if (appDetail?.mode === AppModeEnum.CHAT) {
+    switch (locale) {
+      case LanguagesSupported[1]:
+        return <TemplateChatZh appDetail={appDetail} />
+      case LanguagesSupported[7]:
+        return <TemplateChatJa appDetail={appDetail} />
+      default:
+        return <TemplateChatEn appDetail={appDetail} />
+    }
+  }
+  if (appDetail?.mode === AppModeEnum.ADVANCED_CHAT) {
+    switch (locale) {
+      case LanguagesSupported[1]:
+        return <TemplateAdvancedChatZh appDetail={appDetail} />
+      case LanguagesSupported[7]:
+        return <TemplateAdvancedChatJa appDetail={appDetail} />
+      default:
+        return <TemplateAdvancedChatEn appDetail={appDetail} />
+    }
+  }
+  if (appDetail?.mode === AppModeEnum.WORKFLOW) {
+    // Similar pattern...
+  }
+  return null
+}, [appDetail, locale])
+```
+
+**After** (complexity: ~3):
+
+```typescript
+// Define lookup table outside component
+const TEMPLATE_MAP: Record<AppModeEnum, Record<string, FC<TemplateProps>>> = {
+  [AppModeEnum.CHAT]: {
+    [LanguagesSupported[1]]: TemplateChatZh,
+    [LanguagesSupported[7]]: TemplateChatJa,
+    default: TemplateChatEn,
+  },
+  [AppModeEnum.ADVANCED_CHAT]: {
+    [LanguagesSupported[1]]: TemplateAdvancedChatZh,
+    [LanguagesSupported[7]]: TemplateAdvancedChatJa,
+    default: TemplateAdvancedChatEn,
+  },
+  [AppModeEnum.WORKFLOW]: {
+    [LanguagesSupported[1]]: TemplateWorkflowZh,
+    [LanguagesSupported[7]]: TemplateWorkflowJa,
+    default: TemplateWorkflowEn,
+  },
+  // ...
+}
+
+// Clean component logic
+const Template = useMemo(() => {
+  if (!appDetail?.mode) return null
+  
+  const templates = TEMPLATE_MAP[appDetail.mode]
+  if (!templates) return null
+  
+  const TemplateComponent = templates[locale] ?? templates.default
+  return <TemplateComponent appDetail={appDetail} />
+}, [appDetail, locale])
+```
+
+## Pattern 2: Use Early Returns
+
+**Before** (complexity: ~10):
+
+```typescript
+const handleSubmit = () => {
+  if (isValid) {
+    if (hasChanges) {
+      if (isConnected) {
+        submitData()
+      } else {
+        showConnectionError()
+      }
+    } else {
+      showNoChangesMessage()
+    }
+  } else {
+    showValidationError()
+  }
+}
+```
+
+**After** (complexity: ~4):
+
+```typescript
+const handleSubmit = () => {
+  if (!isValid) {
+    showValidationError()
+    return
+  }
+  
+  if (!hasChanges) {
+    showNoChangesMessage()
+    return
+  }
+  
+  if (!isConnected) {
+    showConnectionError()
+    return
+  }
+  
+  submitData()
+}
+```
+
+## Pattern 3: Extract Complex Conditions
+
+**Before** (complexity: high):
+
+```typescript
+const canPublish = (() => {
+  if (mode !== AppModeEnum.COMPLETION) {
+    if (!isAdvancedMode)
+      return true
+
+    if (modelModeType === ModelModeType.completion) {
+      if (!hasSetBlockStatus.history || !hasSetBlockStatus.query)
+        return false
+      return true
+    }
+    return true
+  }
+  return !promptEmpty
+})()
+```
+
+**After** (complexity: lower):
+
+```typescript
+// Extract to named functions
+const canPublishInCompletionMode = () => !promptEmpty
+
+const canPublishInChatMode = () => {
+  if (!isAdvancedMode) return true
+  if (modelModeType !== ModelModeType.completion) return true
+  return hasSetBlockStatus.history && hasSetBlockStatus.query
+}
+
+// Clean main logic
+const canPublish = mode === AppModeEnum.COMPLETION
+  ? canPublishInCompletionMode()
+  : canPublishInChatMode()
+```
+
+## Pattern 4: Replace Chained Ternaries
+
+**Before** (complexity: ~5):
+
+```typescript
+const statusText = serverActivated
+  ? t('status.running')
+  : serverPublished
+    ? t('status.inactive')
+    : appUnpublished
+      ? t('status.unpublished')
+      : t('status.notConfigured')
+```
+
+**After** (complexity: ~2):
+
+```typescript
+const getStatusText = () => {
+  if (serverActivated) return t('status.running')
+  if (serverPublished) return t('status.inactive')
+  if (appUnpublished) return t('status.unpublished')
+  return t('status.notConfigured')
+}
+
+const statusText = getStatusText()
+```
+
+Or use lookup:
+
+```typescript
+const STATUS_TEXT_MAP = {
+  running: 'status.running',
+  inactive: 'status.inactive',
+  unpublished: 'status.unpublished',
+  notConfigured: 'status.notConfigured',
+} as const
+
+const getStatusKey = (): keyof typeof STATUS_TEXT_MAP => {
+  if (serverActivated) return 'running'
+  if (serverPublished) return 'inactive'
+  if (appUnpublished) return 'unpublished'
+  return 'notConfigured'
+}
+
+const statusText = t(STATUS_TEXT_MAP[getStatusKey()])
+```
+
+## Pattern 5: Flatten Nested Loops
+
+**Before** (complexity: high):
+
+```typescript
+const processData = (items: Item[]) => {
+  const results: ProcessedItem[] = []
+  
+  for (const item of items) {
+    if (item.isValid) {
+      for (const child of item.children) {
+        if (child.isActive) {
+          for (const prop of child.properties) {
+            if (prop.value !== null) {
+              results.push({
+                itemId: item.id,
+                childId: child.id,
+                propValue: prop.value,
+              })
+            }
+          }
+        }
+      }
+    }
+  }
+  
+  return results
+}
+```
+
+**After** (complexity: lower):
+
+```typescript
+// Use functional approach
+const processData = (items: Item[]) => {
+  return items
+    .filter(item => item.isValid)
+    .flatMap(item =>
+      item.children
+        .filter(child => child.isActive)
+        .flatMap(child =>
+          child.properties
+            .filter(prop => prop.value !== null)
+            .map(prop => ({
+              itemId: item.id,
+              childId: child.id,
+              propValue: prop.value,
+            }))
+        )
+    )
+}
+```
+
+## Pattern 6: Extract Event Handler Logic
+
+**Before** (complexity: high in component):
+
+```typescript
+const Component = () => {
+  const handleSelect = (data: DataSet[]) => {
+    if (isEqual(data.map(item => item.id), dataSets.map(item => item.id))) {
+      hideSelectDataSet()
+      return
+    }
+
+    formattingChangedDispatcher()
+    let newDatasets = data
+    if (data.find(item => !item.name)) {
+      const newSelected = produce(data, (draft) => {
+        data.forEach((item, index) => {
+          if (!item.name) {
+            const newItem = dataSets.find(i => i.id === item.id)
+            if (newItem)
+              draft[index] = newItem
+          }
+        })
+      })
+      setDataSets(newSelected)
+      newDatasets = newSelected
+    }
+    else {
+      setDataSets(data)
+    }
+    hideSelectDataSet()
+    
+    // 40 more lines of logic...
+  }
+  
+  return <div>...</div>
+}
+```
+
+**After** (complexity: lower):
+
+```typescript
+// Extract to hook or utility
+const useDatasetSelection = (dataSets: DataSet[], setDataSets: SetState<DataSet[]>) => {
+  const normalizeSelection = (data: DataSet[]) => {
+    const hasUnloadedItem = data.some(item => !item.name)
+    if (!hasUnloadedItem) return data
+    
+    return produce(data, (draft) => {
+      data.forEach((item, index) => {
+        if (!item.name) {
+          const existing = dataSets.find(i => i.id === item.id)
+          if (existing) draft[index] = existing
+        }
+      })
+    })
+  }
+  
+  const hasSelectionChanged = (newData: DataSet[]) => {
+    return !isEqual(
+      newData.map(item => item.id),
+      dataSets.map(item => item.id)
+    )
+  }
+  
+  return { normalizeSelection, hasSelectionChanged }
+}
+
+// Component becomes cleaner
+const Component = () => {
+  const { normalizeSelection, hasSelectionChanged } = useDatasetSelection(dataSets, setDataSets)
+  
+  const handleSelect = (data: DataSet[]) => {
+    if (!hasSelectionChanged(data)) {
+      hideSelectDataSet()
+      return
+    }
+    
+    formattingChangedDispatcher()
+    const normalized = normalizeSelection(data)
+    setDataSets(normalized)
+    hideSelectDataSet()
+  }
+  
+  return <div>...</div>
+}
+```
+
+## Pattern 7: Reduce Boolean Logic Complexity
+
+**Before** (complexity: ~8):
+
+```typescript
+const toggleDisabled = hasInsufficientPermissions
+  || appUnpublished
+  || missingStartNode
+  || triggerModeDisabled
+  || (isAdvancedApp && !currentWorkflow?.graph)
+  || (isBasicApp && !basicAppConfig.updated_at)
+```
+
+**After** (complexity: ~3):
+
+```typescript
+// Extract meaningful boolean functions
+const isAppReady = () => {
+  if (isAdvancedApp) return !!currentWorkflow?.graph
+  return !!basicAppConfig.updated_at
+}
+
+const hasRequiredPermissions = () => {
+  return isCurrentWorkspaceEditor && !hasInsufficientPermissions
+}
+
+const canToggle = () => {
+  if (!hasRequiredPermissions()) return false
+  if (!isAppReady()) return false
+  if (missingStartNode) return false
+  if (triggerModeDisabled) return false
+  return true
+}
+
+const toggleDisabled = !canToggle()
+```
+
+## Pattern 8: Simplify useMemo/useCallback Dependencies
+
+**Before** (complexity: multiple recalculations):
+
+```typescript
+const payload = useMemo(() => {
+  let parameters: Parameter[] = []
+  let outputParameters: OutputParameter[] = []
+
+  if (!published) {
+    parameters = (inputs || []).map((item) => ({
+      name: item.variable,
+      description: '',
+      form: 'llm',
+      required: item.required,
+      type: item.type,
+    }))
+    outputParameters = (outputs || []).map((item) => ({
+      name: item.variable,
+      description: '',
+      type: item.value_type,
+    }))
+  }
+  else if (detail && detail.tool) {
+    parameters = (inputs || []).map((item) => ({
+      // Complex transformation...
+    }))
+    outputParameters = (outputs || []).map((item) => ({
+      // Complex transformation...
+    }))
+  }
+  
+  return {
+    icon: detail?.icon || icon,
+    label: detail?.label || name,
+    // ...more fields
+  }
+}, [detail, published, workflowAppId, icon, name, description, inputs, outputs])
+```
+
+**After** (complexity: separated concerns):
+
+```typescript
+// Separate transformations
+const useParameterTransform = (inputs: InputVar[], detail?: ToolDetail, published?: boolean) => {
+  return useMemo(() => {
+    if (!published) {
+      return inputs.map(item => ({
+        name: item.variable,
+        description: '',
+        form: 'llm',
+        required: item.required,
+        type: item.type,
+      }))
+    }
+    
+    if (!detail?.tool) return []
+    
+    return inputs.map(item => ({
+      name: item.variable,
+      required: item.required,
+      type: item.type === 'paragraph' ? 'string' : item.type,
+      description: detail.tool.parameters.find(p => p.name === item.variable)?.llm_description || '',
+      form: detail.tool.parameters.find(p => p.name === item.variable)?.form || 'llm',
+    }))
+  }, [inputs, detail, published])
+}
+
+// Component uses hook
+const parameters = useParameterTransform(inputs, detail, published)
+const outputParameters = useOutputTransform(outputs, detail, published)
+
+const payload = useMemo(() => ({
+  icon: detail?.icon || icon,
+  label: detail?.label || name,
+  parameters,
+  outputParameters,
+  // ...
+}), [detail, icon, name, parameters, outputParameters])
+```
+
+## Target Metrics After Refactoring
+
+| Metric | Target |
+|--------|--------|
+| Total Complexity | < 50 |
+| Max Function Complexity | < 30 |
+| Function Length | < 30 lines |
+| Nesting Depth | ≤ 3 levels |
+| Conditional Chains | ≤ 3 conditions |

.claude/skills/component-refactoring/references/component-splitting.md

@@ -0,0 +1,477 @@
+# Component Splitting Patterns
+
+This document provides detailed guidance on splitting large components into smaller, focused components in Dify.
+
+## When to Split Components
+
+Split a component when you identify:
+
+1. **Multiple UI sections** - Distinct visual areas with minimal coupling that can be composed independently
+1. **Conditional rendering blocks** - Large `{condition && <JSX />}` blocks
+1. **Repeated patterns** - Similar UI structures used multiple times
+1. **300+ lines** - Component exceeds manageable size
+1. **Modal clusters** - Multiple modals rendered in one component
+
+## Splitting Strategies
+
+### Strategy 1: Section-Based Splitting
+
+Identify visual sections and extract each as a component.
+
+```typescript
+// ❌ Before: Monolithic component (500+ lines)
+const ConfigurationPage = () => {
+  return (
+    <div>
+      {/* Header Section - 50 lines */}
+      <div className="header">
+        <h1>{t('configuration.title')}</h1>
+        <div className="actions">
+          {isAdvancedMode && <Badge>Advanced</Badge>}
+          <ModelParameterModal ... />
+          <AppPublisher ... />
+        </div>
+      </div>
+      
+      {/* Config Section - 200 lines */}
+      <div className="config">
+        <Config />
+      </div>
+      
+      {/* Debug Section - 150 lines */}
+      <div className="debug">
+        <Debug ... />
+      </div>
+      
+      {/* Modals Section - 100 lines */}
+      {showSelectDataSet && <SelectDataSet ... />}
+      {showHistoryModal && <EditHistoryModal ... />}
+      {showUseGPT4Confirm && <Confirm ... />}
+    </div>
+  )
+}
+
+// ✅ After: Split into focused components
+// configuration/
+//   ├── index.tsx              (orchestration)
+//   ├── configuration-header.tsx
+//   ├── configuration-content.tsx
+//   ├── configuration-debug.tsx
+//   └── configuration-modals.tsx
+
+// configuration-header.tsx
+interface ConfigurationHeaderProps {
+  isAdvancedMode: boolean
+  onPublish: () => void
+}
+
+const ConfigurationHeader: FC<ConfigurationHeaderProps> = ({
+  isAdvancedMode,
+  onPublish,
+}) => {
+  const { t } = useTranslation()
+  
+  return (
+    <div className="header">
+      <h1>{t('configuration.title')}</h1>
+      <div className="actions">
+        {isAdvancedMode && <Badge>Advanced</Badge>}
+        <ModelParameterModal ... />
+        <AppPublisher onPublish={onPublish} />
+      </div>
+    </div>
+  )
+}
+
+// index.tsx (orchestration only)
+const ConfigurationPage = () => {
+  const { modelConfig, setModelConfig } = useModelConfig()
+  const { activeModal, openModal, closeModal } = useModalState()
+  
+  return (
+    <div>
+      <ConfigurationHeader
+        isAdvancedMode={isAdvancedMode}
+        onPublish={handlePublish}
+      />
+      <ConfigurationContent
+        modelConfig={modelConfig}
+        onConfigChange={setModelConfig}
+      />
+      {!isMobile && (
+        <ConfigurationDebug
+          inputs={inputs}
+          onSetting={handleSetting}
+        />
+      )}
+      <ConfigurationModals
+        activeModal={activeModal}
+        onClose={closeModal}
+      />
+    </div>
+  )
+}
+```
+
+### Strategy 2: Conditional Block Extraction
+
+Extract large conditional rendering blocks.
+
+```typescript
+// ❌ Before: Large conditional blocks
+const AppInfo = () => {
+  return (
+    <div>
+      {expand ? (
+        <div className="expanded">
+          {/* 100 lines of expanded view */}
+        </div>
+      ) : (
+        <div className="collapsed">
+          {/* 50 lines of collapsed view */}
+        </div>
+      )}
+    </div>
+  )
+}
+
+// ✅ After: Separate view components
+const AppInfoExpanded: FC<AppInfoViewProps> = ({ appDetail, onAction }) => {
+  return (
+    <div className="expanded">
+      {/* Clean, focused expanded view */}
+    </div>
+  )
+}
+
+const AppInfoCollapsed: FC<AppInfoViewProps> = ({ appDetail, onAction }) => {
+  return (
+    <div className="collapsed">
+      {/* Clean, focused collapsed view */}
+    </div>
+  )
+}
+
+const AppInfo = () => {
+  return (
+    <div>
+      {expand
+        ? <AppInfoExpanded appDetail={appDetail} onAction={handleAction} />
+        : <AppInfoCollapsed appDetail={appDetail} onAction={handleAction} />
+      }
+    </div>
+  )
+}
+```
+
+### Strategy 3: Modal Extraction
+
+Extract modals with their trigger logic.
+
+```typescript
+// ❌ Before: Multiple modals in one component
+const AppInfo = () => {
+  const [showEdit, setShowEdit] = useState(false)
+  const [showDuplicate, setShowDuplicate] = useState(false)
+  const [showDelete, setShowDelete] = useState(false)
+  const [showSwitch, setShowSwitch] = useState(false)
+  
+  const onEdit = async (data) => { /* 20 lines */ }
+  const onDuplicate = async (data) => { /* 20 lines */ }
+  const onDelete = async () => { /* 15 lines */ }
+  
+  return (
+    <div>
+      {/* Main content */}
+      
+      {showEdit && <EditModal onConfirm={onEdit} onClose={() => setShowEdit(false)} />}
+      {showDuplicate && <DuplicateModal onConfirm={onDuplicate} onClose={() => setShowDuplicate(false)} />}
+      {showDelete && <DeleteConfirm onConfirm={onDelete} onClose={() => setShowDelete(false)} />}
+      {showSwitch && <SwitchModal ... />}
+    </div>
+  )
+}
+
+// ✅ After: Modal manager component
+// app-info-modals.tsx
+type ModalType = 'edit' | 'duplicate' | 'delete' | 'switch' | null
+
+interface AppInfoModalsProps {
+  appDetail: AppDetail
+  activeModal: ModalType
+  onClose: () => void
+  onSuccess: () => void
+}
+
+const AppInfoModals: FC<AppInfoModalsProps> = ({
+  appDetail,
+  activeModal,
+  onClose,
+  onSuccess,
+}) => {
+  const handleEdit = async (data) => { /* logic */ }
+  const handleDuplicate = async (data) => { /* logic */ }
+  const handleDelete = async () => { /* logic */ }
+
+  return (
+    <>
+      {activeModal === 'edit' && (
+        <EditModal
+          appDetail={appDetail}
+          onConfirm={handleEdit}
+          onClose={onClose}
+        />
+      )}
+      {activeModal === 'duplicate' && (
+        <DuplicateModal
+          appDetail={appDetail}
+          onConfirm={handleDuplicate}
+          onClose={onClose}
+        />
+      )}
+      {activeModal === 'delete' && (
+        <DeleteConfirm
+          onConfirm={handleDelete}
+          onClose={onClose}
+        />
+      )}
+      {activeModal === 'switch' && (
+        <SwitchModal
+          appDetail={appDetail}
+          onClose={onClose}
+        />
+      )}
+    </>
+  )
+}
+
+// Parent component
+const AppInfo = () => {
+  const { activeModal, openModal, closeModal } = useModalState()
+  
+  return (
+    <div>
+      {/* Main content with openModal triggers */}
+      <Button onClick={() => openModal('edit')}>Edit</Button>
+      
+      <AppInfoModals
+        appDetail={appDetail}
+        activeModal={activeModal}
+        onClose={closeModal}
+        onSuccess={handleSuccess}
+      />
+    </div>
+  )
+}
+```
+
+### Strategy 4: List Item Extraction
+
+Extract repeated item rendering.
+
+```typescript
+// ❌ Before: Inline item rendering
+const OperationsList = () => {
+  return (
+    <div>
+      {operations.map(op => (
+        <div key={op.id} className="operation-item">
+          <span className="icon">{op.icon}</span>
+          <span className="title">{op.title}</span>
+          <span className="description">{op.description}</span>
+          <button onClick={() => op.onClick()}>
+            {op.actionLabel}
+          </button>
+          {op.badge && <Badge>{op.badge}</Badge>}
+          {/* More complex rendering... */}
+        </div>
+      ))}
+    </div>
+  )
+}
+
+// ✅ After: Extracted item component
+interface OperationItemProps {
+  operation: Operation
+  onAction: (id: string) => void
+}
+
+const OperationItem: FC<OperationItemProps> = ({ operation, onAction }) => {
+  return (
+    <div className="operation-item">
+      <span className="icon">{operation.icon}</span>
+      <span className="title">{operation.title}</span>
+      <span className="description">{operation.description}</span>
+      <button onClick={() => onAction(operation.id)}>
+        {operation.actionLabel}
+      </button>
+      {operation.badge && <Badge>{operation.badge}</Badge>}
+    </div>
+  )
+}
+
+const OperationsList = () => {
+  const handleAction = useCallback((id: string) => {
+    const op = operations.find(o => o.id === id)
+    op?.onClick()
+  }, [operations])
+
+  return (
+    <div>
+      {operations.map(op => (
+        <OperationItem
+          key={op.id}
+          operation={op}
+          onAction={handleAction}
+        />
+      ))}
+    </div>
+  )
+}
+```
+
+## Directory Structure Patterns
+
+### Pattern A: Flat Structure (Simple Components)
+
+For components with 2-3 sub-components:
+
+```
+component-name/
+  ├── index.tsx           # Main component
+  ├── sub-component-a.tsx
+  ├── sub-component-b.tsx
+  └── types.ts            # Shared types
+```
+
+### Pattern B: Nested Structure (Complex Components)
+
+For components with many sub-components:
+
+```
+component-name/
+  ├── index.tsx           # Main orchestration
+  ├── types.ts            # Shared types
+  ├── hooks/
+  │   ├── use-feature-a.ts
+  │   └── use-feature-b.ts
+  ├── components/
+  │   ├── header/
+  │   │   └── index.tsx
+  │   ├── content/
+  │   │   └── index.tsx
+  │   └── modals/
+  │       └── index.tsx
+  └── utils/
+      └── helpers.ts
+```
+
+### Pattern C: Feature-Based Structure (Dify Standard)
+
+Following Dify's existing patterns:
+
+```
+configuration/
+  ├── index.tsx           # Main page component
+  ├── base/               # Base/shared components
+  │   ├── feature-panel/
+  │   ├── group-name/
+  │   └── operation-btn/
+  ├── config/             # Config section
+  │   ├── index.tsx
+  │   ├── agent/
+  │   └── automatic/
+  ├── dataset-config/     # Dataset section
+  │   ├── index.tsx
+  │   ├── card-item/
+  │   └── params-config/
+  ├── debug/              # Debug section
+  │   ├── index.tsx
+  │   └── hooks.tsx
+  └── hooks/              # Shared hooks
+      └── use-advanced-prompt-config.ts
+```
+
+## Props Design
+
+### Minimal Props Principle
+
+Pass only what's needed:
+
+```typescript
+// ❌ Bad: Passing entire objects when only some fields needed
+<ConfigHeader appDetail={appDetail} modelConfig={modelConfig} />
+
+// ✅ Good: Destructure to minimum required
+<ConfigHeader
+  appName={appDetail.name}
+  isAdvancedMode={modelConfig.isAdvanced}
+  onPublish={handlePublish}
+/>
+```
+
+### Callback Props Pattern
+
+Use callbacks for child-to-parent communication:
+
+```typescript
+// Parent
+const Parent = () => {
+  const [value, setValue] = useState('')
+  
+  return (
+    <Child
+      value={value}
+      onChange={setValue}
+      onSubmit={handleSubmit}
+    />
+  )
+}
+
+// Child
+interface ChildProps {
+  value: string
+  onChange: (value: string) => void
+  onSubmit: () => void
+}
+
+const Child: FC<ChildProps> = ({ value, onChange, onSubmit }) => {
+  return (
+    <div>
+      <input value={value} onChange={e => onChange(e.target.value)} />
+      <button onClick={onSubmit}>Submit</button>
+    </div>
+  )
+}
+```
+
+### Render Props for Flexibility
+
+When sub-components need parent context:
+
+```typescript
+interface ListProps<T> {
+  items: T[]
+  renderItem: (item: T, index: number) => React.ReactNode
+  renderEmpty?: () => React.ReactNode
+}
+
+function List<T>({ items, renderItem, renderEmpty }: ListProps<T>) {
+  if (items.length === 0 && renderEmpty) {
+    return <>{renderEmpty()}</>
+  }
+  
+  return (
+    <div>
+      {items.map((item, index) => renderItem(item, index))}
+    </div>
+  )
+}
+
+// Usage
+<List
+  items={operations}
+  renderItem={(op, i) => <OperationItem key={i} operation={op} />}
+  renderEmpty={() => <EmptyState message="No operations" />}
+/>
+```

.claude/skills/component-refactoring/references/hook-extraction.md

@@ -0,0 +1,317 @@
+# Hook Extraction Patterns
+
+This document provides detailed guidance on extracting custom hooks from complex components in Dify.
+
+## When to Extract Hooks
+
+Extract a custom hook when you identify:
+
+1. **Coupled state groups** - Multiple `useState` hooks that are always used together
+1. **Complex effects** - `useEffect` with multiple dependencies or cleanup logic
+1. **Business logic** - Data transformations, validations, or calculations
+1. **Reusable patterns** - Logic that appears in multiple components
+
+## Extraction Process
+
+### Step 1: Identify State Groups
+
+Look for state variables that are logically related:
+
+```typescript
+// ❌ These belong together - extract to hook
+const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+const [completionParams, setCompletionParams] = useState<FormValue>({})
+const [modelModeType, setModelModeType] = useState<ModelModeType>(...)
+
+// These are model-related state that should be in useModelConfig()
+```
+
+### Step 2: Identify Related Effects
+
+Find effects that modify the grouped state:
+
+```typescript
+// ❌ These effects belong with the state above
+useEffect(() => {
+  if (hasFetchedDetail && !modelModeType) {
+    const mode = currModel?.model_properties.mode
+    if (mode) {
+      const newModelConfig = produce(modelConfig, (draft) => {
+        draft.mode = mode
+      })
+      setModelConfig(newModelConfig)
+    }
+  }
+}, [textGenerationModelList, hasFetchedDetail, modelModeType, currModel])
+```
+
+### Step 3: Create the Hook
+
+```typescript
+// hooks/use-model-config.ts
+import type { FormValue } from '@/app/components/header/account-setting/model-provider-page/declarations'
+import type { ModelConfig } from '@/models/debug'
+import { produce } from 'immer'
+import { useEffect, useState } from 'react'
+import { ModelModeType } from '@/types/app'
+
+interface UseModelConfigParams {
+  initialConfig?: Partial<ModelConfig>
+  currModel?: { model_properties?: { mode?: ModelModeType } }
+  hasFetchedDetail: boolean
+}
+
+interface UseModelConfigReturn {
+  modelConfig: ModelConfig
+  setModelConfig: (config: ModelConfig) => void
+  completionParams: FormValue
+  setCompletionParams: (params: FormValue) => void
+  modelModeType: ModelModeType
+}
+
+export const useModelConfig = ({
+  initialConfig,
+  currModel,
+  hasFetchedDetail,
+}: UseModelConfigParams): UseModelConfigReturn => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>({
+    provider: 'langgenius/openai/openai',
+    model_id: 'gpt-3.5-turbo',
+    mode: ModelModeType.unset,
+    // ... default values
+    ...initialConfig,
+  })
+  
+  const [completionParams, setCompletionParams] = useState<FormValue>({})
+  
+  const modelModeType = modelConfig.mode
+
+  // Fill old app data missing model mode
+  useEffect(() => {
+    if (hasFetchedDetail && !modelModeType) {
+      const mode = currModel?.model_properties?.mode
+      if (mode) {
+        setModelConfig(produce(modelConfig, (draft) => {
+          draft.mode = mode
+        }))
+      }
+    }
+  }, [hasFetchedDetail, modelModeType, currModel])
+
+  return {
+    modelConfig,
+    setModelConfig,
+    completionParams,
+    setCompletionParams,
+    modelModeType,
+  }
+}
+```
+
+### Step 4: Update Component
+
+```typescript
+// Before: 50+ lines of state management
+const Configuration: FC = () => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+  // ... lots of related state and effects
+}
+
+// After: Clean component
+const Configuration: FC = () => {
+  const {
+    modelConfig,
+    setModelConfig,
+    completionParams,
+    setCompletionParams,
+    modelModeType,
+  } = useModelConfig({
+    currModel,
+    hasFetchedDetail,
+  })
+  
+  // Component now focuses on UI
+}
+```
+
+## Naming Conventions
+
+### Hook Names
+
+- Use `use` prefix: `useModelConfig`, `useDatasetConfig`
+- Be specific: `useAdvancedPromptConfig` not `usePrompt`
+- Include domain: `useWorkflowVariables`, `useMCPServer`
+
+### File Names
+
+- Kebab-case: `use-model-config.ts`
+- Place in `hooks/` subdirectory when multiple hooks exist
+- Place alongside component for single-use hooks
+
+### Return Type Names
+
+- Suffix with `Return`: `UseModelConfigReturn`
+- Suffix params with `Params`: `UseModelConfigParams`
+
+## Common Hook Patterns in Dify
+
+### 1. Data Fetching Hook (React Query)
+
+```typescript
+// Pattern: Use @tanstack/react-query for data fetching
+import { useQuery, useQueryClient } from '@tanstack/react-query'
+import { get } from '@/service/base'
+import { useInvalid } from '@/service/use-base'
+
+const NAME_SPACE = 'appConfig'
+
+// Query keys for cache management
+export const appConfigQueryKeys = {
+  detail: (appId: string) => [NAME_SPACE, 'detail', appId] as const,
+}
+
+// Main data hook
+export const useAppConfig = (appId: string) => {
+  return useQuery({
+    enabled: !!appId,
+    queryKey: appConfigQueryKeys.detail(appId),
+    queryFn: () => get<AppDetailResponse>(`/apps/${appId}`),
+    select: data => data?.model_config || null,
+  })
+}
+
+// Invalidation hook for refreshing data
+export const useInvalidAppConfig = () => {
+  return useInvalid([NAME_SPACE])
+}
+
+// Usage in component
+const Component = () => {
+  const { data: config, isLoading, error, refetch } = useAppConfig(appId)
+  const invalidAppConfig = useInvalidAppConfig()
+  
+  const handleRefresh = () => {
+    invalidAppConfig() // Invalidates cache and triggers refetch
+  }
+  
+  return <div>...</div>
+}
+```
+
+### 2. Form State Hook
+
+```typescript
+// Pattern: Form state + validation + submission
+export const useConfigForm = (initialValues: ConfigFormValues) => {
+  const [values, setValues] = useState(initialValues)
+  const [errors, setErrors] = useState<Record<string, string>>({})
+  const [isSubmitting, setIsSubmitting] = useState(false)
+
+  const validate = useCallback(() => {
+    const newErrors: Record<string, string> = {}
+    if (!values.name) newErrors.name = 'Name is required'
+    setErrors(newErrors)
+    return Object.keys(newErrors).length === 0
+  }, [values])
+
+  const handleChange = useCallback((field: string, value: any) => {
+    setValues(prev => ({ ...prev, [field]: value }))
+  }, [])
+
+  const handleSubmit = useCallback(async (onSubmit: (values: ConfigFormValues) => Promise<void>) => {
+    if (!validate()) return
+    setIsSubmitting(true)
+    try {
+      await onSubmit(values)
+    } finally {
+      setIsSubmitting(false)
+    }
+  }, [values, validate])
+
+  return { values, errors, isSubmitting, handleChange, handleSubmit }
+}
+```
+
+### 3. Modal State Hook
+
+```typescript
+// Pattern: Multiple modal management
+type ModalType = 'edit' | 'delete' | 'duplicate' | null
+
+export const useModalState = () => {
+  const [activeModal, setActiveModal] = useState<ModalType>(null)
+  const [modalData, setModalData] = useState<any>(null)
+
+  const openModal = useCallback((type: ModalType, data?: any) => {
+    setActiveModal(type)
+    setModalData(data)
+  }, [])
+
+  const closeModal = useCallback(() => {
+    setActiveModal(null)
+    setModalData(null)
+  }, [])
+
+  return {
+    activeModal,
+    modalData,
+    openModal,
+    closeModal,
+    isOpen: useCallback((type: ModalType) => activeModal === type, [activeModal]),
+  }
+}
+```
+
+### 4. Toggle/Boolean Hook
+
+```typescript
+// Pattern: Boolean state with convenience methods
+export const useToggle = (initialValue = false) => {
+  const [value, setValue] = useState(initialValue)
+
+  const toggle = useCallback(() => setValue(v => !v), [])
+  const setTrue = useCallback(() => setValue(true), [])
+  const setFalse = useCallback(() => setValue(false), [])
+
+  return [value, { toggle, setTrue, setFalse, set: setValue }] as const
+}
+
+// Usage
+const [isExpanded, { toggle, setTrue: expand, setFalse: collapse }] = useToggle()
+```
+
+## Testing Extracted Hooks
+
+After extraction, test hooks in isolation:
+
+```typescript
+// use-model-config.spec.ts
+import { renderHook, act } from '@testing-library/react'
+import { useModelConfig } from './use-model-config'
+
+describe('useModelConfig', () => {
+  it('should initialize with default values', () => {
+    const { result } = renderHook(() => useModelConfig({
+      hasFetchedDetail: false,
+    }))
+
+    expect(result.current.modelConfig.provider).toBe('langgenius/openai/openai')
+    expect(result.current.modelModeType).toBe(ModelModeType.unset)
+  })
+
+  it('should update model config', () => {
+    const { result } = renderHook(() => useModelConfig({
+      hasFetchedDetail: true,
+    }))
+
+    act(() => {
+      result.current.setModelConfig({
+        ...result.current.modelConfig,
+        model_id: 'gpt-4',
+      })
+    })
+
+    expect(result.current.modelConfig.model_id).toBe('gpt-4')
+  })
+})
+```

.claude/skills/frontend-testing/SKILL.md

@@ -318,5 +318,5 @@ For more detailed information, refer to:
 
 - `web/vitest.config.ts` - Vitest configuration
 - `web/vitest.setup.ts` - Test environment setup
-- `web/testing/analyze-component.js` - Component analysis tool
+- `web/scripts/analyze-component.js` - Component analysis tool
 - Modules are not mocked automatically. Global mocks live in `web/vitest.setup.ts` (for example `react-i18next`, `next/image`); mock other modules like `ky` or `mime` locally in test files.

.claude/skills/frontend-testing/assets/component-test.template.tsx

@@ -28,17 +28,14 @@ import userEvent from '@testing-library/user-event'
 
 // i18n (automatically mocked)
 // WHY: Global mock in web/vitest.setup.ts is auto-loaded by Vitest setup
-// No explicit mock needed - it returns translation keys as-is
+// The global mock provides: useTranslation, Trans, useMixedTranslation, useGetLanguage
+// No explicit mock needed for most tests
+//
 // Override only if custom translations are required:
-// vi.mock('react-i18next', () => ({
-//   useTranslation: () => ({
-//     t: (key: string) => {
-//       const customTranslations: Record<string, string> = {
-//         'my.custom.key': 'Custom Translation',
-//       }
-//       return customTranslations[key] || key
-//     },
-//   }),
+// import { createReactI18nextMock } from '@/test/i18n-mock'
+// vi.mock('react-i18next', () => createReactI18nextMock({
+//   'my.custom.key': 'Custom Translation',
+//   'button.save': 'Save',
 // }))
 
 // Router (if component uses useRouter, usePathname, useSearchParams)

.claude/skills/frontend-testing/references/mocking.md

@@ -52,23 +52,29 @@ Modules are not mocked automatically. Use `vi.mock` in test files, or add global
 ### 1. i18n (Auto-loaded via Global Mock)
 
 A global mock is defined in `web/vitest.setup.ts` and is auto-loaded by Vitest setup.
-**No explicit mock needed** for most tests - it returns translation keys as-is.
 
-For tests requiring custom translations, override the mock:
+The global mock provides:
+
+- `useTranslation` - returns translation keys with namespace prefix
+- `Trans` component - renders i18nKey and components
+- `useMixedTranslation` (from `@/app/components/plugins/marketplace/hooks`)
+- `useGetLanguage` (from `@/context/i18n`) - returns `'en-US'`
+
+**Default behavior**: Most tests should use the global mock (no local override needed).
+
+**For custom translations**: Use the helper function from `@/test/i18n-mock`:
 
 ```typescript
-vi.mock('react-i18next', () => ({
-  useTranslation: () => ({
-    t: (key: string) => {
-      const translations: Record<string, string> = {
-        'my.custom.key': 'Custom translation',
-      }
-      return translations[key] || key
-    },
-  }),
+import { createReactI18nextMock } from '@/test/i18n-mock'
+
+vi.mock('react-i18next', () => createReactI18nextMock({
+  'my.custom.key': 'Custom translation',
+  'button.save': 'Save',
 }))
 ```
 
+**Avoid**: Manually defining `useTranslation` mocks that just return the key - the global mock already does this.
+
 ### 2. Next.js Router
 
 ```typescript

.github/workflows/api-tests.yml

@@ -22,12 +22,12 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: ${{ matrix.python-version }}
@@ -57,7 +57,7 @@ jobs:
         run: sh .github/workflows/expose_service_ports.sh
 
       - name: Set up Sandbox
-        uses: hoverkraft-tech/compose-action@v2.0.2
+        uses: hoverkraft-tech/compose-action@v2
         with:
           compose-file: |
             docker/docker-compose.middleware.yaml

.github/workflows/autofix.yml

@@ -12,7 +12,7 @@ jobs:
     if: github.repository == 'langgenius/dify'
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
 
       - name: Check Docker Compose inputs
         id: docker-compose-changes
@@ -27,7 +27,7 @@ jobs:
         with:
           python-version: "3.11"
 
-      - uses: astral-sh/setup-uv@v6
+      - uses: astral-sh/setup-uv@v7
 
       - name: Generate Docker Compose
         if: steps.docker-compose-changes.outputs.any_changed == 'true'

.github/workflows/build-push.yml

@@ -90,7 +90,7 @@ jobs:
           touch "/tmp/digests/${sanitized_digest}"
 
       - name: Upload digest
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v6
         with:
           name: digests-${{ matrix.context }}-${{ env.PLATFORM_PAIR }}
           path: /tmp/digests/*

.github/workflows/db-migration-test.yml

@@ -13,13 +13,13 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: "3.12"
@@ -63,13 +63,13 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: "3.12"

.github/workflows/main-ci.yml

@@ -27,7 +27,7 @@ jobs:
       vdb-changed: ${{ steps.changes.outputs.vdb }}
       migration-changed: ${{ steps.changes.outputs.migration }}
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
       - uses: dorny/paths-filter@v3
         id: changes
         with:
@@ -38,6 +38,7 @@ jobs:
               - '.github/workflows/api-tests.yml'
             web:
               - 'web/**'
+              - '.github/workflows/web-tests.yml'
             vdb:
               - 'api/core/rag/datasource/**'
               - 'docker/**'

.github/workflows/style.yml

@@ -19,13 +19,13 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Check changed files
         id: changed-files
-        uses: tj-actions/changed-files@v46
+        uses: tj-actions/changed-files@v47
         with:
           files: |
             api/**
@@ -33,7 +33,7 @@ jobs:
 
       - name: Setup UV and Python
         if: steps.changed-files.outputs.any_changed == 'true'
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: false
           python-version: "3.12"
@@ -68,15 +68,17 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Check changed files
         id: changed-files
-        uses: tj-actions/changed-files@v46
+        uses: tj-actions/changed-files@v47
         with:
-          files: web/**
+          files: |
+            web/**
+            .github/workflows/style.yml
 
       - name: Install pnpm
         uses: pnpm/action-setup@v4
@@ -85,7 +87,7 @@ jobs:
           run_install: false
 
       - name: Setup NodeJS
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         if: steps.changed-files.outputs.any_changed == 'true'
         with:
           node-version: 22
@@ -108,20 +110,30 @@ jobs:
         working-directory: ./web
         run: pnpm run type-check:tsgo
 
+      - name: Web dead code check
+        if: steps.changed-files.outputs.any_changed == 'true'
+        working-directory: ./web
+        run: pnpm run knip
+
+      - name: Web build check
+        if: steps.changed-files.outputs.any_changed == 'true'
+        working-directory: ./web
+        run: pnpm run build
+
   superlinter:
     name: SuperLinter
     runs-on: ubuntu-latest
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
 
       - name: Check changed files
         id: changed-files
-        uses: tj-actions/changed-files@v46
+        uses: tj-actions/changed-files@v47
         with:
           files: |
             **.sh

.github/workflows/tool-test-sdks.yaml

@@ -25,12 +25,12 @@ jobs:
         working-directory: sdks/nodejs-client
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Use Node.js ${{ matrix.node-version }}
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         with:
           node-version: ${{ matrix.node-version }}
           cache: ''

.github/workflows/translate-i18n-base-on-english.yml

@@ -4,7 +4,7 @@ on:
   push:
     branches: [main]
     paths:
-      - 'web/i18n/en-US/*.ts'
+      - 'web/i18n/en-US/*.json'
 
 permissions:
   contents: write
@@ -18,7 +18,7 @@ jobs:
       run:
         working-directory: web
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
         with:
           fetch-depth: 0
           token: ${{ secrets.GITHUB_TOKEN }}
@@ -28,13 +28,13 @@ jobs:
         run: |
           git fetch origin "${{ github.event.before }}" || true
           git fetch origin "${{ github.sha }}" || true
-          changed_files=$(git diff --name-only "${{ github.event.before }}" "${{ github.sha }}" -- 'i18n/en-US/*.ts')
+          changed_files=$(git diff --name-only "${{ github.event.before }}" "${{ github.sha }}" -- 'i18n/en-US/*.json')
           echo "Changed files: $changed_files"
           if [ -n "$changed_files" ]; then
             echo "FILES_CHANGED=true" >> $GITHUB_ENV
             file_args=""
             for file in $changed_files; do
-              filename=$(basename "$file" .ts)
+              filename=$(basename "$file" .json)
               file_args="$file_args --file $filename"
             done
             echo "FILE_ARGS=$file_args" >> $GITHUB_ENV
@@ -51,7 +51,7 @@ jobs:
 
       - name: Set up Node.js
         if: env.FILES_CHANGED == 'true'
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         with:
           node-version: 'lts/*'
           cache: pnpm
@@ -65,7 +65,7 @@ jobs:
       - name: Generate i18n translations
         if: env.FILES_CHANGED == 'true'
         working-directory: ./web
-        run: pnpm run auto-gen-i18n ${{ env.FILE_ARGS }}
+        run: pnpm run i18n:gen ${{ env.FILE_ARGS }}
 
       - name: Create Pull Request
         if: env.FILES_CHANGED == 'true'

.github/workflows/vdb-tests.yml

@@ -19,19 +19,19 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Free Disk Space
-        uses: endersonmenezes/free-disk-space@v2
+        uses: endersonmenezes/free-disk-space@v3
         with:
           remove_dotnet: true
           remove_haskell: true
           remove_tool_cache: true
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: ${{ matrix.python-version }}

.github/workflows/web-tests.yml

@@ -18,7 +18,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -29,7 +29,7 @@ jobs:
           run_install: false
 
       - name: Setup Node.js
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         with:
           node-version: 22
           cache: pnpm
@@ -360,7 +360,7 @@ jobs:
 
       - name: Upload Coverage Artifact
         if: steps.coverage-summary.outputs.has_coverage == 'true'
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v6
         with:
           name: web-coverage-report
           path: web/coverage

.mcp.json

@@ -1,34 +0,0 @@
-{
-    "mcpServers": {
-      "context7": {
-        "type": "http",
-        "url": "https://mcp.context7.com/mcp"
-      },
-      "sequential-thinking": {
-        "type": "stdio",
-        "command": "npx",
-        "args": ["-y", "@modelcontextprotocol/server-sequential-thinking"],
-        "env": {}
-      },
-      "github": {
-        "type": "stdio",
-        "command": "npx",
-        "args": ["-y", "@modelcontextprotocol/server-github"],
-        "env": {
-          "GITHUB_PERSONAL_ACCESS_TOKEN": "${GITHUB_PERSONAL_ACCESS_TOKEN}"
-        }
-      },
-      "fetch": {
-        "type": "stdio",
-        "command": "uvx",
-        "args": ["mcp-server-fetch"],
-        "env": {}
-      },
-      "playwright": {
-        "type": "stdio",
-        "command": "npx",
-        "args": ["-y", "@playwright/mcp@latest"],
-        "env": {}
-      }
-    }
-  }

api/.env.example

@@ -101,6 +101,15 @@ S3_ACCESS_KEY=your-access-key
 S3_SECRET_KEY=your-secret-key
 S3_REGION=your-region
 
+# Workflow run and Conversation archive storage (S3-compatible)
+ARCHIVE_STORAGE_ENABLED=false
+ARCHIVE_STORAGE_ENDPOINT=
+ARCHIVE_STORAGE_ARCHIVE_BUCKET=
+ARCHIVE_STORAGE_EXPORT_BUCKET=
+ARCHIVE_STORAGE_ACCESS_KEY=
+ARCHIVE_STORAGE_SECRET_KEY=
+ARCHIVE_STORAGE_REGION=auto
+
 # Azure Blob Storage configuration
 AZURE_BLOB_ACCOUNT_NAME=your-account-name
 AZURE_BLOB_ACCOUNT_KEY=your-account-key
@@ -128,6 +137,7 @@ TENCENT_COS_SECRET_KEY=your-secret-key
 TENCENT_COS_SECRET_ID=your-secret-id
 TENCENT_COS_REGION=your-region
 TENCENT_COS_SCHEME=your-scheme
+TENCENT_COS_CUSTOM_DOMAIN=your-custom-domain
 
 # Huawei OBS Storage Configuration
 HUAWEI_OBS_BUCKET_NAME=your-bucket-name

api/.ruff.toml

@@ -1,4 +1,8 @@
-exclude = ["migrations/*"]
+exclude = [
+    "migrations/*",
+    ".git",
+    ".git/**",
+]
 line-length = 120
 
 [format]

api/configs/extra/__init__.py

@@ -1,9 +1,11 @@
+from configs.extra.archive_config import ArchiveStorageConfig
 from configs.extra.notion_config import NotionConfig
 from configs.extra.sentry_config import SentryConfig
 
 
 class ExtraServiceConfig(
     # place the configs in alphabet order
+    ArchiveStorageConfig,
     NotionConfig,
     SentryConfig,
 ):

api/configs/extra/archive_config.py

@@ -0,0 +1,43 @@
+from pydantic import Field
+from pydantic_settings import BaseSettings
+
+
+class ArchiveStorageConfig(BaseSettings):
+    """
+    Configuration settings for workflow run logs archiving storage.
+    """
+
+    ARCHIVE_STORAGE_ENABLED: bool = Field(
+        description="Enable workflow run logs archiving to S3-compatible storage",
+        default=False,
+    )
+
+    ARCHIVE_STORAGE_ENDPOINT: str | None = Field(
+        description="URL of the S3-compatible storage endpoint (e.g., 'https://storage.example.com')",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_ARCHIVE_BUCKET: str | None = Field(
+        description="Name of the bucket to store archived workflow logs",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_EXPORT_BUCKET: str | None = Field(
+        description="Name of the bucket to store exported workflow runs",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_ACCESS_KEY: str | None = Field(
+        description="Access key ID for authenticating with storage",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_SECRET_KEY: str | None = Field(
+        description="Secret access key for authenticating with storage",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_REGION: str = Field(
+        description="Region for storage (use 'auto' if the provider supports it)",
+        default="auto",
+    )

api/configs/middleware/storage/tencent_cos_storage_config.py

@@ -31,3 +31,8 @@ class TencentCloudCOSStorageConfig(BaseSettings):
         description="Protocol scheme for COS requests: 'https' (recommended) or 'http'",
         default=None,
     )
+
+    TENCENT_COS_CUSTOM_DOMAIN: str | None = Field(
+        description="Tencent Cloud COS custom domain setting",
+        default=None,
+    )

api/controllers/console/app/app.py

@@ -1,3 +1,4 @@
+import re
 import uuid
 from typing import Literal
 
@@ -73,6 +74,48 @@ class AppListQuery(BaseModel):
             raise ValueError("Invalid UUID format in tag_ids.") from exc
 
 
+# XSS prevention: patterns that could lead to XSS attacks
+# Includes: script tags, iframe tags, javascript: protocol, SVG with onload, etc.
+_XSS_PATTERNS = [
+    r"<script[^>]*>.*?</script>",  # Script tags
+    r"<iframe\b[^>]*?(?:/>|>.*?</iframe>)",  # Iframe tags (including self-closing)
+    r"javascript:",  # JavaScript protocol
+    r"<svg[^>]*?\s+onload\s*=[^>]*>",  # SVG with onload handler (attribute-aware, flexible whitespace)
+    r"<.*?on\s*\w+\s*=",  # Event handlers like onclick, onerror, etc.
+    r"<object\b[^>]*(?:\s*/>|>.*?</object\s*>)",  # Object tags (opening tag)
+    r"<embed[^>]*>",  # Embed tags (self-closing)
+    r"<link[^>]*>",  # Link tags with javascript
+]
+
+
+def _validate_xss_safe(value: str | None, field_name: str = "Field") -> str | None:
+    """
+    Validate that a string value doesn't contain potential XSS payloads.
+
+    Args:
+        value: The string value to validate
+        field_name: Name of the field for error messages
+
+    Returns:
+        The original value if safe
+
+    Raises:
+        ValueError: If the value contains XSS patterns
+    """
+    if value is None:
+        return None
+
+    value_lower = value.lower()
+    for pattern in _XSS_PATTERNS:
+        if re.search(pattern, value_lower, re.DOTALL | re.IGNORECASE):
+            raise ValueError(
+                f"{field_name} contains invalid characters or patterns. "
+                "HTML tags, JavaScript, and other potentially dangerous content are not allowed."
+            )
+
+    return value
+
+
 class CreateAppPayload(BaseModel):
     name: str = Field(..., min_length=1, description="App name")
     description: str | None = Field(default=None, description="App description (max 400 chars)", max_length=400)
@@ -81,6 +124,11 @@ class CreateAppPayload(BaseModel):
     icon: str | None = Field(default=None, description="Icon")
     icon_background: str | None = Field(default=None, description="Icon background color")
 
+    @field_validator("name", "description", mode="before")
+    @classmethod
+    def validate_xss_safe(cls, value: str | None, info) -> str | None:
+        return _validate_xss_safe(value, info.field_name)
+
 
 class UpdateAppPayload(BaseModel):
     name: str = Field(..., min_length=1, description="App name")
@@ -91,6 +139,11 @@ class UpdateAppPayload(BaseModel):
     use_icon_as_answer_icon: bool | None = Field(default=None, description="Use icon as answer icon")
     max_active_requests: int | None = Field(default=None, description="Maximum active requests")
 
+    @field_validator("name", "description", mode="before")
+    @classmethod
+    def validate_xss_safe(cls, value: str | None, info) -> str | None:
+        return _validate_xss_safe(value, info.field_name)
+
 
 class CopyAppPayload(BaseModel):
     name: str | None = Field(default=None, description="Name for the copied app")
@@ -99,6 +152,11 @@ class CopyAppPayload(BaseModel):
     icon: str | None = Field(default=None, description="Icon")
     icon_background: str | None = Field(default=None, description="Icon background color")
 
+    @field_validator("name", "description", mode="before")
+    @classmethod
+    def validate_xss_safe(cls, value: str | None, info) -> str | None:
+        return _validate_xss_safe(value, info.field_name)
+
 
 class AppExportQuery(BaseModel):
     include_secret: bool = Field(default=False, description="Include secrets in export")

api/controllers/console/auth/oauth.py

@@ -124,7 +124,7 @@ class OAuthCallback(Resource):
             return redirect(f"{dify_config.CONSOLE_WEB_URL}/signin/invite-settings?invite_token={invite_token}")
 
         try:
-            account = _generate_account(provider, user_info)
+            account, oauth_new_user = _generate_account(provider, user_info)
         except AccountNotFoundError:
             return redirect(f"{dify_config.CONSOLE_WEB_URL}/signin?message=Account not found.")
         except (WorkSpaceNotFoundError, WorkSpaceNotAllowedCreateError):
@@ -159,7 +159,10 @@ class OAuthCallback(Resource):
             ip_address=extract_remote_ip(request),
         )
 
-        response = redirect(f"{dify_config.CONSOLE_WEB_URL}")
+        base_url = dify_config.CONSOLE_WEB_URL
+        query_char = "&" if "?" in base_url else "?"
+        target_url = f"{base_url}{query_char}oauth_new_user={str(oauth_new_user).lower()}"
+        response = redirect(target_url)
 
         set_access_token_to_cookie(request, response, token_pair.access_token)
         set_refresh_token_to_cookie(request, response, token_pair.refresh_token)
@@ -177,9 +180,10 @@ def _get_account_by_openid_or_email(provider: str, user_info: OAuthUserInfo) ->
     return account
 
 
-def _generate_account(provider: str, user_info: OAuthUserInfo):
+def _generate_account(provider: str, user_info: OAuthUserInfo) -> tuple[Account, bool]:
     # Get account by openid or email.
     account = _get_account_by_openid_or_email(provider, user_info)
+    oauth_new_user = False
 
     if account:
         tenants = TenantService.get_join_tenants(account)
@@ -193,6 +197,7 @@ def _generate_account(provider: str, user_info: OAuthUserInfo):
                 tenant_was_created.send(new_tenant)
 
     if not account:
+        oauth_new_user = True
         if not FeatureService.get_system_features().is_allow_register:
             if dify_config.BILLING_ENABLED and BillingService.is_email_in_freeze(user_info.email):
                 raise AccountRegisterError(
@@ -220,4 +225,4 @@ def _generate_account(provider: str, user_info: OAuthUserInfo):
     # Link account
     AccountService.link_account_integrate(provider, user_info.id, account)
 
-    return account
+    return account, oauth_new_user

api/controllers/console/billing/billing.py

@@ -1,8 +1,9 @@
 import base64
+from typing import Literal
 
 from flask import request
 from flask_restx import Resource, fields
-from pydantic import BaseModel, Field, field_validator
+from pydantic import BaseModel, Field
 from werkzeug.exceptions import BadRequest
 
 from controllers.console import console_ns
@@ -15,22 +16,8 @@ DEFAULT_REF_TEMPLATE_SWAGGER_2_0 = "#/definitions/{model}"
 
 
 class SubscriptionQuery(BaseModel):
-    plan: str = Field(..., description="Subscription plan")
-    interval: str = Field(..., description="Billing interval")
-
-    @field_validator("plan")
-    @classmethod
-    def validate_plan(cls, value: str) -> str:
-        if value not in [CloudPlan.PROFESSIONAL, CloudPlan.TEAM]:
-            raise ValueError("Invalid plan")
-        return value
-
-    @field_validator("interval")
-    @classmethod
-    def validate_interval(cls, value: str) -> str:
-        if value not in {"month", "year"}:
-            raise ValueError("Invalid interval")
-        return value
+    plan: Literal[CloudPlan.PROFESSIONAL, CloudPlan.TEAM] = Field(..., description="Subscription plan")
+    interval: Literal["month", "year"] = Field(..., description="Billing interval")
 
 
 class PartnerTenantsPayload(BaseModel):

api/controllers/console/datasets/datasets_segments.py

@@ -3,10 +3,12 @@ import uuid
 from flask import request
 from flask_restx import Resource, marshal
 from pydantic import BaseModel, Field
-from sqlalchemy import select
+from sqlalchemy import String, cast, func, or_, select
+from sqlalchemy.dialects.postgresql import JSONB
 from werkzeug.exceptions import Forbidden, NotFound
 
 import services
+from configs import dify_config
 from controllers.common.schema import register_schema_models
 from controllers.console import console_ns
 from controllers.console.app.error import ProviderNotInitializeError
@@ -143,7 +145,29 @@ class DatasetDocumentSegmentListApi(Resource):
             query = query.where(DocumentSegment.hit_count >= hit_count_gte)
 
         if keyword:
-            query = query.where(DocumentSegment.content.ilike(f"%{keyword}%"))
+            # Search in both content and keywords fields
+            # Use database-specific methods for JSON array search
+            if dify_config.SQLALCHEMY_DATABASE_URI_SCHEME == "postgresql":
+                # PostgreSQL: Use jsonb_array_elements_text to properly handle Unicode/Chinese text
+                keywords_condition = func.array_to_string(
+                    func.array(
+                        select(func.jsonb_array_elements_text(cast(DocumentSegment.keywords, JSONB)))
+                        .correlate(DocumentSegment)
+                        .scalar_subquery()
+                    ),
+                    ",",
+                ).ilike(f"%{keyword}%")
+            else:
+                # MySQL: Cast JSON to string for pattern matching
+                # MySQL stores Chinese text directly in JSON without Unicode escaping
+                keywords_condition = cast(DocumentSegment.keywords, String).ilike(f"%{keyword}%")
+
+            query = query.where(
+                or_(
+                    DocumentSegment.content.ilike(f"%{keyword}%"),
+                    keywords_condition,
+                )
+            )
 
         if args.enabled.lower() != "all":
             if args.enabled.lower() == "true":

api/controllers/console/explore/message.py

@@ -1,6 +1,5 @@
 import logging
 from typing import Literal
-from uuid import UUID
 
 from flask import request
 from flask_restx import marshal_with
@@ -26,6 +25,7 @@ from core.errors.error import ModelCurrentlyNotSupportError, ProviderTokenNotIni
 from core.model_runtime.errors.invoke import InvokeError
 from fields.message_fields import message_infinite_scroll_pagination_fields
 from libs import helper
+from libs.helper import UUIDStrOrEmpty
 from libs.login import current_account_with_tenant
 from models.model import AppMode
 from services.app_generate_service import AppGenerateService
@@ -44,8 +44,8 @@ logger = logging.getLogger(__name__)
 
 
 class MessageListQuery(BaseModel):
-    conversation_id: UUID
-    first_id: UUID | None = None
+    conversation_id: UUIDStrOrEmpty
+    first_id: UUIDStrOrEmpty | None = None
     limit: int = Field(default=20, ge=1, le=100)
 
 

api/controllers/console/explore/saved_message.py

@@ -1,5 +1,3 @@
-from uuid import UUID
-
 from flask import request
 from flask_restx import fields, marshal_with
 from pydantic import BaseModel, Field
@@ -10,19 +8,19 @@ from controllers.console import console_ns
 from controllers.console.explore.error import NotCompletionAppError
 from controllers.console.explore.wraps import InstalledAppResource
 from fields.conversation_fields import message_file_fields
-from libs.helper import TimestampField
+from libs.helper import TimestampField, UUIDStrOrEmpty
 from libs.login import current_account_with_tenant
 from services.errors.message import MessageNotExistsError
 from services.saved_message_service import SavedMessageService
 
 
 class SavedMessageListQuery(BaseModel):
-    last_id: UUID | None = None
+    last_id: UUIDStrOrEmpty | None = None
     limit: int = Field(default=20, ge=1, le=100)
 
 
 class SavedMessageCreatePayload(BaseModel):
-    message_id: UUID
+    message_id: UUIDStrOrEmpty
 
 
 register_schema_models(console_ns, SavedMessageListQuery, SavedMessageCreatePayload)

api/controllers/console/workspace/load_balancing_config.py

@@ -1,6 +1,8 @@
-from flask_restx import Resource, reqparse
+from flask_restx import Resource
+from pydantic import BaseModel
 from werkzeug.exceptions import Forbidden
 
+from controllers.common.schema import register_schema_models
 from controllers.console import console_ns
 from controllers.console.wraps import account_initialization_required, setup_required
 from core.model_runtime.entities.model_entities import ModelType
@@ -10,10 +12,20 @@ from models import TenantAccountRole
 from services.model_load_balancing_service import ModelLoadBalancingService
 
 
+class LoadBalancingCredentialPayload(BaseModel):
+    model: str
+    model_type: ModelType
+    credentials: dict[str, object]
+
+
+register_schema_models(console_ns, LoadBalancingCredentialPayload)
+
+
 @console_ns.route(
     "/workspaces/current/model-providers/<path:provider>/models/load-balancing-configs/credentials-validate"
 )
 class LoadBalancingCredentialsValidateApi(Resource):
+    @console_ns.expect(console_ns.models[LoadBalancingCredentialPayload.__name__])
     @setup_required
     @login_required
     @account_initialization_required
@@ -24,20 +36,7 @@ class LoadBalancingCredentialsValidateApi(Resource):
 
         tenant_id = current_tenant_id
 
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("model", type=str, required=True, nullable=False, location="json")
-            .add_argument(
-                "model_type",
-                type=str,
-                required=True,
-                nullable=False,
-                choices=[mt.value for mt in ModelType],
-                location="json",
-            )
-            .add_argument("credentials", type=dict, required=True, nullable=False, location="json")
-        )
-        args = parser.parse_args()
+        payload = LoadBalancingCredentialPayload.model_validate(console_ns.payload or {})
 
         # validate model load balancing credentials
         model_load_balancing_service = ModelLoadBalancingService()
@@ -49,9 +48,9 @@ class LoadBalancingCredentialsValidateApi(Resource):
             model_load_balancing_service.validate_load_balancing_credentials(
                 tenant_id=tenant_id,
                 provider=provider,
-                model=args["model"],
-                model_type=args["model_type"],
-                credentials=args["credentials"],
+                model=payload.model,
+                model_type=payload.model_type,
+                credentials=payload.credentials,
             )
         except CredentialsValidateFailedError as ex:
             result = False
@@ -69,6 +68,7 @@ class LoadBalancingCredentialsValidateApi(Resource):
     "/workspaces/current/model-providers/<path:provider>/models/load-balancing-configs/<string:config_id>/credentials-validate"
 )
 class LoadBalancingConfigCredentialsValidateApi(Resource):
+    @console_ns.expect(console_ns.models[LoadBalancingCredentialPayload.__name__])
     @setup_required
     @login_required
     @account_initialization_required
@@ -79,20 +79,7 @@ class LoadBalancingConfigCredentialsValidateApi(Resource):
 
         tenant_id = current_tenant_id
 
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("model", type=str, required=True, nullable=False, location="json")
-            .add_argument(
-                "model_type",
-                type=str,
-                required=True,
-                nullable=False,
-                choices=[mt.value for mt in ModelType],
-                location="json",
-            )
-            .add_argument("credentials", type=dict, required=True, nullable=False, location="json")
-        )
-        args = parser.parse_args()
+        payload = LoadBalancingCredentialPayload.model_validate(console_ns.payload or {})
 
         # validate model load balancing config credentials
         model_load_balancing_service = ModelLoadBalancingService()
@@ -104,9 +91,9 @@ class LoadBalancingConfigCredentialsValidateApi(Resource):
             model_load_balancing_service.validate_load_balancing_credentials(
                 tenant_id=tenant_id,
                 provider=provider,
-                model=args["model"],
-                model_type=args["model_type"],
-                credentials=args["credentials"],
+                model=payload.model,
+                model_type=payload.model_type,
+                credentials=payload.credentials,
                 config_id=config_id,
             )
         except CredentialsValidateFailedError as ex:

api/controllers/console/workspace/tool_providers.py

@@ -1,4 +1,5 @@
 import io
+import logging
 from urllib.parse import urlparse
 
 from flask import make_response, redirect, request, send_file
@@ -17,8 +18,8 @@ from controllers.console.wraps import (
     is_admin_or_owner_required,
     setup_required,
 )
+from core.db.session_factory import session_factory
 from core.entities.mcp_provider import MCPAuthentication, MCPConfiguration
-from core.helper.tool_provider_cache import ToolProviderListCache
 from core.mcp.auth.auth_flow import auth, handle_callback
 from core.mcp.error import MCPAuthError, MCPError, MCPRefreshTokenError
 from core.mcp.mcp_client import MCPClient
@@ -40,6 +41,8 @@ from services.tools.tools_manage_service import ToolCommonService
 from services.tools.tools_transform_service import ToolTransformService
 from services.tools.workflow_tools_manage_service import WorkflowToolManageService
 
+logger = logging.getLogger(__name__)
+
 
 def is_valid_url(url: str) -> bool:
     if not url:
@@ -945,8 +948,8 @@ class ToolProviderMCPApi(Resource):
         configuration = MCPConfiguration.model_validate(args["configuration"])
         authentication = MCPAuthentication.model_validate(args["authentication"]) if args["authentication"] else None
 
-        # Create provider in transaction
-        with Session(db.engine) as session, session.begin():
+        # 1) Create provider in a short transaction (no network I/O inside)
+        with session_factory.create_session() as session, session.begin():
             service = MCPToolManageService(session=session)
             result = service.create_provider(
                 tenant_id=tenant_id,
@@ -962,8 +965,26 @@ class ToolProviderMCPApi(Resource):
                 authentication=authentication,
             )
 
-        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
-        ToolProviderListCache.invalidate_cache(tenant_id)
+        # 2) Try to fetch tools immediately after creation so they appear without a second save.
+        #    Perform network I/O outside any DB session to avoid holding locks.
+        try:
+            reconnect = MCPToolManageService.reconnect_with_url(
+                server_url=args["server_url"],
+                headers=args.get("headers") or {},
+                timeout=configuration.timeout,
+                sse_read_timeout=configuration.sse_read_timeout,
+            )
+            # Update just-created provider with authed/tools in a new short transaction
+            with session_factory.create_session() as session, session.begin():
+                service = MCPToolManageService(session=session)
+                db_provider = service.get_provider(provider_id=result.id, tenant_id=tenant_id)
+                db_provider.authed = reconnect.authed
+                db_provider.tools = reconnect.tools
+
+                result = ToolTransformService.mcp_provider_to_user_provider(db_provider, for_list=True)
+        except Exception:
+            # Best-effort: if initial fetch fails (e.g., auth required), return created provider as-is
+            logger.warning("Failed to fetch MCP tools after creation", exc_info=True)
 
         return jsonable_encoder(result)
 
@@ -1011,9 +1032,6 @@ class ToolProviderMCPApi(Resource):
                 validation_result=validation_result,
             )
 
-        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
-        ToolProviderListCache.invalidate_cache(current_tenant_id)
-
         return {"result": "success"}
 
     @console_ns.expect(parser_mcp_delete)
@@ -1028,9 +1046,6 @@ class ToolProviderMCPApi(Resource):
             service = MCPToolManageService(session=session)
             service.delete_provider(tenant_id=current_tenant_id, provider_id=args["provider_id"])
 
-        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
-        ToolProviderListCache.invalidate_cache(current_tenant_id)
-
         return {"result": "success"}
 
 
@@ -1081,8 +1096,6 @@ class ToolMCPAuthApi(Resource):
                         credentials=provider_entity.credentials,
                         authed=True,
                     )
-                # Invalidate cache after updating credentials
-                ToolProviderListCache.invalidate_cache(tenant_id)
                 return {"result": "success"}
         except MCPAuthError as e:
             try:
@@ -1096,22 +1109,16 @@ class ToolMCPAuthApi(Resource):
                 with Session(db.engine) as session, session.begin():
                     service = MCPToolManageService(session=session)
                     response = service.execute_auth_actions(auth_result)
-                    # Invalidate cache after auth actions may have updated provider state
-                    ToolProviderListCache.invalidate_cache(tenant_id)
                     return response
             except MCPRefreshTokenError as e:
                 with Session(db.engine) as session, session.begin():
                     service = MCPToolManageService(session=session)
                     service.clear_provider_credentials(provider_id=provider_id, tenant_id=tenant_id)
-                # Invalidate cache after clearing credentials
-                ToolProviderListCache.invalidate_cache(tenant_id)
                 raise ValueError(f"Failed to refresh token, please try to authorize again: {e}") from e
         except (MCPError, ValueError) as e:
             with Session(db.engine) as session, session.begin():
                 service = MCPToolManageService(session=session)
                 service.clear_provider_credentials(provider_id=provider_id, tenant_id=tenant_id)
-            # Invalidate cache after clearing credentials
-            ToolProviderListCache.invalidate_cache(tenant_id)
             raise ValueError(f"Failed to connect to MCP server: {e}") from e
 
 

api/controllers/service_api/dataset/dataset.py

@@ -13,7 +13,6 @@ from controllers.service_api.dataset.error import DatasetInUseError, DatasetName
 from controllers.service_api.wraps import (
     DatasetApiResource,
     cloud_edition_billing_rate_limit_check,
-    validate_dataset_token,
 )
 from core.model_runtime.entities.model_entities import ModelType
 from core.provider_manager import ProviderManager
@@ -460,9 +459,8 @@ class DatasetTagsApi(DatasetApiResource):
             401: "Unauthorized - invalid API token",
         }
     )
-    @validate_dataset_token
     @service_api_ns.marshal_with(build_dataset_tag_fields(service_api_ns))
-    def get(self, _, dataset_id):
+    def get(self, _):
         """Get all knowledge type tags."""
         assert isinstance(current_user, Account)
         cid = current_user.current_tenant_id
@@ -482,8 +480,7 @@ class DatasetTagsApi(DatasetApiResource):
         }
     )
     @service_api_ns.marshal_with(build_dataset_tag_fields(service_api_ns))
-    @validate_dataset_token
-    def post(self, _, dataset_id):
+    def post(self, _):
         """Add a knowledge type tag."""
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
@@ -506,8 +503,7 @@ class DatasetTagsApi(DatasetApiResource):
         }
     )
     @service_api_ns.marshal_with(build_dataset_tag_fields(service_api_ns))
-    @validate_dataset_token
-    def patch(self, _, dataset_id):
+    def patch(self, _):
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
             raise Forbidden()
@@ -533,9 +529,8 @@ class DatasetTagsApi(DatasetApiResource):
             403: "Forbidden - insufficient permissions",
         }
     )
-    @validate_dataset_token
     @edit_permission_required
-    def delete(self, _, dataset_id):
+    def delete(self, _):
         """Delete a knowledge type tag."""
         payload = TagDeletePayload.model_validate(service_api_ns.payload or {})
         TagService.delete_tag(payload.tag_id)
@@ -555,8 +550,7 @@ class DatasetTagBindingApi(DatasetApiResource):
             403: "Forbidden - insufficient permissions",
         }
     )
-    @validate_dataset_token
-    def post(self, _, dataset_id):
+    def post(self, _):
         # The role of the current user in the ta table must be admin, owner, editor, or dataset_operator
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
@@ -580,8 +574,7 @@ class DatasetTagUnbindingApi(DatasetApiResource):
             403: "Forbidden - insufficient permissions",
         }
     )
-    @validate_dataset_token
-    def post(self, _, dataset_id):
+    def post(self, _):
         # The role of the current user in the ta table must be admin, owner, editor, or dataset_operator
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
@@ -604,7 +597,6 @@ class DatasetTagsBindingStatusApi(DatasetApiResource):
             401: "Unauthorized - invalid API token",
         }
     )
-    @validate_dataset_token
     def get(self, _, *args, **kwargs):
         """Get all knowledge type tags."""
         dataset_id = kwargs.get("dataset_id")

api/core/app/apps/base_app_queue_manager.py

@@ -90,6 +90,7 @@ class AppQueueManager:
         """
         self._clear_task_belong_cache()
         self._q.put(None)
+        self._graph_runtime_state = None  # Release reference to allow GC to reclaim memory
 
     def _clear_task_belong_cache(self) -> None:
         """

api/core/helper/code_executor/jinja2/jinja2_transformer.py

@@ -1,9 +1,14 @@
+from collections.abc import Mapping
 from textwrap import dedent
+from typing import Any
 
 from core.helper.code_executor.template_transformer import TemplateTransformer
 
 
 class Jinja2TemplateTransformer(TemplateTransformer):
+    # Use separate placeholder for base64-encoded template to avoid confusion
+    _template_b64_placeholder: str = "{{template_b64}}"
+
     @classmethod
     def transform_response(cls, response: str):
         """
@@ -13,18 +18,35 @@ class Jinja2TemplateTransformer(TemplateTransformer):
         """
         return {"result": cls.extract_result_str_from_response(response)}
 
+    @classmethod
+    def assemble_runner_script(cls, code: str, inputs: Mapping[str, Any]) -> str:
+        """
+        Override base class to use base64 encoding for template code.
+        This prevents issues with special characters (quotes, newlines) in templates
+        breaking the generated Python script. Fixes #26818.
+        """
+        script = cls.get_runner_script()
+        # Encode template as base64 to safely embed any content including quotes
+        code_b64 = cls.serialize_code(code)
+        script = script.replace(cls._template_b64_placeholder, code_b64)
+        inputs_str = cls.serialize_inputs(inputs)
+        script = script.replace(cls._inputs_placeholder, inputs_str)
+        return script
+
     @classmethod
     def get_runner_script(cls) -> str:
         runner_script = dedent(f"""
-            # declare main function
-            def main(**inputs):
-                import jinja2
-                template = jinja2.Template('''{cls._code_placeholder}''')
-                return template.render(**inputs)
-
+            import jinja2
             import json
             from base64 import b64decode
 
+            # declare main function
+            def main(**inputs):
+                # Decode base64-encoded template to handle special characters safely
+                template_code = b64decode('{cls._template_b64_placeholder}').decode('utf-8')
+                template = jinja2.Template(template_code)
+                return template.render(**inputs)
+
             # decode and prepare input dict
             inputs_obj = json.loads(b64decode('{cls._inputs_placeholder}').decode('utf-8'))
 

api/core/helper/code_executor/template_transformer.py

@@ -13,6 +13,15 @@ class TemplateTransformer(ABC):
     _inputs_placeholder: str = "{{inputs}}"
     _result_tag: str = "<<RESULT>>"
 
+    @classmethod
+    def serialize_code(cls, code: str) -> str:
+        """
+        Serialize template code to base64 to safely embed in generated script.
+        This prevents issues with special characters like quotes breaking the script.
+        """
+        code_bytes = code.encode("utf-8")
+        return b64encode(code_bytes).decode("utf-8")
+
     @classmethod
     def transform_caller(cls, code: str, inputs: Mapping[str, Any]) -> tuple[str, str]:
         """

api/core/helper/tool_provider_cache.py

@@ -1,56 +0,0 @@
-import json
-import logging
-from typing import Any
-
-from core.tools.entities.api_entities import ToolProviderTypeApiLiteral
-from extensions.ext_redis import redis_client, redis_fallback
-
-logger = logging.getLogger(__name__)
-
-
-class ToolProviderListCache:
-    """Cache for tool provider lists"""
-
-    CACHE_TTL = 300  # 5 minutes
-
-    @staticmethod
-    def _generate_cache_key(tenant_id: str, typ: ToolProviderTypeApiLiteral = None) -> str:
-        """Generate cache key for tool providers list"""
-        type_filter = typ or "all"
-        return f"tool_providers:tenant_id:{tenant_id}:type:{type_filter}"
-
-    @staticmethod
-    @redis_fallback(default_return=None)
-    def get_cached_providers(tenant_id: str, typ: ToolProviderTypeApiLiteral = None) -> list[dict[str, Any]] | None:
-        """Get cached tool providers"""
-        cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-        cached_data = redis_client.get(cache_key)
-        if cached_data:
-            try:
-                return json.loads(cached_data.decode("utf-8"))
-            except (json.JSONDecodeError, UnicodeDecodeError):
-                logger.warning("Failed to decode cached tool providers data")
-                return None
-        return None
-
-    @staticmethod
-    @redis_fallback()
-    def set_cached_providers(tenant_id: str, typ: ToolProviderTypeApiLiteral, providers: list[dict[str, Any]]):
-        """Cache tool providers"""
-        cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-        redis_client.setex(cache_key, ToolProviderListCache.CACHE_TTL, json.dumps(providers))
-
-    @staticmethod
-    @redis_fallback()
-    def invalidate_cache(tenant_id: str, typ: ToolProviderTypeApiLiteral = None):
-        """Invalidate cache for tool providers"""
-        if typ:
-            # Invalidate specific type cache
-            cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-            redis_client.delete(cache_key)
-        else:
-            # Invalidate all caches for this tenant
-            pattern = f"tool_providers:tenant_id:{tenant_id}:*"
-            keys = list(redis_client.scan_iter(pattern))
-            if keys:
-                redis_client.delete(*keys)

api/core/mcp/client/streamable_client.py

@@ -313,17 +313,20 @@ class StreamableHTTPTransport:
             if is_initialization:
                 self._maybe_extract_session_id_from_response(response)
 
-            content_type = cast(str, response.headers.get(CONTENT_TYPE, "").lower())
+            # Per https://modelcontextprotocol.io/specification/2025-06-18/basic#notifications:
+            # The server MUST NOT send a response to notifications.
+            if isinstance(message.root, JSONRPCRequest):
+                content_type = cast(str, response.headers.get(CONTENT_TYPE, "").lower())
 
-            if content_type.startswith(JSON):
-                self._handle_json_response(response, ctx.server_to_client_queue)
-            elif content_type.startswith(SSE):
-                self._handle_sse_response(response, ctx)
-            else:
-                self._handle_unexpected_content_type(
-                    content_type,
-                    ctx.server_to_client_queue,
-                )
+                if content_type.startswith(JSON):
+                    self._handle_json_response(response, ctx.server_to_client_queue)
+                elif content_type.startswith(SSE):
+                    self._handle_sse_response(response, ctx)
+                else:
+                    self._handle_unexpected_content_type(
+                        content_type,
+                        ctx.server_to_client_queue,
+                    )
 
     def _handle_json_response(
         self,

api/core/plugin/entities/parameters.py

@@ -76,7 +76,7 @@ class PluginParameter(BaseModel):
     auto_generate: PluginParameterAutoGenerate | None = None
     template: PluginParameterTemplate | None = None
     required: bool = False
-    default: Union[float, int, str, bool] | None = None
+    default: Union[float, int, str, bool, list, dict] | None = None
     min: Union[float, int] | None = None
     max: Union[float, int] | None = None
     precision: int | None = None

api/core/rag/cleaner/clean_processor.py

@@ -27,26 +27,44 @@ class CleanProcessor:
                     pattern = r"([a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+)"
                     text = re.sub(pattern, "", text)
 
-                    # Remove URL but keep Markdown image URLs
-                    # First, temporarily replace Markdown image URLs with a placeholder
-                    markdown_image_pattern = r"!\[.*?\]\((https?://[^\s)]+)\)"
-                    placeholders: list[str] = []
+                    # Remove URL but keep Markdown image URLs and link URLs
+                    # Replace the ENTIRE markdown link/image with a single placeholder to protect
+                    # the link text (which might also be a URL) from being removed
+                    markdown_link_pattern = r"\[([^\]]*)\]\((https?://[^)]+)\)"
+                    markdown_image_pattern = r"!\[.*?\]\((https?://[^)]+)\)"
+                    placeholders: list[tuple[str, str, str]] = []  # (type, text, url)
 
-                    def replace_with_placeholder(match, placeholders=placeholders):
+                    def replace_markdown_with_placeholder(match, placeholders=placeholders):
+                        link_type = "link"
+                        link_text = match.group(1)
+                        url = match.group(2)
+                        placeholder = f"__MARKDOWN_PLACEHOLDER_{len(placeholders)}__"
+                        placeholders.append((link_type, link_text, url))
+                        return placeholder
+
+                    def replace_image_with_placeholder(match, placeholders=placeholders):
+                        link_type = "image"
                         url = match.group(1)
-                        placeholder = f"__MARKDOWN_IMAGE_URL_{len(placeholders)}__"
-                        placeholders.append(url)
-                        return f"![image]({placeholder})"
+                        placeholder = f"__MARKDOWN_PLACEHOLDER_{len(placeholders)}__"
+                        placeholders.append((link_type, "image", url))
+                        return placeholder
 
-                    text = re.sub(markdown_image_pattern, replace_with_placeholder, text)
+                    # Protect markdown links first
+                    text = re.sub(markdown_link_pattern, replace_markdown_with_placeholder, text)
+                    # Then protect markdown images
+                    text = re.sub(markdown_image_pattern, replace_image_with_placeholder, text)
 
                     # Now remove all remaining URLs
-                    url_pattern = r"https?://[^\s)]+"
+                    url_pattern = r"https?://\S+"
                     text = re.sub(url_pattern, "", text)
 
-                    # Finally, restore the Markdown image URLs
-                    for i, url in enumerate(placeholders):
-                        text = text.replace(f"__MARKDOWN_IMAGE_URL_{i}__", url)
+                    # Restore the Markdown links and images
+                    for i, (link_type, text_or_alt, url) in enumerate(placeholders):
+                        placeholder = f"__MARKDOWN_PLACEHOLDER_{i}__"
+                        if link_type == "link":
+                            text = text.replace(placeholder, f"[{text_or_alt}]({url})")
+                        else:  # image
+                            text = text.replace(placeholder, f"![{text_or_alt}]({url})")
         return text
 
     def filter_string(self, text):

api/core/rag/datasource/retrieval_service.py

@@ -1,4 +1,5 @@
 import concurrent.futures
+import logging
 from concurrent.futures import ThreadPoolExecutor
 from typing import Any
 
@@ -13,7 +14,7 @@ from core.model_runtime.entities.model_entities import ModelType
 from core.rag.data_post_processor.data_post_processor import DataPostProcessor
 from core.rag.datasource.keyword.keyword_factory import Keyword
 from core.rag.datasource.vdb.vector_factory import Vector
-from core.rag.embedding.retrieval import RetrievalSegments
+from core.rag.embedding.retrieval import RetrievalChildChunk, RetrievalSegments
 from core.rag.entities.metadata_entities import MetadataCondition
 from core.rag.index_processor.constant.doc_type import DocType
 from core.rag.index_processor.constant.index_type import IndexStructureType
@@ -36,6 +37,8 @@ default_retrieval_model = {
     "score_threshold_enabled": False,
 }
 
+logger = logging.getLogger(__name__)
+
 
 class RetrievalService:
     # Cache precompiled regular expressions to avoid repeated compilation
@@ -106,7 +109,12 @@ class RetrievalService:
                         )
                     )
 
-            concurrent.futures.wait(futures, timeout=3600, return_when=concurrent.futures.ALL_COMPLETED)
+            if futures:
+                for future in concurrent.futures.as_completed(futures, timeout=3600):
+                    if exceptions:
+                        for f in futures:
+                            f.cancel()
+                        break
 
         if exceptions:
             raise ValueError(";\n".join(exceptions))
@@ -210,6 +218,7 @@ class RetrievalService:
                 )
                 all_documents.extend(documents)
             except Exception as e:
+                logger.error(e, exc_info=True)
                 exceptions.append(str(e))
 
     @classmethod
@@ -303,6 +312,7 @@ class RetrievalService:
                     else:
                         all_documents.extend(documents)
             except Exception as e:
+                logger.error(e, exc_info=True)
                 exceptions.append(str(e))
 
     @classmethod
@@ -351,6 +361,7 @@ class RetrievalService:
                     else:
                         all_documents.extend(documents)
             except Exception as e:
+                logger.error(e, exc_info=True)
                 exceptions.append(str(e))
 
     @staticmethod
@@ -381,10 +392,9 @@ class RetrievalService:
             records = []
             include_segment_ids = set()
             segment_child_map = {}
-            segment_file_map = {}
 
             valid_dataset_documents = {}
-            image_doc_ids = []
+            image_doc_ids: list[Any] = []
             child_index_node_ids = []
             index_node_ids = []
             doc_to_document_map = {}
@@ -417,28 +427,39 @@ class RetrievalService:
             child_index_node_ids = [i for i in child_index_node_ids if i]
             index_node_ids = [i for i in index_node_ids if i]
 
-            segment_ids = []
+            segment_ids: list[str] = []
             index_node_segments: list[DocumentSegment] = []
             segments: list[DocumentSegment] = []
-            attachment_map = {}
-            child_chunk_map = {}
-            doc_segment_map = {}
+            attachment_map: dict[str, list[dict[str, Any]]] = {}
+            child_chunk_map: dict[str, list[ChildChunk]] = {}
+            doc_segment_map: dict[str, list[str]] = {}
 
             with session_factory.create_session() as session:
                 attachments = cls.get_segment_attachment_infos(image_doc_ids, session)
 
                 for attachment in attachments:
                     segment_ids.append(attachment["segment_id"])
-                    attachment_map[attachment["segment_id"]] = attachment
-                    doc_segment_map[attachment["segment_id"]] = attachment["attachment_id"]
-
+                    if attachment["segment_id"] in attachment_map:
+                        attachment_map[attachment["segment_id"]].append(attachment["attachment_info"])
+                    else:
+                        attachment_map[attachment["segment_id"]] = [attachment["attachment_info"]]
+                    if attachment["segment_id"] in doc_segment_map:
+                        doc_segment_map[attachment["segment_id"]].append(attachment["attachment_id"])
+                    else:
+                        doc_segment_map[attachment["segment_id"]] = [attachment["attachment_id"]]
                 child_chunk_stmt = select(ChildChunk).where(ChildChunk.index_node_id.in_(child_index_node_ids))
                 child_index_nodes = session.execute(child_chunk_stmt).scalars().all()
 
                 for i in child_index_nodes:
                     segment_ids.append(i.segment_id)
-                    child_chunk_map[i.segment_id] = i
-                    doc_segment_map[i.segment_id] = i.index_node_id
+                    if i.segment_id in child_chunk_map:
+                        child_chunk_map[i.segment_id].append(i)
+                    else:
+                        child_chunk_map[i.segment_id] = [i]
+                    if i.segment_id in doc_segment_map:
+                        doc_segment_map[i.segment_id].append(i.index_node_id)
+                    else:
+                        doc_segment_map[i.segment_id] = [i.index_node_id]
 
                 if index_node_ids:
                     document_segment_stmt = select(DocumentSegment).where(
@@ -448,7 +469,7 @@ class RetrievalService:
                     )
                     index_node_segments = session.execute(document_segment_stmt).scalars().all()  # type: ignore
                     for index_node_segment in index_node_segments:
-                        doc_segment_map[index_node_segment.id] = index_node_segment.index_node_id
+                        doc_segment_map[index_node_segment.id] = [index_node_segment.index_node_id]
                 if segment_ids:
                     document_segment_stmt = select(DocumentSegment).where(
                         DocumentSegment.enabled == True,
@@ -461,95 +482,86 @@ class RetrievalService:
                     segments.extend(index_node_segments)
 
             for segment in segments:
-                doc_id = doc_segment_map.get(segment.id)
-                child_chunk = child_chunk_map.get(segment.id)
-                attachment_info = attachment_map.get(segment.id)
+                child_chunks: list[ChildChunk] = child_chunk_map.get(segment.id, [])
+                attachment_infos: list[dict[str, Any]] = attachment_map.get(segment.id, [])
+                ds_dataset_document: DatasetDocument | None = valid_dataset_documents.get(segment.document_id)
 
-                if doc_id:
-                    document = doc_to_document_map[doc_id]
-                    ds_dataset_document: DatasetDocument | None = valid_dataset_documents.get(
-                        document.metadata.get("document_id")
-                    )
-
-                    if ds_dataset_document and ds_dataset_document.doc_form == IndexStructureType.PARENT_CHILD_INDEX:
-                        if segment.id not in include_segment_ids:
-                            include_segment_ids.add(segment.id)
-                            if child_chunk:
+                if ds_dataset_document and ds_dataset_document.doc_form == IndexStructureType.PARENT_CHILD_INDEX:
+                    if segment.id not in include_segment_ids:
+                        include_segment_ids.add(segment.id)
+                        if child_chunks or attachment_infos:
+                            child_chunk_details = []
+                            max_score = 0.0
+                            for child_chunk in child_chunks:
+                                document = doc_to_document_map[child_chunk.index_node_id]
                                 child_chunk_detail = {
                                     "id": child_chunk.id,
                                     "content": child_chunk.content,
                                     "position": child_chunk.position,
                                     "score": document.metadata.get("score", 0.0) if document else 0.0,
                                 }
-                                map_detail = {
-                                    "max_score": document.metadata.get("score", 0.0) if document else 0.0,
-                                    "child_chunks": [child_chunk_detail],
-                                }
-                                segment_child_map[segment.id] = map_detail
-                            record = {
-                                "segment": segment,
+                                child_chunk_details.append(child_chunk_detail)
+                                max_score = max(max_score, document.metadata.get("score", 0.0) if document else 0.0)
+                            for attachment_info in attachment_infos:
+                                file_document = doc_to_document_map[attachment_info["id"]]
+                                max_score = max(
+                                    max_score, file_document.metadata.get("score", 0.0) if file_document else 0.0
+                                )
+
+                            map_detail = {
+                                "max_score": max_score,
+                                "child_chunks": child_chunk_details,
                             }
-                            if attachment_info:
-                                segment_file_map[segment.id] = [attachment_info]
-                            records.append(record)
-                        else:
-                            if child_chunk:
-                                child_chunk_detail = {
-                                    "id": child_chunk.id,
-                                    "content": child_chunk.content,
-                                    "position": child_chunk.position,
-                                    "score": document.metadata.get("score", 0.0),
-                                }
-                                if segment.id in segment_child_map:
-                                    segment_child_map[segment.id]["child_chunks"].append(child_chunk_detail)  # type: ignore
-                                    segment_child_map[segment.id]["max_score"] = max(
-                                        segment_child_map[segment.id]["max_score"],
-                                        document.metadata.get("score", 0.0) if document else 0.0,
-                                    )
-                                else:
-                                    segment_child_map[segment.id] = {
-                                        "max_score": document.metadata.get("score", 0.0) if document else 0.0,
-                                        "child_chunks": [child_chunk_detail],
-                                    }
-                            if attachment_info:
-                                if segment.id in segment_file_map:
-                                    segment_file_map[segment.id].append(attachment_info)
-                                else:
-                                    segment_file_map[segment.id] = [attachment_info]
-                    else:
-                        if segment.id not in include_segment_ids:
-                            include_segment_ids.add(segment.id)
-                            record = {
-                                "segment": segment,
-                                "score": document.metadata.get("score", 0.0),  # type: ignore
-                            }
-                            if attachment_info:
-                                segment_file_map[segment.id] = [attachment_info]
-                            records.append(record)
-                        else:
-                            if attachment_info:
-                                attachment_infos = segment_file_map.get(segment.id, [])
-                                if attachment_info not in attachment_infos:
-                                    attachment_infos.append(attachment_info)
-                                segment_file_map[segment.id] = attachment_infos
+                            segment_child_map[segment.id] = map_detail
+                        record: dict[str, Any] = {
+                            "segment": segment,
+                        }
+                        records.append(record)
+                else:
+                    if segment.id not in include_segment_ids:
+                        include_segment_ids.add(segment.id)
+                        max_score = 0.0
+                        segment_document = doc_to_document_map.get(segment.index_node_id)
+                        if segment_document:
+                            max_score = max(max_score, segment_document.metadata.get("score", 0.0))
+                        for attachment_info in attachment_infos:
+                            file_doc = doc_to_document_map.get(attachment_info["id"])
+                            if file_doc:
+                                max_score = max(max_score, file_doc.metadata.get("score", 0.0))
+                        record = {
+                            "segment": segment,
+                            "score": max_score,
+                        }
+                        records.append(record)
 
             # Add child chunks information to records
             for record in records:
                 if record["segment"].id in segment_child_map:
                     record["child_chunks"] = segment_child_map[record["segment"].id].get("child_chunks")  # type: ignore
                     record["score"] = segment_child_map[record["segment"].id]["max_score"]  # type: ignore
-                if record["segment"].id in segment_file_map:
-                    record["files"] = segment_file_map[record["segment"].id]  # type: ignore[assignment]
+                if record["segment"].id in attachment_map:
+                    record["files"] = attachment_map[record["segment"].id]  # type: ignore[assignment]
 
-            result = []
+            result: list[RetrievalSegments] = []
             for record in records:
                 # Extract segment
                 segment = record["segment"]
 
                 # Extract child_chunks, ensuring it's a list or None
-                child_chunks = record.get("child_chunks")
-                if not isinstance(child_chunks, list):
-                    child_chunks = None
+                raw_child_chunks = record.get("child_chunks")
+                child_chunks_list: list[RetrievalChildChunk] | None = None
+                if isinstance(raw_child_chunks, list):
+                    # Sort by score descending
+                    sorted_chunks = sorted(raw_child_chunks, key=lambda x: x.get("score", 0.0), reverse=True)
+                    child_chunks_list = [
+                        RetrievalChildChunk(
+                            id=chunk["id"],
+                            content=chunk["content"],
+                            score=chunk.get("score", 0.0),
+                            position=chunk["position"],
+                        )
+                        for chunk in sorted_chunks
+                    ]
 
                 # Extract files, ensuring it's a list or None
                 files = record.get("files")
@@ -566,11 +578,11 @@ class RetrievalService:
 
                 # Create RetrievalSegments object
                 retrieval_segment = RetrievalSegments(
-                    segment=segment, child_chunks=child_chunks, score=score, files=files
+                    segment=segment, child_chunks=child_chunks_list, score=score, files=files
                 )
                 result.append(retrieval_segment)
 
-            return result
+            return sorted(result, key=lambda x: x.score if x.score is not None else 0.0, reverse=True)
         except Exception as e:
             db.session.rollback()
             raise e
@@ -662,7 +674,14 @@ class RetrievalService:
                             document_ids_filter=document_ids_filter,
                         )
                     )
-                concurrent.futures.wait(futures, timeout=300, return_when=concurrent.futures.ALL_COMPLETED)
+                # Use as_completed for early error propagation - cancel remaining futures on first error
+                if futures:
+                    for future in concurrent.futures.as_completed(futures, timeout=300):
+                        if future.exception():
+                            # Cancel remaining futures to avoid unnecessary waiting
+                            for f in futures:
+                                f.cancel()
+                            break
 
             if exceptions:
                 raise ValueError(";\n".join(exceptions))

api/core/rag/datasource/vdb/pgvector/pgvector.py

@@ -255,7 +255,10 @@ class PGVector(BaseVector):
                 return
 
             with self._get_cursor() as cur:
-                cur.execute("CREATE EXTENSION IF NOT EXISTS vector")
+                cur.execute("SELECT 1 FROM pg_extension WHERE extname = 'vector'")
+                if not cur.fetchone():
+                    cur.execute("CREATE EXTENSION vector")
+
                 cur.execute(SQL_CREATE_TABLE.format(table_name=self.table_name, dimension=dimension))
                 # PG hnsw index only support 2000 dimension or less
                 # ref: https://github.com/pgvector/pgvector?tab=readme-ov-file#indexing

api/core/rag/extractor/extract_processor.py

@@ -112,7 +112,7 @@ class ExtractProcessor:
                     if file_extension in {".xlsx", ".xls"}:
                         extractor = ExcelExtractor(file_path)
                     elif file_extension == ".pdf":
-                        extractor = PdfExtractor(file_path)
+                        extractor = PdfExtractor(file_path, upload_file.tenant_id, upload_file.created_by)
                     elif file_extension in {".md", ".markdown", ".mdx"}:
                         extractor = (
                             UnstructuredMarkdownExtractor(file_path, unstructured_api_url, unstructured_api_key)
@@ -148,7 +148,7 @@ class ExtractProcessor:
                     if file_extension in {".xlsx", ".xls"}:
                         extractor = ExcelExtractor(file_path)
                     elif file_extension == ".pdf":
-                        extractor = PdfExtractor(file_path)
+                        extractor = PdfExtractor(file_path, upload_file.tenant_id, upload_file.created_by)
                     elif file_extension in {".md", ".markdown", ".mdx"}:
                         extractor = MarkdownExtractor(file_path, autodetect_encoding=True)
                     elif file_extension in {".htm", ".html"}:

api/core/rag/extractor/pdf_extractor.py

@@ -1,25 +1,57 @@
 """Abstract interface for document loader implementations."""
 
 import contextlib
+import io
+import logging
+import uuid
 from collections.abc import Iterator
 
+import pypdfium2
+import pypdfium2.raw as pdfium_c
+
+from configs import dify_config
 from core.rag.extractor.blob.blob import Blob
 from core.rag.extractor.extractor_base import BaseExtractor
 from core.rag.models.document import Document
+from extensions.ext_database import db
 from extensions.ext_storage import storage
+from libs.datetime_utils import naive_utc_now
+from models.enums import CreatorUserRole
+from models.model import UploadFile
+
+logger = logging.getLogger(__name__)
 
 
 class PdfExtractor(BaseExtractor):
-    """Load pdf files.
-
+    """
+    PdfExtractor is used to extract text and images from PDF files.
 
     Args:
-        file_path: Path to the file to load.
+        file_path: Path to the PDF file.
+        tenant_id: Workspace ID.
+        user_id: ID of the user performing the extraction.
+        file_cache_key: Optional cache key for the extracted text.
     """
 
-    def __init__(self, file_path: str, file_cache_key: str | None = None):
-        """Initialize with file path."""
+    # Magic bytes for image format detection: (magic_bytes, extension, mime_type)
+    IMAGE_FORMATS = [
+        (b"\xff\xd8\xff", "jpg", "image/jpeg"),
+        (b"\x89PNG\r\n\x1a\n", "png", "image/png"),
+        (b"\x00\x00\x00\x0c\x6a\x50\x20\x20\x0d\x0a\x87\x0a", "jp2", "image/jp2"),
+        (b"GIF8", "gif", "image/gif"),
+        (b"BM", "bmp", "image/bmp"),
+        (b"II*\x00", "tiff", "image/tiff"),
+        (b"MM\x00*", "tiff", "image/tiff"),
+        (b"II+\x00", "tiff", "image/tiff"),
+        (b"MM\x00+", "tiff", "image/tiff"),
+    ]
+    MAX_MAGIC_LEN = max(len(m) for m, _, _ in IMAGE_FORMATS)
+
+    def __init__(self, file_path: str, tenant_id: str, user_id: str, file_cache_key: str | None = None):
+        """Initialize PdfExtractor."""
         self._file_path = file_path
+        self._tenant_id = tenant_id
+        self._user_id = user_id
         self._file_cache_key = file_cache_key
 
     def extract(self) -> list[Document]:
@@ -50,7 +82,6 @@ class PdfExtractor(BaseExtractor):
 
     def parse(self, blob: Blob) -> Iterator[Document]:
         """Lazily parse the blob."""
-        import pypdfium2  # type: ignore
 
         with blob.as_bytes_io() as file_path:
             pdf_reader = pypdfium2.PdfDocument(file_path, autoclose=True)
@@ -59,8 +90,87 @@ class PdfExtractor(BaseExtractor):
                     text_page = page.get_textpage()
                     content = text_page.get_text_range()
                     text_page.close()
+
+                    image_content = self._extract_images(page)
+                    if image_content:
+                        content += "\n" + image_content
+
                     page.close()
                     metadata = {"source": blob.source, "page": page_number}
                     yield Document(page_content=content, metadata=metadata)
             finally:
                 pdf_reader.close()
+
+    def _extract_images(self, page) -> str:
+        """
+        Extract images from a PDF page, save them to storage and database,
+        and return markdown image links.
+
+        Args:
+            page: pypdfium2 page object.
+
+        Returns:
+            Markdown string containing links to the extracted images.
+        """
+        image_content = []
+        upload_files = []
+        base_url = dify_config.INTERNAL_FILES_URL or dify_config.FILES_URL
+
+        try:
+            image_objects = page.get_objects(filter=(pdfium_c.FPDF_PAGEOBJ_IMAGE,))
+            for obj in image_objects:
+                try:
+                    # Extract image bytes
+                    img_byte_arr = io.BytesIO()
+                    # Extract DCTDecode (JPEG) and JPXDecode (JPEG 2000) images directly
+                    # Fallback to png for other formats
+                    obj.extract(img_byte_arr, fb_format="png")
+                    img_bytes = img_byte_arr.getvalue()
+
+                    if not img_bytes:
+                        continue
+
+                    header = img_bytes[: self.MAX_MAGIC_LEN]
+                    image_ext = None
+                    mime_type = None
+                    for magic, ext, mime in self.IMAGE_FORMATS:
+                        if header.startswith(magic):
+                            image_ext = ext
+                            mime_type = mime
+                            break
+
+                    if not image_ext or not mime_type:
+                        continue
+
+                    file_uuid = str(uuid.uuid4())
+                    file_key = "image_files/" + self._tenant_id + "/" + file_uuid + "." + image_ext
+
+                    storage.save(file_key, img_bytes)
+
+                    # save file to db
+                    upload_file = UploadFile(
+                        tenant_id=self._tenant_id,
+                        storage_type=dify_config.STORAGE_TYPE,
+                        key=file_key,
+                        name=file_key,
+                        size=len(img_bytes),
+                        extension=image_ext,
+                        mime_type=mime_type,
+                        created_by=self._user_id,
+                        created_by_role=CreatorUserRole.ACCOUNT,
+                        created_at=naive_utc_now(),
+                        used=True,
+                        used_by=self._user_id,
+                        used_at=naive_utc_now(),
+                    )
+                    upload_files.append(upload_file)
+                    image_content.append(f"![image]({base_url}/files/{upload_file.id}/file-preview)")
+                except Exception as e:
+                    logger.warning("Failed to extract image from PDF: %s", e)
+                    continue
+        except Exception as e:
+            logger.warning("Failed to get objects from PDF page: %s", e)
+        if upload_files:
+            db.session.add_all(upload_files)
+            db.session.commit()
+        return "\n".join(image_content)

api/core/rag/retrieval/dataset_retrieval.py

@@ -7,7 +7,7 @@ from collections.abc import Generator, Mapping
 from typing import Any, Union, cast
 
 from flask import Flask, current_app
-from sqlalchemy import and_, or_, select
+from sqlalchemy import and_, literal, or_, select
 from sqlalchemy.orm import Session
 
 from core.app.app_config.entities import (
@@ -516,6 +516,9 @@ class DatasetRetrieval:
                     ].embedding_model_provider
                     weights["vector_setting"]["embedding_model_name"] = available_datasets[0].embedding_model
         with measure_time() as timer:
+            cancel_event = threading.Event()
+            thread_exceptions: list[Exception] = []
+
             if query:
                 query_thread = threading.Thread(
                     target=self._multiple_retrieve_thread,
@@ -534,6 +537,8 @@ class DatasetRetrieval:
                         "score_threshold": score_threshold,
                         "query": query,
                         "attachment_id": None,
+                        "cancel_event": cancel_event,
+                        "thread_exceptions": thread_exceptions,
                     },
                 )
                 all_threads.append(query_thread)
@@ -557,12 +562,25 @@ class DatasetRetrieval:
                             "score_threshold": score_threshold,
                             "query": None,
                             "attachment_id": attachment_id,
+                            "cancel_event": cancel_event,
+                            "thread_exceptions": thread_exceptions,
                         },
                     )
                     all_threads.append(attachment_thread)
                     attachment_thread.start()
-            for thread in all_threads:
-                thread.join()
+
+            # Poll threads with short timeout to detect errors quickly (fail-fast)
+            while any(t.is_alive() for t in all_threads):
+                for thread in all_threads:
+                    thread.join(timeout=0.1)
+                    if thread_exceptions:
+                        cancel_event.set()
+                        break
+                if thread_exceptions:
+                    break
+
+            if thread_exceptions:
+                raise thread_exceptions[0]
         self._on_query(query, attachment_ids, dataset_ids, app_id, user_from, user_id)
 
         if all_documents:
@@ -1036,7 +1054,7 @@ class DatasetRetrieval:
             if automatic_metadata_filters:
                 conditions = []
                 for sequence, filter in enumerate(automatic_metadata_filters):
-                    self._process_metadata_filter_func(
+                    self.process_metadata_filter_func(
                         sequence,
                         filter.get("condition"),  # type: ignore
                         filter.get("metadata_name"),  # type: ignore
@@ -1072,7 +1090,7 @@ class DatasetRetrieval:
                             value=expected_value,
                         )
                     )
-                    filters = self._process_metadata_filter_func(
+                    filters = self.process_metadata_filter_func(
                         sequence,
                         condition.comparison_operator,
                         metadata_name,
@@ -1168,8 +1186,9 @@ class DatasetRetrieval:
             return None
         return automatic_metadata_filters
 
-    def _process_metadata_filter_func(
-        self, sequence: int, condition: str, metadata_name: str, value: Any | None, filters: list
+    @classmethod
+    def process_metadata_filter_func(
+        cls, sequence: int, condition: str, metadata_name: str, value: Any | None, filters: list
     ):
         if value is None and condition not in ("empty", "not empty"):
             return filters
@@ -1218,6 +1237,20 @@ class DatasetRetrieval:
 
             case "≥" | ">=":
                 filters.append(DatasetDocument.doc_metadata[metadata_name].as_float() >= value)
+            case "in" | "not in":
+                if isinstance(value, str):
+                    value_list = [v.strip() for v in value.split(",") if v.strip()]
+                elif isinstance(value, (list, tuple)):
+                    value_list = [str(v) for v in value if v is not None]
+                else:
+                    value_list = [str(value)] if value is not None else []
+
+                if not value_list:
+                    # `field in []` is False, `field not in []` is True
+                    filters.append(literal(condition == "not in"))
+                else:
+                    op = json_field.in_ if condition == "in" else json_field.notin_
+                    filters.append(op(value_list))
             case _:
                 pass
 
@@ -1389,40 +1422,53 @@ class DatasetRetrieval:
         score_threshold: float,
         query: str | None,
         attachment_id: str | None,
+        cancel_event: threading.Event | None = None,
+        thread_exceptions: list[Exception] | None = None,
     ):
-        with flask_app.app_context():
-            threads = []
-            all_documents_item: list[Document] = []
-            index_type = None
-            for dataset in available_datasets:
-                index_type = dataset.indexing_technique
-                document_ids_filter = None
-                if dataset.provider != "external":
-                    if metadata_condition and not metadata_filter_document_ids:
-                        continue
-                    if metadata_filter_document_ids:
-                        document_ids = metadata_filter_document_ids.get(dataset.id, [])
-                        if document_ids:
-                            document_ids_filter = document_ids
-                        else:
+        try:
+            with flask_app.app_context():
+                threads = []
+                all_documents_item: list[Document] = []
+                index_type = None
+                for dataset in available_datasets:
+                    # Check for cancellation signal
+                    if cancel_event and cancel_event.is_set():
+                        break
+                    index_type = dataset.indexing_technique
+                    document_ids_filter = None
+                    if dataset.provider != "external":
+                        if metadata_condition and not metadata_filter_document_ids:
                             continue
-                retrieval_thread = threading.Thread(
-                    target=self._retriever,
-                    kwargs={
-                        "flask_app": flask_app,
-                        "dataset_id": dataset.id,
-                        "query": query,
-                        "top_k": top_k,
-                        "all_documents": all_documents_item,
-                        "document_ids_filter": document_ids_filter,
-                        "metadata_condition": metadata_condition,
-                        "attachment_ids": [attachment_id] if attachment_id else None,
-                    },
-                )
-                threads.append(retrieval_thread)
-                retrieval_thread.start()
-            for thread in threads:
-                thread.join()
+                        if metadata_filter_document_ids:
+                            document_ids = metadata_filter_document_ids.get(dataset.id, [])
+                            if document_ids:
+                                document_ids_filter = document_ids
+                            else:
+                                continue
+                    retrieval_thread = threading.Thread(
+                        target=self._retriever,
+                        kwargs={
+                            "flask_app": flask_app,
+                            "dataset_id": dataset.id,
+                            "query": query,
+                            "top_k": top_k,
+                            "all_documents": all_documents_item,
+                            "document_ids_filter": document_ids_filter,
+                            "metadata_condition": metadata_condition,
+                            "attachment_ids": [attachment_id] if attachment_id else None,
+                        },
+                    )
+                    threads.append(retrieval_thread)
+                    retrieval_thread.start()
+
+                # Poll threads with short timeout to respond quickly to cancellation
+                while any(t.is_alive() for t in threads):
+                    for thread in threads:
+                        thread.join(timeout=0.1)
+                        if cancel_event and cancel_event.is_set():
+                            break
+                    if cancel_event and cancel_event.is_set():
+                        break
 
             if reranking_enable:
                 # do rerank for searched documents
@@ -1455,3 +1501,8 @@ class DatasetRetrieval:
                     all_documents_item = all_documents_item[:top_k] if top_k else all_documents_item
             if all_documents_item:
                 all_documents.extend(all_documents_item)
+        except Exception as e:
+            if cancel_event:
+                cancel_event.set()
+            if thread_exceptions is not None:
+                thread_exceptions.append(e)

api/core/tools/mcp_tool/tool.py

@@ -6,7 +6,15 @@ from typing import Any
 
 from core.mcp.auth_client import MCPClientWithAuthRetry
 from core.mcp.error import MCPConnectionError
-from core.mcp.types import AudioContent, CallToolResult, ImageContent, TextContent
+from core.mcp.types import (
+    AudioContent,
+    BlobResourceContents,
+    CallToolResult,
+    EmbeddedResource,
+    ImageContent,
+    TextContent,
+    TextResourceContents,
+)
 from core.tools.__base.tool import Tool
 from core.tools.__base.tool_runtime import ToolRuntime
 from core.tools.entities.tool_entities import ToolEntity, ToolInvokeMessage, ToolProviderType
@@ -53,10 +61,19 @@ class MCPTool(Tool):
         for content in result.content:
             if isinstance(content, TextContent):
                 yield from self._process_text_content(content)
-            elif isinstance(content, ImageContent):
-                yield self._process_image_content(content)
-            elif isinstance(content, AudioContent):
-                yield self._process_audio_content(content)
+            elif isinstance(content, ImageContent | AudioContent):
+                yield self.create_blob_message(
+                    blob=base64.b64decode(content.data), meta={"mime_type": content.mimeType}
+                )
+            elif isinstance(content, EmbeddedResource):
+                resource = content.resource
+                if isinstance(resource, TextResourceContents):
+                    yield self.create_text_message(resource.text)
+                elif isinstance(resource, BlobResourceContents):
+                    mime_type = resource.mimeType or "application/octet-stream"
+                    yield self.create_blob_message(blob=base64.b64decode(resource.blob), meta={"mime_type": mime_type})
+                else:
+                    raise ToolInvokeError(f"Unsupported embedded resource type: {type(resource)}")
             else:
                 logger.warning("Unsupported content type=%s", type(content))
 
@@ -101,14 +118,6 @@ class MCPTool(Tool):
         for item in json_list:
             yield self.create_json_message(item)
 
-    def _process_image_content(self, content: ImageContent) -> ToolInvokeMessage:
-        """Process image content and return a blob message."""
-        return self.create_blob_message(blob=base64.b64decode(content.data), meta={"mime_type": content.mimeType})
-
-    def _process_audio_content(self, content: AudioContent) -> ToolInvokeMessage:
-        """Process audio content and return a blob message."""
-        return self.create_blob_message(blob=base64.b64decode(content.data), meta={"mime_type": content.mimeType})
-
     def fork_tool_runtime(self, runtime: ToolRuntime) -> "MCPTool":
         return MCPTool(
             entity=self.entity,

api/core/tools/utils/parser.py

@@ -378,7 +378,7 @@ class ApiBasedToolSchemaParser:
     @staticmethod
     def auto_parse_to_tool_bundle(
         content: str, extra_info: dict | None = None, warning: dict | None = None
-    ) -> tuple[list[ApiToolBundle], str]:
+    ) -> tuple[list[ApiToolBundle], ApiProviderSchemaType]:
         """
         auto parse to tool bundle
 

api/core/tools/utils/text_processing_utils.py

@@ -4,6 +4,7 @@ import re
 def remove_leading_symbols(text: str) -> str:
     """
     Remove leading punctuation or symbols from the given text.
+    Preserves markdown links like [text](url) at the start.
 
     Args:
         text (str): The input text to process.
@@ -11,6 +12,11 @@ def remove_leading_symbols(text: str) -> str:
     Returns:
         str: The text with leading punctuation or symbols removed.
     """
+    # Check if text starts with a markdown link - preserve it
+    markdown_link_pattern = r"^\[([^\]]+)\]\((https?://[^)]+)\)"
+    if re.match(markdown_link_pattern, text):
+        return text
+
     # Match Unicode ranges for punctuation and symbols
     # FIXME this pattern is confused quick fix for #11868 maybe refactor it later
     pattern = r'^[\[\]\u2000-\u2025\u2027-\u206F\u2E00-\u2E7F\u3000-\u300F\u3011-\u303F"#$%&\'()*+,./:;<=>?@^_`~]+'

api/core/tools/workflow_as_tool/provider.py

@@ -5,6 +5,7 @@ from sqlalchemy.orm import Session
 
 from core.app.app_config.entities import VariableEntity, VariableEntityType
 from core.app.apps.workflow.app_config_manager import WorkflowAppConfigManager
+from core.db.session_factory import session_factory
 from core.plugin.entities.parameters import PluginParameterOption
 from core.tools.__base.tool_provider import ToolProviderController
 from core.tools.__base.tool_runtime import ToolRuntime
@@ -47,33 +48,29 @@ class WorkflowToolProviderController(ToolProviderController):
 
     @classmethod
     def from_db(cls, db_provider: WorkflowToolProvider) -> "WorkflowToolProviderController":
-        with Session(db.engine, expire_on_commit=False) as session, session.begin():
-            provider = session.get(WorkflowToolProvider, db_provider.id) if db_provider.id else None
-            if not provider:
-                raise ValueError("workflow provider not found")
-            app = session.get(App, provider.app_id)
+        with session_factory.create_session() as session, session.begin():
+            app = session.get(App, db_provider.app_id)
             if not app:
                 raise ValueError("app not found")
 
-            user = session.get(Account, provider.user_id) if provider.user_id else None
-
+            user = session.get(Account, db_provider.user_id) if db_provider.user_id else None
             controller = WorkflowToolProviderController(
                 entity=ToolProviderEntity(
                     identity=ToolProviderIdentity(
                         author=user.name if user else "",
-                        name=provider.label,
-                        label=I18nObject(en_US=provider.label, zh_Hans=provider.label),
-                        description=I18nObject(en_US=provider.description, zh_Hans=provider.description),
-                        icon=provider.icon,
+                        name=db_provider.label,
+                        label=I18nObject(en_US=db_provider.label, zh_Hans=db_provider.label),
+                        description=I18nObject(en_US=db_provider.description, zh_Hans=db_provider.description),
+                        icon=db_provider.icon,
                     ),
                     credentials_schema=[],
                     plugin_id=None,
                 ),
-                provider_id=provider.id or "",
+                provider_id=db_provider.id,
             )
 
             controller.tools = [
-                controller._get_db_provider_tool(provider, app, session=session, user=user),
+                controller._get_db_provider_tool(db_provider, app, session=session, user=user),
             ]
 
         return controller

api/core/workflow/graph_engine/graph_traversal/skip_propagator.py

@@ -60,6 +60,7 @@ class SkipPropagator:
         if edge_states["has_taken"]:
             # Enqueue node
             self._state_manager.enqueue_node(downstream_node_id)
+            self._state_manager.start_execution(downstream_node_id)
             return
 
         # All edges are skipped, propagate skip to this node

api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py

@@ -6,7 +6,7 @@ from collections import defaultdict
 from collections.abc import Mapping, Sequence
 from typing import TYPE_CHECKING, Any, cast
 
-from sqlalchemy import and_, func, literal, or_, select
+from sqlalchemy import and_, func, or_, select
 from sqlalchemy.orm import sessionmaker
 
 from core.app.app_config.entities import DatasetRetrieveConfigEntity
@@ -460,7 +460,7 @@ class KnowledgeRetrievalNode(LLMUsageTrackingMixin, Node[KnowledgeRetrievalNodeD
             if automatic_metadata_filters:
                 conditions = []
                 for sequence, filter in enumerate(automatic_metadata_filters):
-                    self._process_metadata_filter_func(
+                    DatasetRetrieval.process_metadata_filter_func(
                         sequence,
                         filter.get("condition", ""),
                         filter.get("metadata_name", ""),
@@ -504,7 +504,7 @@ class KnowledgeRetrievalNode(LLMUsageTrackingMixin, Node[KnowledgeRetrievalNodeD
                             value=expected_value,
                         )
                     )
-                    filters = self._process_metadata_filter_func(
+                    filters = DatasetRetrieval.process_metadata_filter_func(
                         sequence,
                         condition.comparison_operator,
                         metadata_name,
@@ -603,87 +603,6 @@ class KnowledgeRetrievalNode(LLMUsageTrackingMixin, Node[KnowledgeRetrievalNodeD
             return [], usage
         return automatic_metadata_filters, usage
 
-    def _process_metadata_filter_func(
-        self, sequence: int, condition: str, metadata_name: str, value: Any, filters: list[Any]
-    ) -> list[Any]:
-        if value is None and condition not in ("empty", "not empty"):
-            return filters
-
-        json_field = Document.doc_metadata[metadata_name].as_string()
-
-        match condition:
-            case "contains":
-                filters.append(json_field.like(f"%{value}%"))
-
-            case "not contains":
-                filters.append(json_field.notlike(f"%{value}%"))
-
-            case "start with":
-                filters.append(json_field.like(f"{value}%"))
-
-            case "end with":
-                filters.append(json_field.like(f"%{value}"))
-            case "in":
-                if isinstance(value, str):
-                    value_list = [v.strip() for v in value.split(",") if v.strip()]
-                elif isinstance(value, (list, tuple)):
-                    value_list = [str(v) for v in value if v is not None]
-                else:
-                    value_list = [str(value)] if value is not None else []
-
-                if not value_list:
-                    filters.append(literal(False))
-                else:
-                    filters.append(json_field.in_(value_list))
-
-            case "not in":
-                if isinstance(value, str):
-                    value_list = [v.strip() for v in value.split(",") if v.strip()]
-                elif isinstance(value, (list, tuple)):
-                    value_list = [str(v) for v in value if v is not None]
-                else:
-                    value_list = [str(value)] if value is not None else []
-
-                if not value_list:
-                    filters.append(literal(True))
-                else:
-                    filters.append(json_field.notin_(value_list))
-
-            case "is" | "=":
-                if isinstance(value, str):
-                    filters.append(json_field == value)
-                elif isinstance(value, (int, float)):
-                    filters.append(Document.doc_metadata[metadata_name].as_float() == value)
-
-            case "is not" | "≠":
-                if isinstance(value, str):
-                    filters.append(json_field != value)
-                elif isinstance(value, (int, float)):
-                    filters.append(Document.doc_metadata[metadata_name].as_float() != value)
-
-            case "empty":
-                filters.append(Document.doc_metadata[metadata_name].is_(None))
-
-            case "not empty":
-                filters.append(Document.doc_metadata[metadata_name].isnot(None))
-
-            case "before" | "<":
-                filters.append(Document.doc_metadata[metadata_name].as_float() < value)
-
-            case "after" | ">":
-                filters.append(Document.doc_metadata[metadata_name].as_float() > value)
-
-            case "≤" | "<=":
-                filters.append(Document.doc_metadata[metadata_name].as_float() <= value)
-
-            case "≥" | ">=":
-                filters.append(Document.doc_metadata[metadata_name].as_float() >= value)
-
-            case _:
-                pass
-
-        return filters
-
     @classmethod
     def _extract_variable_selector_to_variable_mapping(
         cls,

api/extensions/ext_celery.py

@@ -12,9 +12,8 @@ from dify_app import DifyApp
 
 def _get_celery_ssl_options() -> dict[str, Any] | None:
     """Get SSL configuration for Celery broker/backend connections."""
-    # Use REDIS_USE_SSL for consistency with the main Redis client
     # Only apply SSL if we're using Redis as broker/backend
-    if not dify_config.REDIS_USE_SSL:
+    if not dify_config.BROKER_USE_SSL:
         return None
 
     # Check if Celery is actually using Redis

api/extensions/storage/tencent_cos_storage.py

@@ -13,12 +13,20 @@ class TencentCosStorage(BaseStorage):
         super().__init__()
 
         self.bucket_name = dify_config.TENCENT_COS_BUCKET_NAME
-        config = CosConfig(
-            Region=dify_config.TENCENT_COS_REGION,
-            SecretId=dify_config.TENCENT_COS_SECRET_ID,
-            SecretKey=dify_config.TENCENT_COS_SECRET_KEY,
-            Scheme=dify_config.TENCENT_COS_SCHEME,
-        )
+        if dify_config.TENCENT_COS_CUSTOM_DOMAIN:
+            config = CosConfig(
+                Domain=dify_config.TENCENT_COS_CUSTOM_DOMAIN,
+                SecretId=dify_config.TENCENT_COS_SECRET_ID,
+                SecretKey=dify_config.TENCENT_COS_SECRET_KEY,
+                Scheme=dify_config.TENCENT_COS_SCHEME,
+            )
+        else:
+            config = CosConfig(
+                Region=dify_config.TENCENT_COS_REGION,
+                SecretId=dify_config.TENCENT_COS_SECRET_ID,
+                SecretKey=dify_config.TENCENT_COS_SECRET_KEY,
+                Scheme=dify_config.TENCENT_COS_SCHEME,
+            )
         self.client = CosS3Client(config)
 
     def save(self, filename, data):

api/libs/archive_storage.py

@@ -0,0 +1,347 @@
+"""
+Archive Storage Client for S3-compatible storage.
+
+This module provides a dedicated storage client for archiving or exporting logs
+to S3-compatible object storage.
+"""
+
+import base64
+import datetime
+import gzip
+import hashlib
+import logging
+from collections.abc import Generator
+from typing import Any, cast
+
+import boto3
+import orjson
+from botocore.client import Config
+from botocore.exceptions import ClientError
+
+from configs import dify_config
+
+logger = logging.getLogger(__name__)
+
+
+class ArchiveStorageError(Exception):
+    """Base exception for archive storage operations."""
+
+    pass
+
+
+class ArchiveStorageNotConfiguredError(ArchiveStorageError):
+    """Raised when archive storage is not properly configured."""
+
+    pass
+
+
+class ArchiveStorage:
+    """
+    S3-compatible storage client for archiving or exporting.
+
+    This client provides methods for storing and retrieving archived data in JSONL+gzip format.
+    """
+
+    def __init__(self, bucket: str):
+        if not dify_config.ARCHIVE_STORAGE_ENABLED:
+            raise ArchiveStorageNotConfiguredError("Archive storage is not enabled")
+
+        if not bucket:
+            raise ArchiveStorageNotConfiguredError("Archive storage bucket is not configured")
+        if not all(
+            [
+                dify_config.ARCHIVE_STORAGE_ENDPOINT,
+                bucket,
+                dify_config.ARCHIVE_STORAGE_ACCESS_KEY,
+                dify_config.ARCHIVE_STORAGE_SECRET_KEY,
+            ]
+        ):
+            raise ArchiveStorageNotConfiguredError(
+                "Archive storage configuration is incomplete. "
+                "Required: ARCHIVE_STORAGE_ENDPOINT, ARCHIVE_STORAGE_ACCESS_KEY, "
+                "ARCHIVE_STORAGE_SECRET_KEY, and a bucket name"
+            )
+
+        self.bucket = bucket
+        self.client = boto3.client(
+            "s3",
+            endpoint_url=dify_config.ARCHIVE_STORAGE_ENDPOINT,
+            aws_access_key_id=dify_config.ARCHIVE_STORAGE_ACCESS_KEY,
+            aws_secret_access_key=dify_config.ARCHIVE_STORAGE_SECRET_KEY,
+            region_name=dify_config.ARCHIVE_STORAGE_REGION,
+            config=Config(s3={"addressing_style": "path"}),
+        )
+
+        # Verify bucket accessibility
+        try:
+            self.client.head_bucket(Bucket=self.bucket)
+        except ClientError as e:
+            error_code = e.response.get("Error", {}).get("Code")
+            if error_code == "404":
+                raise ArchiveStorageNotConfiguredError(f"Archive bucket '{self.bucket}' does not exist")
+            elif error_code == "403":
+                raise ArchiveStorageNotConfiguredError(f"Access denied to archive bucket '{self.bucket}'")
+            else:
+                raise ArchiveStorageError(f"Failed to access archive bucket: {e}")
+
+    def put_object(self, key: str, data: bytes) -> str:
+        """
+        Upload an object to the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+            data: Binary data to upload
+
+        Returns:
+            MD5 checksum of the uploaded data
+
+        Raises:
+            ArchiveStorageError: If upload fails
+        """
+        checksum = hashlib.md5(data).hexdigest()
+        try:
+            self.client.put_object(
+                Bucket=self.bucket,
+                Key=key,
+                Body=data,
+                ContentMD5=self._content_md5(data),
+            )
+            logger.debug("Uploaded object: %s (size=%d, checksum=%s)", key, len(data), checksum)
+            return checksum
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to upload object '{key}': {e}")
+
+    def get_object(self, key: str) -> bytes:
+        """
+        Download an object from the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Returns:
+            Binary data of the object
+
+        Raises:
+            ArchiveStorageError: If download fails
+            FileNotFoundError: If object does not exist
+        """
+        try:
+            response = self.client.get_object(Bucket=self.bucket, Key=key)
+            return response["Body"].read()
+        except ClientError as e:
+            error_code = e.response.get("Error", {}).get("Code")
+            if error_code == "NoSuchKey":
+                raise FileNotFoundError(f"Archive object not found: {key}")
+            raise ArchiveStorageError(f"Failed to download object '{key}': {e}")
+
+    def get_object_stream(self, key: str) -> Generator[bytes, None, None]:
+        """
+        Stream an object from the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Yields:
+            Chunks of binary data
+
+        Raises:
+            ArchiveStorageError: If download fails
+            FileNotFoundError: If object does not exist
+        """
+        try:
+            response = self.client.get_object(Bucket=self.bucket, Key=key)
+            yield from response["Body"].iter_chunks()
+        except ClientError as e:
+            error_code = e.response.get("Error", {}).get("Code")
+            if error_code == "NoSuchKey":
+                raise FileNotFoundError(f"Archive object not found: {key}")
+            raise ArchiveStorageError(f"Failed to stream object '{key}': {e}")
+
+    def object_exists(self, key: str) -> bool:
+        """
+        Check if an object exists in the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Returns:
+            True if object exists, False otherwise
+        """
+        try:
+            self.client.head_object(Bucket=self.bucket, Key=key)
+            return True
+        except ClientError:
+            return False
+
+    def delete_object(self, key: str) -> None:
+        """
+        Delete an object from the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Raises:
+            ArchiveStorageError: If deletion fails
+        """
+        try:
+            self.client.delete_object(Bucket=self.bucket, Key=key)
+            logger.debug("Deleted object: %s", key)
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to delete object '{key}': {e}")
+
+    def generate_presigned_url(self, key: str, expires_in: int = 3600) -> str:
+        """
+        Generate a pre-signed URL for downloading an object.
+
+        Args:
+            key: Object key (path) within the bucket
+            expires_in: URL validity duration in seconds (default: 1 hour)
+
+        Returns:
+            Pre-signed URL string.
+
+        Raises:
+            ArchiveStorageError: If generation fails
+        """
+        try:
+            return self.client.generate_presigned_url(
+                ClientMethod="get_object",
+                Params={"Bucket": self.bucket, "Key": key},
+                ExpiresIn=expires_in,
+            )
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to generate pre-signed URL for '{key}': {e}")
+
+    def list_objects(self, prefix: str) -> list[str]:
+        """
+        List objects under a given prefix.
+
+        Args:
+            prefix: Object key prefix to filter by
+
+        Returns:
+            List of object keys matching the prefix
+        """
+        keys = []
+        paginator = self.client.get_paginator("list_objects_v2")
+
+        try:
+            for page in paginator.paginate(Bucket=self.bucket, Prefix=prefix):
+                for obj in page.get("Contents", []):
+                    keys.append(obj["Key"])
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to list objects with prefix '{prefix}': {e}")
+
+        return keys
+
+    @staticmethod
+    def _content_md5(data: bytes) -> str:
+        """Calculate base64-encoded MD5 for Content-MD5 header."""
+        return base64.b64encode(hashlib.md5(data).digest()).decode()
+
+    @staticmethod
+    def serialize_to_jsonl_gz(records: list[dict[str, Any]]) -> bytes:
+        """
+        Serialize records to gzipped JSONL format.
+
+        Args:
+            records: List of dictionaries to serialize
+
+        Returns:
+            Gzipped JSONL bytes
+        """
+        lines = []
+        for record in records:
+            # Convert datetime objects to ISO format strings
+            serialized = ArchiveStorage._serialize_record(record)
+            lines.append(orjson.dumps(serialized))
+
+        jsonl_content = b"\n".join(lines)
+        if jsonl_content:
+            jsonl_content += b"\n"
+
+        return gzip.compress(jsonl_content)
+
+    @staticmethod
+    def deserialize_from_jsonl_gz(data: bytes) -> list[dict[str, Any]]:
+        """
+        Deserialize gzipped JSONL data to records.
+
+        Args:
+            data: Gzipped JSONL bytes
+
+        Returns:
+            List of dictionaries
+        """
+        jsonl_content = gzip.decompress(data)
+        records = []
+
+        for line in jsonl_content.splitlines():
+            if line:
+                records.append(orjson.loads(line))
+
+        return records
+
+    @staticmethod
+    def _serialize_record(record: dict[str, Any]) -> dict[str, Any]:
+        """Serialize a single record, converting special types."""
+
+        def _serialize(item: Any) -> Any:
+            if isinstance(item, datetime.datetime):
+                return item.isoformat()
+            if isinstance(item, dict):
+                return {key: _serialize(value) for key, value in item.items()}
+            if isinstance(item, list):
+                return [_serialize(value) for value in item]
+            return item
+
+        return cast(dict[str, Any], _serialize(record))
+
+    @staticmethod
+    def compute_checksum(data: bytes) -> str:
+        """Compute MD5 checksum of data."""
+        return hashlib.md5(data).hexdigest()
+
+
+# Singleton instance (lazy initialization)
+_archive_storage: ArchiveStorage | None = None
+_export_storage: ArchiveStorage | None = None
+
+
+def get_archive_storage() -> ArchiveStorage:
+    """
+    Get the archive storage singleton instance.
+
+    Returns:
+        ArchiveStorage instance
+
+    Raises:
+        ArchiveStorageNotConfiguredError: If archive storage is not configured
+    """
+    global _archive_storage
+    if _archive_storage is None:
+        archive_bucket = dify_config.ARCHIVE_STORAGE_ARCHIVE_BUCKET
+        if not archive_bucket:
+            raise ArchiveStorageNotConfiguredError(
+                "Archive storage bucket is not configured. Required: ARCHIVE_STORAGE_ARCHIVE_BUCKET"
+            )
+        _archive_storage = ArchiveStorage(bucket=archive_bucket)
+    return _archive_storage
+
+
+def get_export_storage() -> ArchiveStorage:
+    """
+    Get the export storage singleton instance.
+
+    Returns:
+        ArchiveStorage instance
+    """
+    global _export_storage
+    if _export_storage is None:
+        export_bucket = dify_config.ARCHIVE_STORAGE_EXPORT_BUCKET
+        if not export_bucket:
+            raise ArchiveStorageNotConfiguredError(
+                "Archive export bucket is not configured. Required: ARCHIVE_STORAGE_EXPORT_BUCKET"
+            )
+        _export_storage = ArchiveStorage(bucket=export_bucket)
+    return _export_storage

api/schedule/queue_monitor_task.py

@@ -16,6 +16,11 @@ celery_redis = Redis(
     port=redis_config.get("port") or 6379,
     password=redis_config.get("password") or None,
     db=int(redis_config.get("virtual_host")) if redis_config.get("virtual_host") else 1,
+    ssl=bool(dify_config.BROKER_USE_SSL),
+    ssl_ca_certs=dify_config.REDIS_SSL_CA_CERTS if dify_config.BROKER_USE_SSL else None,
+    ssl_cert_reqs=getattr(dify_config, "REDIS_SSL_CERT_REQS", None) if dify_config.BROKER_USE_SSL else None,
+    ssl_certfile=getattr(dify_config, "REDIS_SSL_CERTFILE", None) if dify_config.BROKER_USE_SSL else None,
+    ssl_keyfile=getattr(dify_config, "REDIS_SSL_KEYFILE", None) if dify_config.BROKER_USE_SSL else None,
 )
 
 logger = logging.getLogger(__name__)

api/services/app_generate_service.py

@@ -14,7 +14,8 @@ from enums.quota_type import QuotaType, unlimited
 from extensions.otel import AppGenerateHandler, trace_span
 from models.model import Account, App, AppMode, EndUser
 from models.workflow import Workflow
-from services.errors.app import InvokeRateLimitError, QuotaExceededError, WorkflowIdFormatError, WorkflowNotFoundError
+from services.errors.app import QuotaExceededError, WorkflowIdFormatError, WorkflowNotFoundError
+from services.errors.llm import InvokeRateLimitError
 from services.workflow_service import WorkflowService
 
 

api/services/async_workflow_service.py

@@ -21,7 +21,7 @@ from models.model import App, EndUser
 from models.trigger import WorkflowTriggerLog
 from models.workflow import Workflow
 from repositories.sqlalchemy_workflow_trigger_log_repository import SQLAlchemyWorkflowTriggerLogRepository
-from services.errors.app import InvokeRateLimitError, QuotaExceededError, WorkflowNotFoundError
+from services.errors.app import QuotaExceededError, WorkflowNotFoundError, WorkflowQuotaLimitError
 from services.workflow.entities import AsyncTriggerResponse, TriggerData, WorkflowTaskData
 from services.workflow.queue_dispatcher import QueueDispatcherManager, QueuePriority
 from services.workflow_service import WorkflowService
@@ -141,7 +141,7 @@ class AsyncWorkflowService:
             trigger_log_repo.update(trigger_log)
             session.commit()
 
-            raise InvokeRateLimitError(
+            raise WorkflowQuotaLimitError(
                 f"Workflow execution quota limit reached for tenant {trigger_data.tenant_id}"
             ) from e
 

api/services/billing_service.py

@@ -1,3 +1,4 @@
+import json
 import logging
 import os
 from collections.abc import Sequence
@@ -31,6 +32,11 @@ class BillingService:
 
     compliance_download_rate_limiter = RateLimiter("compliance_download_rate_limiter", 4, 60)
 
+    # Redis key prefix for tenant plan cache
+    _PLAN_CACHE_KEY_PREFIX = "tenant_plan:"
+    # Cache TTL: 10 minutes
+    _PLAN_CACHE_TTL = 600
+
     @classmethod
     def get_info(cls, tenant_id: str):
         params = {"tenant_id": tenant_id}
@@ -272,14 +278,110 @@ class BillingService:
                 data = resp.get("data", {})
 
                 for tenant_id, plan in data.items():
-                    subscription_plan = subscription_adapter.validate_python(plan)
-                    results[tenant_id] = subscription_plan
+                    try:
+                        subscription_plan = subscription_adapter.validate_python(plan)
+                        results[tenant_id] = subscription_plan
+                    except Exception:
+                        logger.exception(
+                            "get_plan_bulk: failed to validate subscription plan for tenant(%s)", tenant_id
+                        )
+                        continue
             except Exception:
-                logger.exception("Failed to fetch billing info batch for tenants: %s", chunk)
+                logger.exception("get_plan_bulk: failed to fetch billing info batch for tenants: %s", chunk)
                 continue
 
         return results
 
+    @classmethod
+    def _make_plan_cache_key(cls, tenant_id: str) -> str:
+        return f"{cls._PLAN_CACHE_KEY_PREFIX}{tenant_id}"
+
+    @classmethod
+    def get_plan_bulk_with_cache(cls, tenant_ids: Sequence[str]) -> dict[str, SubscriptionPlan]:
+        """
+        Bulk fetch billing subscription plan with cache to reduce billing API loads in batch job scenarios.
+
+        NOTE: if you want to high data consistency, use get_plan_bulk instead.
+
+        Returns:
+            Mapping of tenant_id -> {plan: str, expiration_date: int}
+        """
+        tenant_plans: dict[str, SubscriptionPlan] = {}
+
+        if not tenant_ids:
+            return tenant_plans
+
+        subscription_adapter = TypeAdapter(SubscriptionPlan)
+
+        # Step 1: Batch fetch from Redis cache using mget
+        redis_keys = [cls._make_plan_cache_key(tenant_id) for tenant_id in tenant_ids]
+        try:
+            cached_values = redis_client.mget(redis_keys)
+
+            if len(cached_values) != len(tenant_ids):
+                raise Exception(
+                    "get_plan_bulk_with_cache: unexpected error: redis mget failed: cached values length mismatch"
+                )
+
+            # Map cached values back to tenant_ids
+            cache_misses: list[str] = []
+
+            for tenant_id, cached_value in zip(tenant_ids, cached_values):
+                if cached_value:
+                    try:
+                        # Redis returns bytes, decode to string and parse JSON
+                        json_str = cached_value.decode("utf-8") if isinstance(cached_value, bytes) else cached_value
+                        plan_dict = json.loads(json_str)
+                        subscription_plan = subscription_adapter.validate_python(plan_dict)
+                        tenant_plans[tenant_id] = subscription_plan
+                    except Exception:
+                        logger.exception(
+                            "get_plan_bulk_with_cache: process tenant(%s) failed, add to cache misses", tenant_id
+                        )
+                        cache_misses.append(tenant_id)
+                else:
+                    cache_misses.append(tenant_id)
+
+            logger.info(
+                "get_plan_bulk_with_cache: cache hits=%s, cache misses=%s",
+                len(tenant_plans),
+                len(cache_misses),
+            )
+        except Exception:
+            logger.exception("get_plan_bulk_with_cache: redis mget failed, falling back to API")
+            cache_misses = list(tenant_ids)
+
+        # Step 2: Fetch missing plans from billing API
+        if cache_misses:
+            bulk_plans = BillingService.get_plan_bulk(cache_misses)
+
+            if bulk_plans:
+                plans_to_cache: dict[str, SubscriptionPlan] = {}
+
+                for tenant_id, subscription_plan in bulk_plans.items():
+                    tenant_plans[tenant_id] = subscription_plan
+                    plans_to_cache[tenant_id] = subscription_plan
+
+                # Step 3: Batch update Redis cache using pipeline
+                if plans_to_cache:
+                    try:
+                        pipe = redis_client.pipeline()
+                        for tenant_id, subscription_plan in plans_to_cache.items():
+                            redis_key = cls._make_plan_cache_key(tenant_id)
+                            # Serialize dict to JSON string
+                            json_str = json.dumps(subscription_plan)
+                            pipe.setex(redis_key, cls._PLAN_CACHE_TTL, json_str)
+                        pipe.execute()
+
+                        logger.info(
+                            "get_plan_bulk_with_cache: cached %s new tenant plans to Redis",
+                            len(plans_to_cache),
+                        )
+                    except Exception:
+                        logger.exception("get_plan_bulk_with_cache: redis pipeline failed")
+
+        return tenant_plans
+
     @classmethod
     def get_expired_subscription_cleanup_whitelist(cls) -> Sequence[str]:
         resp = cls._send_request("GET", "/subscription/cleanup/whitelist")

api/services/enterprise/enterprise_service.py

@@ -110,5 +110,5 @@ class EnterpriseService:
             if not app_id:
                 raise ValueError("app_id must be provided.")
 
-            body = {"appId": app_id}
-            EnterpriseRequest.send_request("DELETE", "/webapp/clean", json=body)
+            params = {"appId": app_id}
+            EnterpriseRequest.send_request("DELETE", "/webapp/clean", params=params)

api/services/errors/app.py

@@ -18,8 +18,8 @@ class WorkflowIdFormatError(Exception):
     pass
 
 
-class InvokeRateLimitError(Exception):
-    """Raised when rate limit is exceeded for workflow invocations."""
+class WorkflowQuotaLimitError(Exception):
+    """Raised when workflow execution quota is exceeded (for async/background workflows)."""
 
     pass
 

api/services/plugin/plugin_parameter_service.py

@@ -146,7 +146,7 @@ class PluginParameterService:
                 provider,
                 action,
                 resolved_credentials,
-                CredentialType.API_KEY.value,
+                original_subscription.credential_type or CredentialType.UNAUTHORIZED.value,
                 parameter,
             )
             .options

api/services/tools/api_tools_manage_service.py

@@ -7,7 +7,6 @@ from httpx import get
 from sqlalchemy import select
 
 from core.entities.provider_entities import ProviderConfig
-from core.helper.tool_provider_cache import ToolProviderListCache
 from core.model_runtime.utils.encoders import jsonable_encoder
 from core.tools.__base.tool_runtime import ToolRuntime
 from core.tools.custom_tool.provider import ApiToolProviderController
@@ -86,7 +85,9 @@ class ApiToolManageService:
             raise ValueError(f"invalid schema: {str(e)}")
 
     @staticmethod
-    def convert_schema_to_tool_bundles(schema: str, extra_info: dict | None = None) -> tuple[list[ApiToolBundle], str]:
+    def convert_schema_to_tool_bundles(
+        schema: str, extra_info: dict | None = None
+    ) -> tuple[list[ApiToolBundle], ApiProviderSchemaType]:
         """
         convert schema to tool bundles
 
@@ -104,7 +105,7 @@ class ApiToolManageService:
         provider_name: str,
         icon: dict,
         credentials: dict,
-        schema_type: str,
+        schema_type: ApiProviderSchemaType,
         schema: str,
         privacy_policy: str,
         custom_disclaimer: str,
@@ -113,9 +114,6 @@ class ApiToolManageService:
         """
         create api tool provider
         """
-        if schema_type not in [member.value for member in ApiProviderSchemaType]:
-            raise ValueError(f"invalid schema type {schema}")
-
         provider_name = provider_name.strip()
 
         # check if the provider exists
@@ -178,9 +176,6 @@ class ApiToolManageService:
         # update labels
         ToolLabelManager.update_tool_labels(provider_controller, labels)
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
         return {"result": "success"}
 
     @staticmethod
@@ -245,18 +240,15 @@ class ApiToolManageService:
         original_provider: str,
         icon: dict,
         credentials: dict,
-        schema_type: str,
+        _schema_type: ApiProviderSchemaType,
         schema: str,
-        privacy_policy: str,
+        privacy_policy: str | None,
         custom_disclaimer: str,
         labels: list[str],
     ):
         """
         update api tool provider
         """
-        if schema_type not in [member.value for member in ApiProviderSchemaType]:
-            raise ValueError(f"invalid schema type {schema}")
-
         provider_name = provider_name.strip()
 
         # check if the provider exists
@@ -281,7 +273,7 @@ class ApiToolManageService:
         provider.icon = json.dumps(icon)
         provider.schema = schema
         provider.description = extra_info.get("description", "")
-        provider.schema_type_str = ApiProviderSchemaType.OPENAPI
+        provider.schema_type_str = schema_type
         provider.tools_str = json.dumps(jsonable_encoder(tool_bundles))
         provider.privacy_policy = privacy_policy
         provider.custom_disclaimer = custom_disclaimer
@@ -322,9 +314,6 @@ class ApiToolManageService:
         # update labels
         ToolLabelManager.update_tool_labels(provider_controller, labels)
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
         return {"result": "success"}
 
     @staticmethod
@@ -347,9 +336,6 @@ class ApiToolManageService:
         db.session.delete(provider)
         db.session.commit()
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
         return {"result": "success"}
 
     @staticmethod
@@ -366,7 +352,7 @@ class ApiToolManageService:
         tool_name: str,
         credentials: dict,
         parameters: dict,
-        schema_type: str,
+        schema_type: ApiProviderSchemaType,
         schema: str,
     ):
         """

api/services/tools/builtin_tools_manage_service.py

@@ -12,7 +12,6 @@ from constants import HIDDEN_VALUE, UNKNOWN_VALUE
 from core.helper.name_generator import generate_incremental_name
 from core.helper.position_helper import is_filtered
 from core.helper.provider_cache import NoOpProviderCredentialCache, ToolProviderCredentialsCache
-from core.helper.tool_provider_cache import ToolProviderListCache
 from core.plugin.entities.plugin_daemon import CredentialType
 from core.tools.builtin_tool.provider import BuiltinToolProviderController
 from core.tools.builtin_tool.providers._positions import BuiltinToolProviderSort
@@ -205,9 +204,6 @@ class BuiltinToolManageService:
                     db_provider.name = name
 
                 session.commit()
-
-                # Invalidate tool providers cache
-                ToolProviderListCache.invalidate_cache(tenant_id)
             except Exception as e:
                 session.rollback()
                 raise ValueError(str(e))
@@ -290,8 +286,6 @@ class BuiltinToolManageService:
                 session.rollback()
                 raise ValueError(str(e))
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id, "builtin")
         return {"result": "success"}
 
     @staticmethod
@@ -409,9 +403,6 @@ class BuiltinToolManageService:
             )
             cache.delete()
 
-            # Invalidate tool providers cache
-            ToolProviderListCache.invalidate_cache(tenant_id)
-
         return {"result": "success"}
 
     @staticmethod
@@ -434,8 +425,6 @@ class BuiltinToolManageService:
             target_provider.is_default = True
             session.commit()
 
-            # Invalidate tool providers cache
-            ToolProviderListCache.invalidate_cache(tenant_id)
         return {"result": "success"}
 
     @staticmethod

api/services/tools/mcp_tools_manage_service.py

@@ -319,8 +319,14 @@ class MCPToolManageService:
         except MCPError as e:
             raise ValueError(f"Failed to connect to MCP server: {e}")
 
-        # Update database with retrieved tools
-        db_provider.tools = json.dumps([tool.model_dump() for tool in tools])
+        # Update database with retrieved tools (ensure description is a non-null string)
+        tools_payload = []
+        for tool in tools:
+            data = tool.model_dump()
+            if data.get("description") is None:
+                data["description"] = ""
+            tools_payload.append(data)
+        db_provider.tools = json.dumps(tools_payload)
         db_provider.authed = True
         db_provider.updated_at = datetime.now()
         self._session.flush()
@@ -620,6 +626,21 @@ class MCPToolManageService:
             server_url_hash=new_server_url_hash,
         )
 
+    @staticmethod
+    def reconnect_with_url(
+        *,
+        server_url: str,
+        headers: dict[str, str],
+        timeout: float | None,
+        sse_read_timeout: float | None,
+    ) -> ReconnectResult:
+        return MCPToolManageService._reconnect_with_url(
+            server_url=server_url,
+            headers=headers,
+            timeout=timeout,
+            sse_read_timeout=sse_read_timeout,
+        )
+
     @staticmethod
     def _reconnect_with_url(
         *,
@@ -642,9 +663,16 @@ class MCPToolManageService:
                 sse_read_timeout=sse_read_timeout,
             ) as mcp_client:
                 tools = mcp_client.list_tools()
+                # Ensure tool descriptions are non-null in payload
+                tools_payload = []
+                for t in tools:
+                    d = t.model_dump()
+                    if d.get("description") is None:
+                        d["description"] = ""
+                    tools_payload.append(d)
                 return ReconnectResult(
                     authed=True,
-                    tools=json.dumps([tool.model_dump() for tool in tools]),
+                    tools=json.dumps(tools_payload),
                     encrypted_credentials=EMPTY_CREDENTIALS_JSON,
                 )
         except MCPAuthError:

api/services/tools/tools_manage_service.py

@@ -1,6 +1,5 @@
 import logging
 
-from core.helper.tool_provider_cache import ToolProviderListCache
 from core.tools.entities.api_entities import ToolProviderTypeApiLiteral
 from core.tools.tool_manager import ToolManager
 from services.tools.tools_transform_service import ToolTransformService
@@ -16,14 +15,6 @@ class ToolCommonService:
 
         :return: the list of tool providers
         """
-        # Try to get from cache first
-        cached_result = ToolProviderListCache.get_cached_providers(tenant_id, typ)
-        if cached_result is not None:
-            logger.debug("Returning cached tool providers for tenant %s, type %s", tenant_id, typ)
-            return cached_result
-
-        # Cache miss - fetch from database
-        logger.debug("Cache miss for tool providers, fetching from database for tenant %s, type %s", tenant_id, typ)
         providers = ToolManager.list_providers_from_api(user_id, tenant_id, typ)
 
         # add icon
@@ -32,7 +23,4 @@ class ToolCommonService:
 
         result = [provider.to_dict() for provider in providers]
 
-        # Cache the result
-        ToolProviderListCache.set_cached_providers(tenant_id, typ, result)
-
         return result

api/services/tools/workflow_tools_manage_service.py

@@ -7,7 +7,6 @@ from typing import Any
 from sqlalchemy import or_, select
 from sqlalchemy.orm import Session
 
-from core.helper.tool_provider_cache import ToolProviderListCache
 from core.model_runtime.utils.encoders import jsonable_encoder
 from core.tools.__base.tool_provider import ToolProviderController
 from core.tools.entities.api_entities import ToolApiEntity, ToolProviderApiEntity
@@ -68,34 +67,31 @@ class WorkflowToolManageService:
         if workflow is None:
             raise ValueError(f"Workflow not found for app {workflow_app_id}")
 
-        with Session(db.engine, expire_on_commit=False) as session, session.begin():
-            workflow_tool_provider = WorkflowToolProvider(
-                tenant_id=tenant_id,
-                user_id=user_id,
-                app_id=workflow_app_id,
-                name=name,
-                label=label,
-                icon=json.dumps(icon),
-                description=description,
-                parameter_configuration=json.dumps(parameters),
-                privacy_policy=privacy_policy,
-                version=workflow.version,
-            )
-            session.add(workflow_tool_provider)
+        workflow_tool_provider = WorkflowToolProvider(
+            tenant_id=tenant_id,
+            user_id=user_id,
+            app_id=workflow_app_id,
+            name=name,
+            label=label,
+            icon=json.dumps(icon),
+            description=description,
+            parameter_configuration=json.dumps(parameters),
+            privacy_policy=privacy_policy,
+            version=workflow.version,
+        )
 
         try:
             WorkflowToolProviderController.from_db(workflow_tool_provider)
         except Exception as e:
             raise ValueError(str(e))
 
+        with Session(db.engine, expire_on_commit=False) as session, session.begin():
+            session.add(workflow_tool_provider)
+
         if labels is not None:
             ToolLabelManager.update_tool_labels(
                 ToolTransformService.workflow_provider_to_controller(workflow_tool_provider), labels
             )
-
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
         return {"result": "success"}
 
     @classmethod
@@ -183,9 +179,6 @@ class WorkflowToolManageService:
                 ToolTransformService.workflow_provider_to_controller(workflow_tool_provider), labels
             )
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
         return {"result": "success"}
 
     @classmethod
@@ -248,9 +241,6 @@ class WorkflowToolManageService:
 
         db.session.commit()
 
-        # Invalidate tool providers cache
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
         return {"result": "success"}
 
     @classmethod

api/services/trigger/trigger_provider_service.py

@@ -868,48 +868,111 @@ class TriggerProviderService:
         if not provider_controller:
             raise ValueError(f"Provider {provider_id} not found")
 
-        subscription = TriggerProviderService.get_subscription_by_id(
-            tenant_id=tenant_id,
-            subscription_id=subscription_id,
-        )
-        if not subscription:
-            raise ValueError(f"Subscription {subscription_id} not found")
+        # Use distributed lock to prevent race conditions on the same subscription
+        lock_key = f"trigger_subscription_rebuild_lock:{tenant_id}_{subscription_id}"
+        with redis_client.lock(lock_key, timeout=20):
+            with Session(db.engine, expire_on_commit=False) as session:
+                try:
+                    # Get subscription within the transaction
+                    subscription: TriggerSubscription | None = (
+                        session.query(TriggerSubscription).filter_by(tenant_id=tenant_id, id=subscription_id).first()
+                    )
+                    if not subscription:
+                        raise ValueError(f"Subscription {subscription_id} not found")
 
-        credential_type = CredentialType.of(subscription.credential_type)
-        if credential_type not in [CredentialType.OAUTH2, CredentialType.API_KEY]:
-            raise ValueError("Credential type not supported for rebuild")
+                    credential_type = CredentialType.of(subscription.credential_type)
+                    if credential_type not in [CredentialType.OAUTH2, CredentialType.API_KEY]:
+                        raise ValueError("Credential type not supported for rebuild")
 
-        # TODO: Trying to invoke update api of the plugin trigger provider
+                    # Decrypt existing credentials for merging
+                    credential_encrypter, _ = create_trigger_provider_encrypter_for_subscription(
+                        tenant_id=tenant_id,
+                        controller=provider_controller,
+                        subscription=subscription,
+                    )
+                    decrypted_credentials = dict(credential_encrypter.decrypt(subscription.credentials))
 
-        # FALLBACK: If the update api is not implemented, delete the previous subscription and create a new one
+                    # Merge credentials: if caller passed HIDDEN_VALUE, retain existing decrypted value
+                    merged_credentials: dict[str, Any] = {
+                        key: value if value != HIDDEN_VALUE else decrypted_credentials.get(key, UNKNOWN_VALUE)
+                        for key, value in credentials.items()
+                    }
 
-        # Delete the previous subscription
-        user_id = subscription.user_id
-        TriggerManager.unsubscribe_trigger(
-            tenant_id=tenant_id,
-            user_id=user_id,
-            provider_id=provider_id,
-            subscription=subscription.to_entity(),
-            credentials=subscription.credentials,
-            credential_type=credential_type,
-        )
+                    user_id = subscription.user_id
 
-        # Create a new subscription with the same subscription_id and endpoint_id
-        new_subscription: TriggerSubscriptionEntity = TriggerManager.subscribe_trigger(
-            tenant_id=tenant_id,
-            user_id=user_id,
-            provider_id=provider_id,
-            endpoint=generate_plugin_trigger_endpoint_url(subscription.endpoint_id),
-            parameters=parameters,
-            credentials=credentials,
-            credential_type=credential_type,
-        )
-        TriggerProviderService.update_trigger_subscription(
-            tenant_id=tenant_id,
-            subscription_id=subscription.id,
-            name=name,
-            parameters=parameters,
-            credentials=credentials,
-            properties=new_subscription.properties,
-            expires_at=new_subscription.expires_at,
-        )
+                    # TODO: Trying to invoke update api of the plugin trigger provider
+
+                    # FALLBACK: If the update api is not implemented,
+                    # delete the previous subscription and create a new one
+
+                    # Unsubscribe the previous subscription (external call, but we'll handle errors)
+                    try:
+                        TriggerManager.unsubscribe_trigger(
+                            tenant_id=tenant_id,
+                            user_id=user_id,
+                            provider_id=provider_id,
+                            subscription=subscription.to_entity(),
+                            credentials=decrypted_credentials,
+                            credential_type=credential_type,
+                        )
+                    except Exception as e:
+                        logger.exception("Error unsubscribing trigger during rebuild", exc_info=e)
+                        # Continue anyway - the subscription might already be deleted externally
+
+                    # Create a new subscription with the same subscription_id and endpoint_id (external call)
+                    new_subscription: TriggerSubscriptionEntity = TriggerManager.subscribe_trigger(
+                        tenant_id=tenant_id,
+                        user_id=user_id,
+                        provider_id=provider_id,
+                        endpoint=generate_plugin_trigger_endpoint_url(subscription.endpoint_id),
+                        parameters=parameters,
+                        credentials=merged_credentials,
+                        credential_type=credential_type,
+                    )
+
+                    # Update the subscription in the same transaction
+                    # Inline update logic to reuse the same session
+                    if name is not None and name != subscription.name:
+                        existing = (
+                            session.query(TriggerSubscription)
+                            .filter_by(tenant_id=tenant_id, provider_id=str(provider_id), name=name)
+                            .first()
+                        )
+                        if existing and existing.id != subscription.id:
+                            raise ValueError(f"Subscription name '{name}' already exists for this provider")
+                        subscription.name = name
+
+                    # Update parameters
+                    subscription.parameters = dict(parameters)
+
+                    # Update credentials with merged (and encrypted) values
+                    subscription.credentials = dict(credential_encrypter.encrypt(merged_credentials))
+
+                    # Update properties
+                    if new_subscription.properties:
+                        properties_encrypter, _ = create_provider_encrypter(
+                            tenant_id=tenant_id,
+                            config=provider_controller.get_properties_schema(),
+                            cache=NoOpProviderCredentialCache(),
+                        )
+                        subscription.properties = dict(properties_encrypter.encrypt(dict(new_subscription.properties)))
+
+                    # Update expiration timestamp
+                    if new_subscription.expires_at is not None:
+                        subscription.expires_at = new_subscription.expires_at
+
+                    # Commit the transaction
+                    session.commit()
+
+                    # Clear subscription cache
+                    delete_cache_for_subscription(
+                        tenant_id=tenant_id,
+                        provider_id=subscription.provider_id,
+                        subscription_id=subscription.id,
+                    )
+
+                except Exception as e:
+                    # Rollback on any error
+                    session.rollback()
+                    logger.exception("Failed to rebuild trigger subscription", exc_info=e)
+                    raise

api/services/trigger/webhook_service.py

@@ -863,10 +863,18 @@ class WebhookService:
                 not_found_in_cache.append(node_id)
                 continue
 
-        with Session(db.engine) as session:
-            try:
-                # lock the concurrent webhook trigger creation
-                redis_client.lock(f"{cls.__WEBHOOK_NODE_CACHE_KEY__}:apps:{app.id}:lock", timeout=10)
+        lock_key = f"{cls.__WEBHOOK_NODE_CACHE_KEY__}:apps:{app.id}:lock"
+        lock = redis_client.lock(lock_key, timeout=10)
+        lock_acquired = False
+
+        try:
+            # acquire the lock with blocking and timeout
+            lock_acquired = lock.acquire(blocking=True, blocking_timeout=10)
+            if not lock_acquired:
+                logger.warning("Failed to acquire lock for webhook sync, app %s", app.id)
+                raise RuntimeError("Failed to acquire lock for webhook trigger synchronization")
+
+            with Session(db.engine) as session:
                 # fetch the non-cached nodes from DB
                 all_records = session.scalars(
                     select(WorkflowWebhookTrigger).where(
@@ -903,11 +911,16 @@ class WebhookService:
                         session.delete(nodes_id_in_db[node_id])
                         redis_client.delete(f"{cls.__WEBHOOK_NODE_CACHE_KEY__}:{app.id}:{node_id}")
                 session.commit()
-            except Exception:
-                logger.exception("Failed to sync webhook relationships for app %s", app.id)
-                raise
-            finally:
-                redis_client.delete(f"{cls.__WEBHOOK_NODE_CACHE_KEY__}:apps:{app.id}:lock")
+        except Exception:
+            logger.exception("Failed to sync webhook relationships for app %s", app.id)
+            raise
+        finally:
+            # release the lock only if it was acquired
+            if lock_acquired:
+                try:
+                    lock.release()
+                except Exception:
+                    logger.exception("Failed to release lock for webhook sync, app %s", app.id)
 
     @classmethod
     def generate_webhook_id(cls) -> str:

api/tests/integration_tests/workflow/nodes/code_executor/test_code_jinja2.py

@@ -7,11 +7,14 @@ CODE_LANGUAGE = CodeLanguage.JINJA2
 
 
 def test_jinja2():
+    """Test basic Jinja2 template rendering."""
     template = "Hello {{template}}"
+    # Template must be base64 encoded to match the new safe embedding approach
+    template_b64 = base64.b64encode(template.encode("utf-8")).decode("utf-8")
     inputs = base64.b64encode(b'{"template": "World"}').decode("utf-8")
     code = (
         Jinja2TemplateTransformer.get_runner_script()
-        .replace(Jinja2TemplateTransformer._code_placeholder, template)
+        .replace(Jinja2TemplateTransformer._template_b64_placeholder, template_b64)
         .replace(Jinja2TemplateTransformer._inputs_placeholder, inputs)
     )
     result = CodeExecutor.execute_code(
@@ -21,6 +24,7 @@ def test_jinja2():
 
 
 def test_jinja2_with_code_template():
+    """Test template rendering via the high-level workflow API."""
     result = CodeExecutor.execute_workflow_code_template(
         language=CODE_LANGUAGE, code="Hello {{template}}", inputs={"template": "World"}
     )
@@ -28,7 +32,64 @@ def test_jinja2_with_code_template():
 
 
 def test_jinja2_get_runner_script():
+    """Test that runner script contains required placeholders."""
     runner_script = Jinja2TemplateTransformer.get_runner_script()
-    assert runner_script.count(Jinja2TemplateTransformer._code_placeholder) == 1
+    assert runner_script.count(Jinja2TemplateTransformer._template_b64_placeholder) == 1
     assert runner_script.count(Jinja2TemplateTransformer._inputs_placeholder) == 1
     assert runner_script.count(Jinja2TemplateTransformer._result_tag) == 2
+
+
+def test_jinja2_template_with_special_characters():
+    """
+    Test that templates with special characters (quotes, newlines) render correctly.
+    This is a regression test for issue #26818 where textarea pre-fill values
+    containing special characters would break template rendering.
+    """
+    # Template with triple quotes, single quotes, double quotes, and newlines
+    template = """<html>
+<body>
+    <input value="{{ task.get('Task ID', '') }}"/>
+    <textarea>{{ task.get('Issues', 'No issues reported') }}</textarea>
+    <p>Status: "{{ status }}"</p>
+    <pre>'''code block'''</pre>
+</body>
+</html>"""
+    inputs = {"task": {"Task ID": "TASK-123", "Issues": "Line 1\nLine 2\nLine 3"}, "status": "completed"}
+
+    result = CodeExecutor.execute_workflow_code_template(language=CODE_LANGUAGE, code=template, inputs=inputs)
+
+    # Verify the template rendered correctly with all special characters
+    output = result["result"]
+    assert 'value="TASK-123"' in output
+    assert "<textarea>Line 1\nLine 2\nLine 3</textarea>" in output
+    assert 'Status: "completed"' in output
+    assert "'''code block'''" in output
+
+
+def test_jinja2_template_with_html_textarea_prefill():
+    """
+    Specific test for HTML textarea with Jinja2 variable pre-fill.
+    Verifies fix for issue #26818.
+    """
+    template = "<textarea name='notes'>{{ notes }}</textarea>"
+    notes_content = "This is a multi-line note.\nWith special chars: 'single' and \"double\" quotes."
+    inputs = {"notes": notes_content}
+
+    result = CodeExecutor.execute_workflow_code_template(language=CODE_LANGUAGE, code=template, inputs=inputs)
+
+    expected_output = f"<textarea name='notes'>{notes_content}</textarea>"
+    assert result["result"] == expected_output
+
+
+def test_jinja2_assemble_runner_script_encodes_template():
+    """Test that assemble_runner_script properly base64 encodes the template."""
+    template = "Hello {{ name }}!"
+    inputs = {"name": "World"}
+
+    script = Jinja2TemplateTransformer.assemble_runner_script(template, inputs)
+
+    # The template should be base64 encoded in the script
+    template_b64 = base64.b64encode(template.encode("utf-8")).decode("utf-8")
+    assert template_b64 in script
+    # The raw template should NOT appear in the script (it's encoded)
+    assert "Hello {{ name }}!" not in script

api/tests/test_containers_integration_tests/services/test_billing_service.py

@@ -0,0 +1,365 @@
+import json
+from unittest.mock import patch
+
+import pytest
+
+from extensions.ext_redis import redis_client
+from services.billing_service import BillingService
+
+
+class TestBillingServiceGetPlanBulkWithCache:
+    """
+    Comprehensive integration tests for get_plan_bulk_with_cache using testcontainers.
+
+    This test class covers all major scenarios:
+    - Cache hit/miss scenarios
+    - Redis operation failures and fallback behavior
+    - Invalid cache data handling
+    - TTL expiration handling
+    - Error recovery and logging
+    """
+
+    @pytest.fixture(autouse=True)
+    def setup_redis_cleanup(self, flask_app_with_containers):
+        """Clean up Redis cache before and after each test."""
+        with flask_app_with_containers.app_context():
+            # Clean up before test
+            yield
+            # Clean up after test
+            # Delete all test cache keys
+            pattern = f"{BillingService._PLAN_CACHE_KEY_PREFIX}*"
+            keys = redis_client.keys(pattern)
+            if keys:
+                redis_client.delete(*keys)
+
+    def _create_test_plan_data(self, plan: str = "sandbox", expiration_date: int = 1735689600):
+        """Helper to create test SubscriptionPlan data."""
+        return {"plan": plan, "expiration_date": expiration_date}
+
+    def _set_cache(self, tenant_id: str, plan_data: dict, ttl: int = 600):
+        """Helper to set cache data in Redis."""
+        cache_key = BillingService._make_plan_cache_key(tenant_id)
+        json_str = json.dumps(plan_data)
+        redis_client.setex(cache_key, ttl, json_str)
+
+    def _get_cache(self, tenant_id: str):
+        """Helper to get cache data from Redis."""
+        cache_key = BillingService._make_plan_cache_key(tenant_id)
+        value = redis_client.get(cache_key)
+        if value:
+            if isinstance(value, bytes):
+                return value.decode("utf-8")
+            return value
+        return None
+
+    def test_get_plan_bulk_with_cache_all_cache_hit(self, flask_app_with_containers):
+        """Test bulk plan retrieval when all tenants are in cache."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2", "tenant-3"]
+            expected_plans = {
+                "tenant-1": self._create_test_plan_data("sandbox", 1735689600),
+                "tenant-2": self._create_test_plan_data("professional", 1767225600),
+                "tenant-3": self._create_test_plan_data("team", 1798761600),
+            }
+
+            # Pre-populate cache
+            for tenant_id, plan_data in expected_plans.items():
+                self._set_cache(tenant_id, plan_data)
+
+            # Act
+            with patch.object(BillingService, "get_plan_bulk") as mock_get_plan_bulk:
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert
+            assert len(result) == 3
+            assert result["tenant-1"]["plan"] == "sandbox"
+            assert result["tenant-1"]["expiration_date"] == 1735689600
+            assert result["tenant-2"]["plan"] == "professional"
+            assert result["tenant-2"]["expiration_date"] == 1767225600
+            assert result["tenant-3"]["plan"] == "team"
+            assert result["tenant-3"]["expiration_date"] == 1798761600
+
+            # Verify API was not called
+            mock_get_plan_bulk.assert_not_called()
+
+    def test_get_plan_bulk_with_cache_all_cache_miss(self, flask_app_with_containers):
+        """Test bulk plan retrieval when all tenants are not in cache."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2"]
+            expected_plans = {
+                "tenant-1": self._create_test_plan_data("sandbox", 1735689600),
+                "tenant-2": self._create_test_plan_data("professional", 1767225600),
+            }
+
+            # Act
+            with patch.object(BillingService, "get_plan_bulk", return_value=expected_plans) as mock_get_plan_bulk:
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert
+            assert len(result) == 2
+            assert result["tenant-1"]["plan"] == "sandbox"
+            assert result["tenant-2"]["plan"] == "professional"
+
+            # Verify API was called with correct tenant_ids
+            mock_get_plan_bulk.assert_called_once_with(tenant_ids)
+
+            # Verify data was written to cache
+            cached_1 = self._get_cache("tenant-1")
+            cached_2 = self._get_cache("tenant-2")
+            assert cached_1 is not None
+            assert cached_2 is not None
+
+            # Verify cache content
+            cached_data_1 = json.loads(cached_1)
+            cached_data_2 = json.loads(cached_2)
+            assert cached_data_1 == expected_plans["tenant-1"]
+            assert cached_data_2 == expected_plans["tenant-2"]
+
+            # Verify TTL is set
+            cache_key_1 = BillingService._make_plan_cache_key("tenant-1")
+            ttl_1 = redis_client.ttl(cache_key_1)
+            assert ttl_1 > 0
+            assert ttl_1 <= 600  # Should be <= 600 seconds
+
+    def test_get_plan_bulk_with_cache_partial_cache_hit(self, flask_app_with_containers):
+        """Test bulk plan retrieval when some tenants are in cache, some are not."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2", "tenant-3"]
+            # Pre-populate cache for tenant-1 and tenant-2
+            self._set_cache("tenant-1", self._create_test_plan_data("sandbox", 1735689600))
+            self._set_cache("tenant-2", self._create_test_plan_data("professional", 1767225600))
+
+            # tenant-3 is not in cache
+            missing_plan = {"tenant-3": self._create_test_plan_data("team", 1798761600)}
+
+            # Act
+            with patch.object(BillingService, "get_plan_bulk", return_value=missing_plan) as mock_get_plan_bulk:
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert
+            assert len(result) == 3
+            assert result["tenant-1"]["plan"] == "sandbox"
+            assert result["tenant-2"]["plan"] == "professional"
+            assert result["tenant-3"]["plan"] == "team"
+
+            # Verify API was called only for missing tenant
+            mock_get_plan_bulk.assert_called_once_with(["tenant-3"])
+
+            # Verify tenant-3 data was written to cache
+            cached_3 = self._get_cache("tenant-3")
+            assert cached_3 is not None
+            cached_data_3 = json.loads(cached_3)
+            assert cached_data_3 == missing_plan["tenant-3"]
+
+    def test_get_plan_bulk_with_cache_redis_mget_failure(self, flask_app_with_containers):
+        """Test fallback to API when Redis mget fails."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2"]
+            expected_plans = {
+                "tenant-1": self._create_test_plan_data("sandbox", 1735689600),
+                "tenant-2": self._create_test_plan_data("professional", 1767225600),
+            }
+
+            # Act
+            with (
+                patch.object(redis_client, "mget", side_effect=Exception("Redis connection error")),
+                patch.object(BillingService, "get_plan_bulk", return_value=expected_plans) as mock_get_plan_bulk,
+            ):
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert
+            assert len(result) == 2
+            assert result["tenant-1"]["plan"] == "sandbox"
+            assert result["tenant-2"]["plan"] == "professional"
+
+            # Verify API was called for all tenants (fallback)
+            mock_get_plan_bulk.assert_called_once_with(tenant_ids)
+
+            # Verify data was written to cache after fallback
+            cached_1 = self._get_cache("tenant-1")
+            cached_2 = self._get_cache("tenant-2")
+            assert cached_1 is not None
+            assert cached_2 is not None
+
+    def test_get_plan_bulk_with_cache_invalid_json_in_cache(self, flask_app_with_containers):
+        """Test fallback to API when cache contains invalid JSON."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2", "tenant-3"]
+
+            # Set valid cache for tenant-1
+            self._set_cache("tenant-1", self._create_test_plan_data("sandbox", 1735689600))
+
+            # Set invalid JSON for tenant-2
+            cache_key_2 = BillingService._make_plan_cache_key("tenant-2")
+            redis_client.setex(cache_key_2, 600, "invalid json {")
+
+            # tenant-3 is not in cache
+            expected_plans = {
+                "tenant-2": self._create_test_plan_data("professional", 1767225600),
+                "tenant-3": self._create_test_plan_data("team", 1798761600),
+            }
+
+            # Act
+            with patch.object(BillingService, "get_plan_bulk", return_value=expected_plans) as mock_get_plan_bulk:
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert
+            assert len(result) == 3
+            assert result["tenant-1"]["plan"] == "sandbox"  # From cache
+            assert result["tenant-2"]["plan"] == "professional"  # From API (fallback)
+            assert result["tenant-3"]["plan"] == "team"  # From API
+
+            # Verify API was called for tenant-2 and tenant-3
+            mock_get_plan_bulk.assert_called_once_with(["tenant-2", "tenant-3"])
+
+            # Verify tenant-2's invalid JSON was replaced with correct data in cache
+            cached_2 = self._get_cache("tenant-2")
+            assert cached_2 is not None
+            cached_data_2 = json.loads(cached_2)
+            assert cached_data_2 == expected_plans["tenant-2"]
+            assert cached_data_2["plan"] == "professional"
+            assert cached_data_2["expiration_date"] == 1767225600
+
+            # Verify tenant-2 cache has correct TTL
+            cache_key_2_new = BillingService._make_plan_cache_key("tenant-2")
+            ttl_2 = redis_client.ttl(cache_key_2_new)
+            assert ttl_2 > 0
+            assert ttl_2 <= 600
+
+            # Verify tenant-3 data was also written to cache
+            cached_3 = self._get_cache("tenant-3")
+            assert cached_3 is not None
+            cached_data_3 = json.loads(cached_3)
+            assert cached_data_3 == expected_plans["tenant-3"]
+
+    def test_get_plan_bulk_with_cache_invalid_plan_data_in_cache(self, flask_app_with_containers):
+        """Test fallback to API when cache data doesn't match SubscriptionPlan schema."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2", "tenant-3"]
+
+            # Set valid cache for tenant-1
+            self._set_cache("tenant-1", self._create_test_plan_data("sandbox", 1735689600))
+
+            # Set invalid plan data for tenant-2 (missing expiration_date)
+            cache_key_2 = BillingService._make_plan_cache_key("tenant-2")
+            invalid_data = json.dumps({"plan": "professional"})  # Missing expiration_date
+            redis_client.setex(cache_key_2, 600, invalid_data)
+
+            # tenant-3 is not in cache
+            expected_plans = {
+                "tenant-2": self._create_test_plan_data("professional", 1767225600),
+                "tenant-3": self._create_test_plan_data("team", 1798761600),
+            }
+
+            # Act
+            with patch.object(BillingService, "get_plan_bulk", return_value=expected_plans) as mock_get_plan_bulk:
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert
+            assert len(result) == 3
+            assert result["tenant-1"]["plan"] == "sandbox"  # From cache
+            assert result["tenant-2"]["plan"] == "professional"  # From API (fallback)
+            assert result["tenant-3"]["plan"] == "team"  # From API
+
+            # Verify API was called for tenant-2 and tenant-3
+            mock_get_plan_bulk.assert_called_once_with(["tenant-2", "tenant-3"])
+
+    def test_get_plan_bulk_with_cache_redis_pipeline_failure(self, flask_app_with_containers):
+        """Test that pipeline failure doesn't affect return value."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2"]
+            expected_plans = {
+                "tenant-1": self._create_test_plan_data("sandbox", 1735689600),
+                "tenant-2": self._create_test_plan_data("professional", 1767225600),
+            }
+
+            # Act
+            with (
+                patch.object(BillingService, "get_plan_bulk", return_value=expected_plans),
+                patch.object(redis_client, "pipeline") as mock_pipeline,
+            ):
+                # Create a mock pipeline that fails on execute
+                mock_pipe = mock_pipeline.return_value
+                mock_pipe.execute.side_effect = Exception("Pipeline execution failed")
+
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert - Function should still return correct result despite pipeline failure
+            assert len(result) == 2
+            assert result["tenant-1"]["plan"] == "sandbox"
+            assert result["tenant-2"]["plan"] == "professional"
+
+            # Verify pipeline was attempted
+            mock_pipeline.assert_called_once()
+
+    def test_get_plan_bulk_with_cache_empty_tenant_ids(self, flask_app_with_containers):
+        """Test with empty tenant_ids list."""
+        with flask_app_with_containers.app_context():
+            # Act
+            with patch.object(BillingService, "get_plan_bulk") as mock_get_plan_bulk:
+                result = BillingService.get_plan_bulk_with_cache([])
+
+            # Assert
+            assert result == {}
+            assert len(result) == 0
+
+            # Verify no API calls
+            mock_get_plan_bulk.assert_not_called()
+
+            # Verify no Redis operations (mget with empty list would return empty list)
+            # But we should check that mget was not called at all
+            # Since we can't easily verify this without more mocking, we just verify the result
+
+    def test_get_plan_bulk_with_cache_ttl_expired(self, flask_app_with_containers):
+        """Test that expired cache keys are treated as cache misses."""
+        with flask_app_with_containers.app_context():
+            # Arrange
+            tenant_ids = ["tenant-1", "tenant-2"]
+
+            # Set cache for tenant-1 with very short TTL (1 second) to simulate expiration
+            self._set_cache("tenant-1", self._create_test_plan_data("sandbox", 1735689600), ttl=1)
+
+            # Wait for TTL to expire (key will be deleted by Redis)
+            import time
+
+            time.sleep(2)
+
+            # Verify cache is expired (key doesn't exist)
+            cache_key_1 = BillingService._make_plan_cache_key("tenant-1")
+            exists = redis_client.exists(cache_key_1)
+            assert exists == 0  # Key doesn't exist (expired)
+
+            # tenant-2 is not in cache
+            expected_plans = {
+                "tenant-1": self._create_test_plan_data("sandbox", 1735689600),
+                "tenant-2": self._create_test_plan_data("professional", 1767225600),
+            }
+
+            # Act
+            with patch.object(BillingService, "get_plan_bulk", return_value=expected_plans) as mock_get_plan_bulk:
+                result = BillingService.get_plan_bulk_with_cache(tenant_ids)
+
+            # Assert
+            assert len(result) == 2
+            assert result["tenant-1"]["plan"] == "sandbox"
+            assert result["tenant-2"]["plan"] == "professional"
+
+            # Verify API was called for both tenants (tenant-1 expired, tenant-2 missing)
+            mock_get_plan_bulk.assert_called_once_with(tenant_ids)
+
+            # Verify both were written to cache with correct TTL
+            cache_key_1_new = BillingService._make_plan_cache_key("tenant-1")
+            cache_key_2 = BillingService._make_plan_cache_key("tenant-2")
+            ttl_1_new = redis_client.ttl(cache_key_1_new)
+            ttl_2 = redis_client.ttl(cache_key_2)
+            assert ttl_1_new > 0
+            assert ttl_1_new <= 600
+            assert ttl_2 > 0
+            assert ttl_2 <= 600

api/tests/test_containers_integration_tests/services/test_trigger_provider_service.py

@@ -0,0 +1,682 @@
+from unittest.mock import MagicMock, patch
+
+import pytest
+from faker import Faker
+
+from constants import HIDDEN_VALUE, UNKNOWN_VALUE
+from core.plugin.entities.plugin_daemon import CredentialType
+from core.trigger.entities.entities import Subscription as TriggerSubscriptionEntity
+from extensions.ext_database import db
+from models.provider_ids import TriggerProviderID
+from models.trigger import TriggerSubscription
+from services.trigger.trigger_provider_service import TriggerProviderService
+
+
+class TestTriggerProviderService:
+    """Integration tests for TriggerProviderService using testcontainers."""
+
+    @pytest.fixture
+    def mock_external_service_dependencies(self):
+        """Mock setup for external service dependencies."""
+        with (
+            patch("services.trigger.trigger_provider_service.TriggerManager") as mock_trigger_manager,
+            patch("services.trigger.trigger_provider_service.redis_client") as mock_redis_client,
+            patch("services.trigger.trigger_provider_service.delete_cache_for_subscription") as mock_delete_cache,
+            patch("services.account_service.FeatureService") as mock_account_feature_service,
+        ):
+            # Setup default mock returns
+            mock_provider_controller = MagicMock()
+            mock_provider_controller.get_credential_schema_config.return_value = MagicMock()
+            mock_provider_controller.get_properties_schema.return_value = MagicMock()
+            mock_trigger_manager.get_trigger_provider.return_value = mock_provider_controller
+
+            # Mock redis lock
+            mock_lock = MagicMock()
+            mock_lock.__enter__ = MagicMock(return_value=None)
+            mock_lock.__exit__ = MagicMock(return_value=None)
+            mock_redis_client.lock.return_value = mock_lock
+
+            # Setup account feature service mock
+            mock_account_feature_service.get_system_features.return_value.is_allow_register = True
+
+            yield {
+                "trigger_manager": mock_trigger_manager,
+                "redis_client": mock_redis_client,
+                "delete_cache": mock_delete_cache,
+                "provider_controller": mock_provider_controller,
+                "account_feature_service": mock_account_feature_service,
+            }
+
+    def _create_test_account_and_tenant(self, db_session_with_containers, mock_external_service_dependencies):
+        """
+        Helper method to create a test account and tenant for testing.
+
+        Args:
+            db_session_with_containers: Database session from testcontainers infrastructure
+            mock_external_service_dependencies: Mock dependencies
+
+        Returns:
+            tuple: (account, tenant) - Created account and tenant instances
+        """
+        fake = Faker()
+
+        from services.account_service import AccountService, TenantService
+
+        # Setup mocks for account creation
+        mock_external_service_dependencies[
+            "account_feature_service"
+        ].get_system_features.return_value.is_allow_register = True
+        mock_external_service_dependencies[
+            "trigger_manager"
+        ].get_trigger_provider.return_value = mock_external_service_dependencies["provider_controller"]
+
+        # Create account and tenant
+        account = AccountService.create_account(
+            email=fake.email(),
+            name=fake.name(),
+            interface_language="en-US",
+            password=fake.password(length=12),
+        )
+        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
+        tenant = account.current_tenant
+
+        return account, tenant
+
+    def _create_test_subscription(
+        self,
+        db_session_with_containers,
+        tenant_id,
+        user_id,
+        provider_id,
+        credential_type,
+        credentials,
+        mock_external_service_dependencies,
+    ):
+        """
+        Helper method to create a test trigger subscription.
+
+        Args:
+            db_session_with_containers: Database session
+            tenant_id: Tenant ID
+            user_id: User ID
+            provider_id: Provider ID
+            credential_type: Credential type
+            credentials: Credentials dict
+            mock_external_service_dependencies: Mock dependencies
+
+        Returns:
+            TriggerSubscription: Created subscription instance
+        """
+        fake = Faker()
+        from core.helper.provider_cache import NoOpProviderCredentialCache
+        from core.helper.provider_encryption import create_provider_encrypter
+
+        # Use mock provider controller to encrypt credentials
+        provider_controller = mock_external_service_dependencies["provider_controller"]
+
+        # Create encrypter for credentials
+        credential_encrypter, _ = create_provider_encrypter(
+            tenant_id=tenant_id,
+            config=provider_controller.get_credential_schema_config(credential_type),
+            cache=NoOpProviderCredentialCache(),
+        )
+
+        subscription = TriggerSubscription(
+            name=fake.word(),
+            tenant_id=tenant_id,
+            user_id=user_id,
+            provider_id=str(provider_id),
+            endpoint_id=fake.uuid4(),
+            parameters={"param1": "value1"},
+            properties={"prop1": "value1"},
+            credentials=dict(credential_encrypter.encrypt(credentials)),
+            credential_type=credential_type.value,
+            credential_expires_at=-1,
+            expires_at=-1,
+        )
+
+        db.session.add(subscription)
+        db.session.commit()
+        db.session.refresh(subscription)
+
+        return subscription
+
+    def test_rebuild_trigger_subscription_success_with_merged_credentials(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test successful rebuild with credential merging (HIDDEN_VALUE handling).
+
+        This test verifies:
+        - Credentials are properly merged (HIDDEN_VALUE replaced with existing values)
+        - Single transaction wraps all operations
+        - Merged credentials are used for subscribe and update
+        - Database state is correctly updated
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.API_KEY
+
+        # Create initial subscription with credentials
+        original_credentials = {"api_key": "original-secret-key", "api_secret": "original-secret"}
+        subscription = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            original_credentials,
+            mock_external_service_dependencies,
+        )
+
+        # Prepare new credentials with HIDDEN_VALUE for api_key (should keep original)
+        # and new value for api_secret (should update)
+        new_credentials = {
+            "api_key": HIDDEN_VALUE,  # Should be replaced with original
+            "api_secret": "new-secret-value",  # Should be updated
+        }
+
+        # Mock subscribe_trigger to return a new subscription entity
+        new_subscription_entity = TriggerSubscriptionEntity(
+            endpoint=subscription.endpoint_id,
+            parameters={"param1": "value1"},
+            properties={"prop1": "new_prop_value"},
+            expires_at=1234567890,
+        )
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.return_value = new_subscription_entity
+
+        # Mock unsubscribe_trigger
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.return_value = MagicMock()
+
+        # Execute rebuild
+        TriggerProviderService.rebuild_trigger_subscription(
+            tenant_id=tenant.id,
+            provider_id=provider_id,
+            subscription_id=subscription.id,
+            credentials=new_credentials,
+            parameters={"param1": "updated_value"},
+            name="updated_name",
+        )
+
+        # Verify unsubscribe was called with decrypted original credentials
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.assert_called_once()
+        unsubscribe_call_args = mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.call_args
+        assert unsubscribe_call_args.kwargs["tenant_id"] == tenant.id
+        assert unsubscribe_call_args.kwargs["provider_id"] == provider_id
+        assert unsubscribe_call_args.kwargs["credential_type"] == credential_type
+
+        # Verify subscribe was called with merged credentials (api_key from original, api_secret new)
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.assert_called_once()
+        subscribe_call_args = mock_external_service_dependencies["trigger_manager"].subscribe_trigger.call_args
+        subscribe_credentials = subscribe_call_args.kwargs["credentials"]
+        assert subscribe_credentials["api_key"] == original_credentials["api_key"]  # Merged from original
+        assert subscribe_credentials["api_secret"] == "new-secret-value"  # New value
+
+        # Verify database state was updated
+        db.session.refresh(subscription)
+        assert subscription.name == "updated_name"
+        assert subscription.parameters == {"param1": "updated_value"}
+
+        # Verify credentials in DB were updated with merged values (decrypt to check)
+        from core.helper.provider_cache import NoOpProviderCredentialCache
+        from core.helper.provider_encryption import create_provider_encrypter
+
+        # Use mock provider controller to decrypt credentials
+        provider_controller = mock_external_service_dependencies["provider_controller"]
+        credential_encrypter, _ = create_provider_encrypter(
+            tenant_id=tenant.id,
+            config=provider_controller.get_credential_schema_config(credential_type),
+            cache=NoOpProviderCredentialCache(),
+        )
+        decrypted_db_credentials = dict(credential_encrypter.decrypt(subscription.credentials))
+        assert decrypted_db_credentials["api_key"] == original_credentials["api_key"]
+        assert decrypted_db_credentials["api_secret"] == "new-secret-value"
+
+        # Verify cache was cleared
+        mock_external_service_dependencies["delete_cache"].assert_called_once_with(
+            tenant_id=tenant.id,
+            provider_id=subscription.provider_id,
+            subscription_id=subscription.id,
+        )
+
+    def test_rebuild_trigger_subscription_with_all_new_credentials(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test rebuild when all credentials are new (no HIDDEN_VALUE).
+
+        This test verifies:
+        - All new credentials are used when no HIDDEN_VALUE is present
+        - Merged credentials contain only new values
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.API_KEY
+
+        # Create initial subscription
+        original_credentials = {"api_key": "original-key", "api_secret": "original-secret"}
+        subscription = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            original_credentials,
+            mock_external_service_dependencies,
+        )
+
+        # All new credentials (no HIDDEN_VALUE)
+        new_credentials = {
+            "api_key": "completely-new-key",
+            "api_secret": "completely-new-secret",
+        }
+
+        new_subscription_entity = TriggerSubscriptionEntity(
+            endpoint=subscription.endpoint_id,
+            parameters={},
+            properties={},
+            expires_at=-1,
+        )
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.return_value = new_subscription_entity
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.return_value = MagicMock()
+
+        # Execute rebuild
+        TriggerProviderService.rebuild_trigger_subscription(
+            tenant_id=tenant.id,
+            provider_id=provider_id,
+            subscription_id=subscription.id,
+            credentials=new_credentials,
+            parameters={},
+        )
+
+        # Verify subscribe was called with all new credentials
+        subscribe_call_args = mock_external_service_dependencies["trigger_manager"].subscribe_trigger.call_args
+        subscribe_credentials = subscribe_call_args.kwargs["credentials"]
+        assert subscribe_credentials["api_key"] == "completely-new-key"
+        assert subscribe_credentials["api_secret"] == "completely-new-secret"
+
+    def test_rebuild_trigger_subscription_with_all_hidden_values(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test rebuild when all credentials are HIDDEN_VALUE (preserve all existing).
+
+        This test verifies:
+        - All HIDDEN_VALUE credentials are replaced with existing values
+        - Original credentials are preserved
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.API_KEY
+
+        original_credentials = {"api_key": "original-key", "api_secret": "original-secret"}
+        subscription = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            original_credentials,
+            mock_external_service_dependencies,
+        )
+
+        # All HIDDEN_VALUE (should preserve all original)
+        new_credentials = {
+            "api_key": HIDDEN_VALUE,
+            "api_secret": HIDDEN_VALUE,
+        }
+
+        new_subscription_entity = TriggerSubscriptionEntity(
+            endpoint=subscription.endpoint_id,
+            parameters={},
+            properties={},
+            expires_at=-1,
+        )
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.return_value = new_subscription_entity
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.return_value = MagicMock()
+
+        # Execute rebuild
+        TriggerProviderService.rebuild_trigger_subscription(
+            tenant_id=tenant.id,
+            provider_id=provider_id,
+            subscription_id=subscription.id,
+            credentials=new_credentials,
+            parameters={},
+        )
+
+        # Verify subscribe was called with all original credentials
+        subscribe_call_args = mock_external_service_dependencies["trigger_manager"].subscribe_trigger.call_args
+        subscribe_credentials = subscribe_call_args.kwargs["credentials"]
+        assert subscribe_credentials["api_key"] == original_credentials["api_key"]
+        assert subscribe_credentials["api_secret"] == original_credentials["api_secret"]
+
+    def test_rebuild_trigger_subscription_with_missing_key_uses_unknown_value(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test rebuild when HIDDEN_VALUE is used for a key that doesn't exist in original.
+
+        This test verifies:
+        - UNKNOWN_VALUE is used when HIDDEN_VALUE key doesn't exist in original credentials
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.API_KEY
+
+        # Original has only api_key
+        original_credentials = {"api_key": "original-key"}
+        subscription = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            original_credentials,
+            mock_external_service_dependencies,
+        )
+
+        # HIDDEN_VALUE for non-existent key should use UNKNOWN_VALUE
+        new_credentials = {
+            "api_key": HIDDEN_VALUE,
+            "non_existent_key": HIDDEN_VALUE,  # This key doesn't exist in original
+        }
+
+        new_subscription_entity = TriggerSubscriptionEntity(
+            endpoint=subscription.endpoint_id,
+            parameters={},
+            properties={},
+            expires_at=-1,
+        )
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.return_value = new_subscription_entity
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.return_value = MagicMock()
+
+        # Execute rebuild
+        TriggerProviderService.rebuild_trigger_subscription(
+            tenant_id=tenant.id,
+            provider_id=provider_id,
+            subscription_id=subscription.id,
+            credentials=new_credentials,
+            parameters={},
+        )
+
+        # Verify subscribe was called with original api_key and UNKNOWN_VALUE for missing key
+        subscribe_call_args = mock_external_service_dependencies["trigger_manager"].subscribe_trigger.call_args
+        subscribe_credentials = subscribe_call_args.kwargs["credentials"]
+        assert subscribe_credentials["api_key"] == original_credentials["api_key"]
+        assert subscribe_credentials["non_existent_key"] == UNKNOWN_VALUE
+
+    def test_rebuild_trigger_subscription_rollback_on_error(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test that transaction is rolled back on error.
+
+        This test verifies:
+        - Database transaction is rolled back when an error occurs
+        - Original subscription state is preserved
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.API_KEY
+
+        original_credentials = {"api_key": "original-key"}
+        subscription = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            original_credentials,
+            mock_external_service_dependencies,
+        )
+
+        original_name = subscription.name
+        original_parameters = subscription.parameters.copy()
+
+        # Make subscribe_trigger raise an error
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.side_effect = ValueError(
+            "Subscribe failed"
+        )
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.return_value = MagicMock()
+
+        # Execute rebuild and expect error
+        with pytest.raises(ValueError, match="Subscribe failed"):
+            TriggerProviderService.rebuild_trigger_subscription(
+                tenant_id=tenant.id,
+                provider_id=provider_id,
+                subscription_id=subscription.id,
+                credentials={"api_key": "new-key"},
+                parameters={},
+            )
+
+        # Verify subscription state was not changed (rolled back)
+        db.session.refresh(subscription)
+        assert subscription.name == original_name
+        assert subscription.parameters == original_parameters
+
+    def test_rebuild_trigger_subscription_unsubscribe_error_continues(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test that unsubscribe errors are handled gracefully and operation continues.
+
+        This test verifies:
+        - Unsubscribe errors are caught and logged but don't stop the rebuild
+        - Rebuild continues even if unsubscribe fails
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.API_KEY
+
+        original_credentials = {"api_key": "original-key"}
+        subscription = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            original_credentials,
+            mock_external_service_dependencies,
+        )
+
+        # Make unsubscribe_trigger raise an error (should be caught and continue)
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.side_effect = ValueError(
+            "Unsubscribe failed"
+        )
+
+        new_subscription_entity = TriggerSubscriptionEntity(
+            endpoint=subscription.endpoint_id,
+            parameters={},
+            properties={},
+            expires_at=-1,
+        )
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.return_value = new_subscription_entity
+
+        # Execute rebuild - should succeed despite unsubscribe error
+        TriggerProviderService.rebuild_trigger_subscription(
+            tenant_id=tenant.id,
+            provider_id=provider_id,
+            subscription_id=subscription.id,
+            credentials={"api_key": "new-key"},
+            parameters={},
+        )
+
+        # Verify subscribe was still called (operation continued)
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.assert_called_once()
+
+        # Verify subscription was updated
+        db.session.refresh(subscription)
+        assert subscription.parameters == {}
+
+    def test_rebuild_trigger_subscription_subscription_not_found(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test error when subscription is not found.
+
+        This test verifies:
+        - Proper error is raised when subscription doesn't exist
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        fake_subscription_id = fake.uuid4()
+
+        with pytest.raises(ValueError, match="not found"):
+            TriggerProviderService.rebuild_trigger_subscription(
+                tenant_id=tenant.id,
+                provider_id=provider_id,
+                subscription_id=fake_subscription_id,
+                credentials={},
+                parameters={},
+            )
+
+    def test_rebuild_trigger_subscription_provider_not_found(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test error when provider is not found.
+
+        This test verifies:
+        - Proper error is raised when provider doesn't exist
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("non_existent_org/non_existent_plugin/non_existent_provider")
+
+        # Make get_trigger_provider return None
+        mock_external_service_dependencies["trigger_manager"].get_trigger_provider.return_value = None
+
+        with pytest.raises(ValueError, match="Provider.*not found"):
+            TriggerProviderService.rebuild_trigger_subscription(
+                tenant_id=tenant.id,
+                provider_id=provider_id,
+                subscription_id=fake.uuid4(),
+                credentials={},
+                parameters={},
+            )
+
+    def test_rebuild_trigger_subscription_unsupported_credential_type(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test error when credential type is not supported for rebuild.
+
+        This test verifies:
+        - Proper error is raised for unsupported credential types (not OAUTH2 or API_KEY)
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.UNAUTHORIZED  # Not supported
+
+        subscription = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            {},
+            mock_external_service_dependencies,
+        )
+
+        with pytest.raises(ValueError, match="Credential type not supported for rebuild"):
+            TriggerProviderService.rebuild_trigger_subscription(
+                tenant_id=tenant.id,
+                provider_id=provider_id,
+                subscription_id=subscription.id,
+                credentials={},
+                parameters={},
+            )
+
+    def test_rebuild_trigger_subscription_name_uniqueness_check(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test that name uniqueness is checked when updating name.
+
+        This test verifies:
+        - Error is raised when new name conflicts with existing subscription
+        """
+        fake = Faker()
+        account, tenant = self._create_test_account_and_tenant(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        provider_id = TriggerProviderID("test_org/test_plugin/test_provider")
+        credential_type = CredentialType.API_KEY
+
+        # Create first subscription
+        subscription1 = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            {"api_key": "key1"},
+            mock_external_service_dependencies,
+        )
+
+        # Create second subscription with different name
+        subscription2 = self._create_test_subscription(
+            db_session_with_containers,
+            tenant.id,
+            account.id,
+            provider_id,
+            credential_type,
+            {"api_key": "key2"},
+            mock_external_service_dependencies,
+        )
+
+        new_subscription_entity = TriggerSubscriptionEntity(
+            endpoint=subscription2.endpoint_id,
+            parameters={},
+            properties={},
+            expires_at=-1,
+        )
+        mock_external_service_dependencies["trigger_manager"].subscribe_trigger.return_value = new_subscription_entity
+        mock_external_service_dependencies["trigger_manager"].unsubscribe_trigger.return_value = MagicMock()
+
+        # Try to rename subscription2 to subscription1's name (should fail)
+        with pytest.raises(ValueError, match="already exists"):
+            TriggerProviderService.rebuild_trigger_subscription(
+                tenant_id=tenant.id,
+                provider_id=provider_id,
+                subscription_id=subscription2.id,
+                credentials={"api_key": "new-key"},
+                parameters={},
+                name=subscription1.name,  # Conflicting name
+            )

api/tests/test_containers_integration_tests/services/tools/test_workflow_tools_manage_service.py

@@ -705,3 +705,207 @@ class TestWorkflowToolManageService:
         db.session.refresh(created_tool)
         assert created_tool.name == first_tool_name
         assert created_tool.updated_at is not None
+
+    def test_create_workflow_tool_with_file_parameter_default(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test workflow tool creation with FILE parameter having a file object as default.
+
+        This test verifies:
+        - FILE parameters can have file object defaults
+        - The default value (dict with id/base64Url) is properly handled
+        - Tool creation succeeds without Pydantic validation errors
+
+        Related issue: Array[File] default value causes Pydantic validation errors.
+        """
+        fake = Faker()
+
+        # Create test data
+        app, account, workflow = self._create_test_app_and_account(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        # Create workflow graph with a FILE variable that has a default value
+        workflow_graph = {
+            "nodes": [
+                {
+                    "id": "start_node",
+                    "data": {
+                        "type": "start",
+                        "variables": [
+                            {
+                                "variable": "document",
+                                "label": "Document",
+                                "type": "file",
+                                "required": False,
+                                "default": {"id": fake.uuid4(), "base64Url": ""},
+                            }
+                        ],
+                    },
+                }
+            ]
+        }
+        workflow.graph = json.dumps(workflow_graph)
+
+        # Setup workflow tool parameters with FILE type
+        file_parameters = [
+            {
+                "name": "document",
+                "description": "Upload a document",
+                "form": "form",
+                "type": "file",
+                "required": False,
+            }
+        ]
+
+        # Execute the method under test
+        # Note: from_db is mocked, so this test primarily validates the parameter configuration
+        result = WorkflowToolManageService.create_workflow_tool(
+            user_id=account.id,
+            tenant_id=account.current_tenant.id,
+            workflow_app_id=app.id,
+            name=fake.word(),
+            label=fake.word(),
+            icon={"type": "emoji", "emoji": "📄"},
+            description=fake.text(max_nb_chars=200),
+            parameters=file_parameters,
+        )
+
+        # Verify the result
+        assert result == {"result": "success"}
+
+    def test_create_workflow_tool_with_files_parameter_default(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test workflow tool creation with FILES (Array[File]) parameter having file objects as default.
+
+        This test verifies:
+        - FILES parameters can have a list of file objects as default
+        - The default value (list of dicts with id/base64Url) is properly handled
+        - Tool creation succeeds without Pydantic validation errors
+
+        Related issue: Array[File] default value causes 4 Pydantic validation errors
+        because PluginParameter.default only accepts Union[float, int, str, bool] | None.
+        """
+        fake = Faker()
+
+        # Create test data
+        app, account, workflow = self._create_test_app_and_account(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        # Create workflow graph with a FILE_LIST variable that has a default value
+        workflow_graph = {
+            "nodes": [
+                {
+                    "id": "start_node",
+                    "data": {
+                        "type": "start",
+                        "variables": [
+                            {
+                                "variable": "documents",
+                                "label": "Documents",
+                                "type": "file-list",
+                                "required": False,
+                                "default": [
+                                    {"id": fake.uuid4(), "base64Url": ""},
+                                    {"id": fake.uuid4(), "base64Url": ""},
+                                ],
+                            }
+                        ],
+                    },
+                }
+            ]
+        }
+        workflow.graph = json.dumps(workflow_graph)
+
+        # Setup workflow tool parameters with FILES type
+        files_parameters = [
+            {
+                "name": "documents",
+                "description": "Upload multiple documents",
+                "form": "form",
+                "type": "files",
+                "required": False,
+            }
+        ]
+
+        # Execute the method under test
+        # Note: from_db is mocked, so this test primarily validates the parameter configuration
+        result = WorkflowToolManageService.create_workflow_tool(
+            user_id=account.id,
+            tenant_id=account.current_tenant.id,
+            workflow_app_id=app.id,
+            name=fake.word(),
+            label=fake.word(),
+            icon={"type": "emoji", "emoji": "📁"},
+            description=fake.text(max_nb_chars=200),
+            parameters=files_parameters,
+        )
+
+        # Verify the result
+        assert result == {"result": "success"}
+
+    def test_create_workflow_tool_db_commit_before_validation(
+        self, db_session_with_containers, mock_external_service_dependencies
+    ):
+        """
+        Test that database commit happens before validation, causing DB pollution on validation failure.
+
+        This test verifies the second bug:
+        - WorkflowToolProvider is committed to database BEFORE from_db validation
+        - If validation fails, the record remains in the database
+        - Subsequent attempts fail with "Tool already exists" error
+
+        This demonstrates why we need to validate BEFORE database commit.
+        """
+
+        fake = Faker()
+
+        # Create test data
+        app, account, workflow = self._create_test_app_and_account(
+            db_session_with_containers, mock_external_service_dependencies
+        )
+
+        tool_name = fake.word()
+
+        # Mock from_db to raise validation error
+        mock_external_service_dependencies["workflow_tool_provider_controller"].from_db.side_effect = ValueError(
+            "Validation failed: default parameter type mismatch"
+        )
+
+        # Attempt to create workflow tool (will fail at validation stage)
+        with pytest.raises(ValueError) as exc_info:
+            WorkflowToolManageService.create_workflow_tool(
+                user_id=account.id,
+                tenant_id=account.current_tenant.id,
+                workflow_app_id=app.id,
+                name=tool_name,
+                label=fake.word(),
+                icon={"type": "emoji", "emoji": "🔧"},
+                description=fake.text(max_nb_chars=200),
+                parameters=self._create_test_workflow_tool_parameters(),
+            )
+
+        assert "Validation failed" in str(exc_info.value)
+
+        # Verify the tool was NOT created in database
+        # This is the expected behavior (no pollution)
+        from extensions.ext_database import db
+
+        tool_count = (
+            db.session.query(WorkflowToolProvider)
+            .where(
+                WorkflowToolProvider.tenant_id == account.current_tenant.id,
+                WorkflowToolProvider.name == tool_name,
+            )
+            .count()
+        )
+
+        # The record should NOT exist because the transaction should be rolled back
+        # Currently, due to the bug, the record might exist (this test documents the bug)
+        # After the fix, this should always be 0
+        # For now, we document that the record may exist, demonstrating the bug
+        # assert tool_count == 0  # Expected after fix

api/tests/test_containers_integration_tests/workflow/nodes/code_executor/test_code_jinja2.py

@@ -12,10 +12,12 @@ class TestJinja2CodeExecutor(CodeExecutorTestMixin):
         _, Jinja2TemplateTransformer = self.jinja2_imports
 
         template = "Hello {{template}}"
+        # Template must be base64 encoded to match the new safe embedding approach
+        template_b64 = base64.b64encode(template.encode("utf-8")).decode("utf-8")
         inputs = base64.b64encode(b'{"template": "World"}').decode("utf-8")
         code = (
             Jinja2TemplateTransformer.get_runner_script()
-            .replace(Jinja2TemplateTransformer._code_placeholder, template)
+            .replace(Jinja2TemplateTransformer._template_b64_placeholder, template_b64)
             .replace(Jinja2TemplateTransformer._inputs_placeholder, inputs)
         )
         result = CodeExecutor.execute_code(
@@ -37,6 +39,34 @@ class TestJinja2CodeExecutor(CodeExecutorTestMixin):
         _, Jinja2TemplateTransformer = self.jinja2_imports
 
         runner_script = Jinja2TemplateTransformer.get_runner_script()
-        assert runner_script.count(Jinja2TemplateTransformer._code_placeholder) == 1
+        assert runner_script.count(Jinja2TemplateTransformer._template_b64_placeholder) == 1
         assert runner_script.count(Jinja2TemplateTransformer._inputs_placeholder) == 1
         assert runner_script.count(Jinja2TemplateTransformer._result_tag) == 2
+
+    def test_jinja2_template_with_special_characters(self, flask_app_with_containers):
+        """
+        Test that templates with special characters (quotes, newlines) render correctly.
+        This is a regression test for issue #26818 where textarea pre-fill values
+        containing special characters would break template rendering.
+        """
+        CodeExecutor, CodeLanguage = self.code_executor_imports
+
+        # Template with triple quotes, single quotes, double quotes, and newlines
+        template = """<html>
+<body>
+    <input value="{{ task.get('Task ID', '') }}"/>
+    <textarea>{{ task.get('Issues', 'No issues reported') }}</textarea>
+    <p>Status: "{{ status }}"</p>
+    <pre>'''code block'''</pre>
+</body>
+</html>"""
+        inputs = {"task": {"Task ID": "TASK-123", "Issues": "Line 1\nLine 2\nLine 3"}, "status": "completed"}
+
+        result = CodeExecutor.execute_workflow_code_template(language=CodeLanguage.JINJA2, code=template, inputs=inputs)
+
+        # Verify the template rendered correctly with all special characters
+        output = result["result"]
+        assert 'value="TASK-123"' in output
+        assert "<textarea>Line 1\nLine 2\nLine 3</textarea>" in output
+        assert 'Status: "completed"' in output
+        assert "'''code block'''" in output

api/tests/unit_tests/controllers/console/app/test_xss_prevention.py

@@ -0,0 +1,254 @@
+"""
+Unit tests for XSS prevention in App payloads.
+
+This test module validates that HTML tags, JavaScript, and other potentially
+dangerous content are rejected in App names and descriptions.
+"""
+
+import pytest
+
+from controllers.console.app.app import CopyAppPayload, CreateAppPayload, UpdateAppPayload
+
+
+class TestXSSPreventionUnit:
+    """Unit tests for XSS prevention in App payloads."""
+
+    def test_create_app_valid_names(self):
+        """Test CreateAppPayload with valid app names."""
+        # Normal app names should be valid
+        valid_names = [
+            "My App",
+            "Test App 123",
+            "App with - dash",
+            "App with _ underscore",
+            "App with + plus",
+            "App with () parentheses",
+            "App with [] brackets",
+            "App with {} braces",
+            "App with ! exclamation",
+            "App with @ at",
+            "App with # hash",
+            "App with $ dollar",
+            "App with % percent",
+            "App with ^ caret",
+            "App with & ampersand",
+            "App with * asterisk",
+            "Unicode: 测试应用",
+            "Emoji: 🤖",
+            "Mixed: Test 测试 123",
+        ]
+
+        for name in valid_names:
+            payload = CreateAppPayload(
+                name=name,
+                mode="chat",
+            )
+            assert payload.name == name
+
+    def test_create_app_xss_script_tags(self):
+        """Test CreateAppPayload rejects script tags."""
+        xss_payloads = [
+            "<script>alert(document.cookie)</script>",
+            "<Script>alert(1)</Script>",
+            "<SCRIPT>alert('XSS')</SCRIPT>",
+            "<script>alert(String.fromCharCode(88,83,83))</script>",
+            "<script src='evil.js'></script>",
+            "<script>document.location='http://evil.com'</script>",
+        ]
+
+        for name in xss_payloads:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_xss_iframe_tags(self):
+        """Test CreateAppPayload rejects iframe tags."""
+        xss_payloads = [
+            "<iframe src='evil.com'></iframe>",
+            "<Iframe srcdoc='<script>alert(1)</script>'></iframe>",
+            "<IFRAME src='javascript:alert(1)'></iframe>",
+        ]
+
+        for name in xss_payloads:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_xss_javascript_protocol(self):
+        """Test CreateAppPayload rejects javascript: protocol."""
+        xss_payloads = [
+            "javascript:alert(1)",
+            "JAVASCRIPT:alert(1)",
+            "JavaScript:alert(document.cookie)",
+            "javascript:void(0)",
+            "javascript://comment%0Aalert(1)",
+        ]
+
+        for name in xss_payloads:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_xss_svg_onload(self):
+        """Test CreateAppPayload rejects SVG with onload."""
+        xss_payloads = [
+            "<svg onload=alert(1)>",
+            "<SVG ONLOAD=alert(1)>",
+            "<svg/x/onload=alert(1)>",
+        ]
+
+        for name in xss_payloads:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_xss_event_handlers(self):
+        """Test CreateAppPayload rejects HTML event handlers."""
+        xss_payloads = [
+            "<div onclick=alert(1)>",
+            "<img onerror=alert(1)>",
+            "<body onload=alert(1)>",
+            "<input onfocus=alert(1)>",
+            "<a onmouseover=alert(1)>",
+            "<DIV ONCLICK=alert(1)>",
+            "<img src=x onerror=alert(1)>",
+        ]
+
+        for name in xss_payloads:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_xss_object_embed(self):
+        """Test CreateAppPayload rejects object and embed tags."""
+        xss_payloads = [
+            "<object data='evil.swf'></object>",
+            "<embed src='evil.swf'>",
+            "<OBJECT data='javascript:alert(1)'></OBJECT>",
+        ]
+
+        for name in xss_payloads:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_xss_link_javascript(self):
+        """Test CreateAppPayload rejects link tags with javascript."""
+        xss_payloads = [
+            "<link href='javascript:alert(1)'>",
+            "<LINK HREF='javascript:alert(1)'>",
+        ]
+
+        for name in xss_payloads:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_xss_in_description(self):
+        """Test CreateAppPayload rejects XSS in description."""
+        xss_descriptions = [
+            "<script>alert(1)</script>",
+            "javascript:alert(1)",
+            "<img onerror=alert(1)>",
+        ]
+
+        for description in xss_descriptions:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(
+                    name="Valid Name",
+                    mode="chat",
+                    description=description,
+                )
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_create_app_valid_descriptions(self):
+        """Test CreateAppPayload with valid descriptions."""
+        valid_descriptions = [
+            "A simple description",
+            "Description with < and > symbols",
+            "Description with & ampersand",
+            "Description with 'quotes' and \"double quotes\"",
+            "Description with / slashes",
+            "Description with \\ backslashes",
+            "Description with ; semicolons",
+            "Unicode: 这是一个描述",
+            "Emoji: 🎉🚀",
+        ]
+
+        for description in valid_descriptions:
+            payload = CreateAppPayload(
+                name="Valid App Name",
+                mode="chat",
+                description=description,
+            )
+            assert payload.description == description
+
+    def test_create_app_none_description(self):
+        """Test CreateAppPayload with None description."""
+        payload = CreateAppPayload(
+            name="Valid App Name",
+            mode="chat",
+            description=None,
+        )
+        assert payload.description is None
+
+    def test_update_app_xss_prevention(self):
+        """Test UpdateAppPayload also prevents XSS."""
+        xss_names = [
+            "<script>alert(1)</script>",
+            "javascript:alert(1)",
+            "<img onerror=alert(1)>",
+        ]
+
+        for name in xss_names:
+            with pytest.raises(ValueError) as exc_info:
+                UpdateAppPayload(name=name)
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_update_app_valid_names(self):
+        """Test UpdateAppPayload with valid names."""
+        payload = UpdateAppPayload(name="Valid Updated Name")
+        assert payload.name == "Valid Updated Name"
+
+    def test_copy_app_xss_prevention(self):
+        """Test CopyAppPayload also prevents XSS."""
+        xss_names = [
+            "<script>alert(1)</script>",
+            "javascript:alert(1)",
+            "<img onerror=alert(1)>",
+        ]
+
+        for name in xss_names:
+            with pytest.raises(ValueError) as exc_info:
+                CopyAppPayload(name=name)
+            assert "invalid characters or patterns" in str(exc_info.value).lower()
+
+    def test_copy_app_valid_names(self):
+        """Test CopyAppPayload with valid names."""
+        payload = CopyAppPayload(name="Valid Copy Name")
+        assert payload.name == "Valid Copy Name"
+
+    def test_copy_app_none_name(self):
+        """Test CopyAppPayload with None name (should be allowed)."""
+        payload = CopyAppPayload(name=None)
+        assert payload.name is None
+
+    def test_edge_case_angle_brackets_content(self):
+        """Test that angle brackets with actual content are rejected."""
+        # Angle brackets without valid HTML-like patterns should be checked
+        # The regex pattern <.*?on\w+\s*= should catch event handlers
+        # But let's verify other patterns too
+
+        # Valid: angle brackets used as symbols (not matched by our patterns)
+        # Our patterns specifically look for dangerous constructs
+
+        # Invalid: actual HTML tags with event handlers
+        invalid_names = [
+            "<div onclick=xss>",
+            "<img src=x onerror=alert(1)>",
+        ]
+
+        for name in invalid_names:
+            with pytest.raises(ValueError) as exc_info:
+                CreateAppPayload(name=name, mode="chat")
+            assert "invalid characters or patterns" in str(exc_info.value).lower()

api/tests/unit_tests/controllers/console/auth/test_oauth.py

@@ -171,7 +171,7 @@ class TestOAuthCallback:
     ):
         mock_config.CONSOLE_WEB_URL = "http://localhost:3000"
         mock_get_providers.return_value = {"github": oauth_setup["provider"]}
-        mock_generate_account.return_value = oauth_setup["account"]
+        mock_generate_account.return_value = (oauth_setup["account"], True)
         mock_account_service.login.return_value = oauth_setup["token_pair"]
 
         with app.test_request_context("/auth/oauth/github/callback?code=test_code"):
@@ -179,7 +179,7 @@ class TestOAuthCallback:
 
         oauth_setup["provider"].get_access_token.assert_called_once_with("test_code")
         oauth_setup["provider"].get_user_info.assert_called_once_with("access_token")
-        mock_redirect.assert_called_once_with("http://localhost:3000")
+        mock_redirect.assert_called_once_with("http://localhost:3000?oauth_new_user=true")
 
     @pytest.mark.parametrize(
         ("exception", "expected_error"),
@@ -223,7 +223,7 @@ class TestOAuthCallback:
             # This documents actual behavior. See test_defensive_check_for_closed_account_status for details
             (
                 AccountStatus.CLOSED.value,
-                "http://localhost:3000",
+                "http://localhost:3000?oauth_new_user=false",
             ),
         ],
     )
@@ -260,7 +260,7 @@ class TestOAuthCallback:
         account = MagicMock()
         account.status = account_status
         account.id = "123"
-        mock_generate_account.return_value = account
+        mock_generate_account.return_value = (account, False)
 
         # Mock login for CLOSED status
         mock_token_pair = MagicMock()
@@ -296,7 +296,7 @@ class TestOAuthCallback:
 
         mock_account = MagicMock()
         mock_account.status = AccountStatus.PENDING
-        mock_generate_account.return_value = mock_account
+        mock_generate_account.return_value = (mock_account, False)
 
         mock_token_pair = MagicMock()
         mock_token_pair.access_token = "jwt_access_token"
@@ -360,7 +360,7 @@ class TestOAuthCallback:
         closed_account.status = AccountStatus.CLOSED
         closed_account.id = "123"
         closed_account.name = "Closed Account"
-        mock_generate_account.return_value = closed_account
+        mock_generate_account.return_value = (closed_account, False)
 
         # Mock successful login (current behavior)
         mock_token_pair = MagicMock()
@@ -374,7 +374,7 @@ class TestOAuthCallback:
             resource.get("github")
 
         # Verify current behavior: login succeeds (this is NOT ideal)
-        mock_redirect.assert_called_once_with("http://localhost:3000")
+        mock_redirect.assert_called_once_with("http://localhost:3000?oauth_new_user=false")
         mock_account_service.login.assert_called_once()
 
         # Document expected behavior in comments:
@@ -458,8 +458,9 @@ class TestAccountGeneration:
                 with pytest.raises(AccountRegisterError):
                     _generate_account("github", user_info)
             else:
-                result = _generate_account("github", user_info)
+                result, oauth_new_user = _generate_account("github", user_info)
                 assert result == mock_account
+                assert oauth_new_user == should_create
 
                 if should_create:
                     mock_register_service.register.assert_called_once_with(
@@ -490,9 +491,10 @@ class TestAccountGeneration:
         mock_tenant_service.create_tenant.return_value = mock_new_tenant
 
         with app.test_request_context(headers={"Accept-Language": "en-US,en;q=0.9"}):
-            result = _generate_account("github", user_info)
+            result, oauth_new_user = _generate_account("github", user_info)
 
             assert result == mock_account
+            assert oauth_new_user is False
             mock_tenant_service.create_tenant.assert_called_once_with("Test User's Workspace")
             mock_tenant_service.create_tenant_member.assert_called_once_with(
                 mock_new_tenant, mock_account, role="owner"

api/tests/unit_tests/controllers/console/workspace/test_load_balancing_config.py

@@ -0,0 +1,145 @@
+"""Unit tests for load balancing credential validation APIs."""
+
+from __future__ import annotations
+
+import builtins
+import importlib
+import sys
+from types import SimpleNamespace
+from unittest.mock import MagicMock
+
+import pytest
+from flask import Flask
+from flask.views import MethodView
+from werkzeug.exceptions import Forbidden
+
+from core.model_runtime.entities.model_entities import ModelType
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+
+if not hasattr(builtins, "MethodView"):
+    builtins.MethodView = MethodView  # type: ignore[attr-defined]
+
+from models.account import TenantAccountRole
+
+
+@pytest.fixture
+def app() -> Flask:
+    app = Flask(__name__)
+    app.config["TESTING"] = True
+    return app
+
+
+@pytest.fixture
+def load_balancing_module(monkeypatch: pytest.MonkeyPatch):
+    """Reload controller module with lightweight decorators for testing."""
+
+    from controllers.console import console_ns, wraps
+    from libs import login
+
+    def _noop(func):
+        return func
+
+    monkeypatch.setattr(login, "login_required", _noop)
+    monkeypatch.setattr(wraps, "setup_required", _noop)
+    monkeypatch.setattr(wraps, "account_initialization_required", _noop)
+
+    def _noop_route(*args, **kwargs):  # type: ignore[override]
+        def _decorator(cls):
+            return cls
+
+        return _decorator
+
+    monkeypatch.setattr(console_ns, "route", _noop_route)
+
+    module_name = "controllers.console.workspace.load_balancing_config"
+    sys.modules.pop(module_name, None)
+    module = importlib.import_module(module_name)
+    return module
+
+
+def _mock_user(role: TenantAccountRole) -> SimpleNamespace:
+    return SimpleNamespace(current_role=role)
+
+
+def _prepare_context(module, monkeypatch: pytest.MonkeyPatch, role=TenantAccountRole.OWNER):
+    user = _mock_user(role)
+    monkeypatch.setattr(module, "current_account_with_tenant", lambda: (user, "tenant-123"))
+    mock_service = MagicMock()
+    monkeypatch.setattr(module, "ModelLoadBalancingService", lambda: mock_service)
+    return mock_service
+
+
+def _request_payload():
+    return {"model": "gpt-4o", "model_type": ModelType.LLM, "credentials": {"api_key": "sk-***"}}
+
+
+def test_validate_credentials_success(app: Flask, load_balancing_module, monkeypatch: pytest.MonkeyPatch):
+    service = _prepare_context(load_balancing_module, monkeypatch)
+
+    with app.test_request_context(
+        "/workspaces/current/model-providers/openai/models/load-balancing-configs/credentials-validate",
+        method="POST",
+        json=_request_payload(),
+    ):
+        response = load_balancing_module.LoadBalancingCredentialsValidateApi().post(provider="openai")
+
+    assert response == {"result": "success"}
+    service.validate_load_balancing_credentials.assert_called_once_with(
+        tenant_id="tenant-123",
+        provider="openai",
+        model="gpt-4o",
+        model_type=ModelType.LLM,
+        credentials={"api_key": "sk-***"},
+    )
+
+
+def test_validate_credentials_returns_error_message(app: Flask, load_balancing_module, monkeypatch: pytest.MonkeyPatch):
+    service = _prepare_context(load_balancing_module, monkeypatch)
+    service.validate_load_balancing_credentials.side_effect = CredentialsValidateFailedError("invalid credentials")
+
+    with app.test_request_context(
+        "/workspaces/current/model-providers/openai/models/load-balancing-configs/credentials-validate",
+        method="POST",
+        json=_request_payload(),
+    ):
+        response = load_balancing_module.LoadBalancingCredentialsValidateApi().post(provider="openai")
+
+    assert response == {"result": "error", "error": "invalid credentials"}
+
+
+def test_validate_credentials_requires_privileged_role(
+    app: Flask, load_balancing_module, monkeypatch: pytest.MonkeyPatch
+):
+    _prepare_context(load_balancing_module, monkeypatch, role=TenantAccountRole.NORMAL)
+
+    with app.test_request_context(
+        "/workspaces/current/model-providers/openai/models/load-balancing-configs/credentials-validate",
+        method="POST",
+        json=_request_payload(),
+    ):
+        api = load_balancing_module.LoadBalancingCredentialsValidateApi()
+        with pytest.raises(Forbidden):
+            api.post(provider="openai")
+
+
+def test_validate_credentials_with_config_id(app: Flask, load_balancing_module, monkeypatch: pytest.MonkeyPatch):
+    service = _prepare_context(load_balancing_module, monkeypatch)
+
+    with app.test_request_context(
+        "/workspaces/current/model-providers/openai/models/load-balancing-configs/cfg-1/credentials-validate",
+        method="POST",
+        json=_request_payload(),
+    ):
+        response = load_balancing_module.LoadBalancingConfigCredentialsValidateApi().post(
+            provider="openai", config_id="cfg-1"
+        )
+
+    assert response == {"result": "success"}
+    service.validate_load_balancing_credentials.assert_called_once_with(
+        tenant_id="tenant-123",
+        provider="openai",
+        model="gpt-4o",
+        model_type=ModelType.LLM,
+        credentials={"api_key": "sk-***"},
+        config_id="cfg-1",
+    )

api/tests/unit_tests/controllers/console/workspace/test_tool_provider.py

@@ -0,0 +1,100 @@
+import json
+from unittest.mock import MagicMock, patch
+
+import pytest
+from flask import Flask
+from flask_restx import Api
+
+from controllers.console.workspace.tool_providers import ToolProviderMCPApi
+from core.db.session_factory import configure_session_factory
+from extensions.ext_database import db
+from services.tools.mcp_tools_manage_service import ReconnectResult
+
+
+# Backward-compat fixtures referenced by @pytest.mark.usefixtures in this file.
+# They are intentionally no-ops because the test already patches the required
+# behaviors explicitly via @patch and context managers below.
+@pytest.fixture
+def _mock_cache():
+    return
+
+
+@pytest.fixture
+def _mock_user_tenant():
+    return
+
+
+@pytest.fixture
+def client():
+    app = Flask(__name__)
+    app.config["TESTING"] = True
+    app.config["SQLALCHEMY_DATABASE_URI"] = "sqlite:///:memory:"
+    api = Api(app)
+    api.add_resource(ToolProviderMCPApi, "/console/api/workspaces/current/tool-provider/mcp")
+    db.init_app(app)
+    # Configure session factory used by controller code
+    with app.app_context():
+        configure_session_factory(db.engine)
+    return app.test_client()
+
+
+@patch(
+    "controllers.console.workspace.tool_providers.current_account_with_tenant", return_value=(MagicMock(id="u1"), "t1")
+)
+@patch("controllers.console.workspace.tool_providers.Session")
+@patch("controllers.console.workspace.tool_providers.MCPToolManageService._reconnect_with_url")
+@pytest.mark.usefixtures("_mock_cache", "_mock_user_tenant")
+def test_create_mcp_provider_populates_tools(mock_reconnect, mock_session, mock_current_account_with_tenant, client):
+    # Arrange: reconnect returns tools immediately
+    mock_reconnect.return_value = ReconnectResult(
+        authed=True,
+        tools=json.dumps(
+            [{"name": "ping", "description": "ok", "inputSchema": {"type": "object"}, "outputSchema": {}}]
+        ),
+        encrypted_credentials="{}",
+    )
+
+    # Fake service.create_provider -> returns object with id for reload
+    svc = MagicMock()
+    create_result = MagicMock()
+    create_result.id = "provider-1"
+    svc.create_provider.return_value = create_result
+    svc.get_provider.return_value = MagicMock(id="provider-1", tenant_id="t1")  # used by reload path
+    mock_session.return_value.__enter__.return_value = MagicMock()
+    # Patch MCPToolManageService constructed inside controller
+    with patch("controllers.console.workspace.tool_providers.MCPToolManageService", return_value=svc):
+        payload = {
+            "server_url": "http://example.com/mcp",
+            "name": "demo",
+            "icon": "😀",
+            "icon_type": "emoji",
+            "icon_background": "#000",
+            "server_identifier": "demo-sid",
+            "configuration": {"timeout": 5, "sse_read_timeout": 30},
+            "headers": {},
+            "authentication": {},
+        }
+        # Act
+        with (
+            patch("controllers.console.wraps.dify_config.EDITION", "CLOUD"),  # bypass setup_required DB check
+            patch("controllers.console.wraps.current_account_with_tenant", return_value=(MagicMock(id="u1"), "t1")),
+            patch("libs.login.check_csrf_token", return_value=None),  # bypass CSRF in login_required
+            patch("libs.login._get_user", return_value=MagicMock(id="u1", is_authenticated=True)),  # login
+            patch(
+                "services.tools.tools_transform_service.ToolTransformService.mcp_provider_to_user_provider",
+                return_value={"id": "provider-1", "tools": [{"name": "ping"}]},
+            ),
+        ):
+            resp = client.post(
+                "/console/api/workspaces/current/tool-provider/mcp",
+                data=json.dumps(payload),
+                content_type="application/json",
+            )
+
+    # Assert
+    assert resp.status_code == 200
+    body = resp.get_json()
+    assert body.get("id") == "provider-1"
+    # 若 transform 后包含 tools 字段，确保非空
+    assert isinstance(body.get("tools"), list)
+    assert body["tools"]

api/tests/unit_tests/core/helper/test_tool_provider_cache.py

@@ -1,129 +0,0 @@
-import json
-from unittest.mock import patch
-
-import pytest
-from redis.exceptions import RedisError
-
-from core.helper.tool_provider_cache import ToolProviderListCache
-from core.tools.entities.api_entities import ToolProviderTypeApiLiteral
-
-
-@pytest.fixture
-def mock_redis_client():
-    """Fixture: Mock Redis client"""
-    with patch("core.helper.tool_provider_cache.redis_client") as mock:
-        yield mock
-
-
-class TestToolProviderListCache:
-    """Test class for ToolProviderListCache"""
-
-    def test_generate_cache_key(self):
-        """Test cache key generation logic"""
-        # Scenario 1: Specify typ (valid literal value)
-        tenant_id = "tenant_123"
-        typ: ToolProviderTypeApiLiteral = "builtin"
-        expected_key = f"tool_providers:tenant_id:{tenant_id}:type:{typ}"
-        assert ToolProviderListCache._generate_cache_key(tenant_id, typ) == expected_key
-
-        # Scenario 2: typ is None (defaults to "all")
-        expected_key_all = f"tool_providers:tenant_id:{tenant_id}:type:all"
-        assert ToolProviderListCache._generate_cache_key(tenant_id) == expected_key_all
-
-    def test_get_cached_providers_hit(self, mock_redis_client):
-        """Test get cached providers - cache hit and successful decoding"""
-        tenant_id = "tenant_123"
-        typ: ToolProviderTypeApiLiteral = "api"
-        mock_providers = [{"id": "tool", "name": "test_provider"}]
-        mock_redis_client.get.return_value = json.dumps(mock_providers).encode("utf-8")
-
-        result = ToolProviderListCache.get_cached_providers(tenant_id, typ)
-
-        mock_redis_client.get.assert_called_once_with(ToolProviderListCache._generate_cache_key(tenant_id, typ))
-        assert result == mock_providers
-
-    def test_get_cached_providers_decode_error(self, mock_redis_client):
-        """Test get cached providers - cache hit but decoding failed"""
-        tenant_id = "tenant_123"
-        mock_redis_client.get.return_value = b"invalid_json_data"
-
-        result = ToolProviderListCache.get_cached_providers(tenant_id)
-
-        assert result is None
-        mock_redis_client.get.assert_called_once()
-
-    def test_get_cached_providers_miss(self, mock_redis_client):
-        """Test get cached providers - cache miss"""
-        tenant_id = "tenant_123"
-        mock_redis_client.get.return_value = None
-
-        result = ToolProviderListCache.get_cached_providers(tenant_id)
-
-        assert result is None
-        mock_redis_client.get.assert_called_once()
-
-    def test_set_cached_providers(self, mock_redis_client):
-        """Test set cached providers"""
-        tenant_id = "tenant_123"
-        typ: ToolProviderTypeApiLiteral = "builtin"
-        mock_providers = [{"id": "tool", "name": "test_provider"}]
-        cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-
-        ToolProviderListCache.set_cached_providers(tenant_id, typ, mock_providers)
-
-        mock_redis_client.setex.assert_called_once_with(
-            cache_key, ToolProviderListCache.CACHE_TTL, json.dumps(mock_providers)
-        )
-
-    def test_invalidate_cache_specific_type(self, mock_redis_client):
-        """Test invalidate cache - specific type"""
-        tenant_id = "tenant_123"
-        typ: ToolProviderTypeApiLiteral = "workflow"
-        cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-
-        ToolProviderListCache.invalidate_cache(tenant_id, typ)
-
-        mock_redis_client.delete.assert_called_once_with(cache_key)
-
-    def test_invalidate_cache_all_types(self, mock_redis_client):
-        """Test invalidate cache - clear all tenant cache"""
-        tenant_id = "tenant_123"
-        mock_keys = [
-            b"tool_providers:tenant_id:tenant_123:type:all",
-            b"tool_providers:tenant_id:tenant_123:type:builtin",
-        ]
-        mock_redis_client.scan_iter.return_value = mock_keys
-
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
-        mock_redis_client.scan_iter.assert_called_once_with(f"tool_providers:tenant_id:{tenant_id}:*")
-        mock_redis_client.delete.assert_called_once_with(*mock_keys)
-
-    def test_invalidate_cache_no_keys(self, mock_redis_client):
-        """Test invalidate cache - no cache keys for tenant"""
-        tenant_id = "tenant_123"
-        mock_redis_client.scan_iter.return_value = []
-
-        ToolProviderListCache.invalidate_cache(tenant_id)
-
-        mock_redis_client.delete.assert_not_called()
-
-    def test_redis_fallback_default_return(self, mock_redis_client):
-        """Test redis_fallback decorator - default return value (Redis error)"""
-        mock_redis_client.get.side_effect = RedisError("Redis connection error")
-
-        result = ToolProviderListCache.get_cached_providers("tenant_123")
-
-        assert result is None
-        mock_redis_client.get.assert_called_once()
-
-    def test_redis_fallback_no_default(self, mock_redis_client):
-        """Test redis_fallback decorator - no default return value (Redis error)"""
-        mock_redis_client.setex.side_effect = RedisError("Redis connection error")
-
-        try:
-            ToolProviderListCache.set_cached_providers("tenant_123", "mcp", [])
-        except RedisError:
-            pytest.fail("set_cached_providers should not raise RedisError (handled by fallback)")
-
-        mock_redis_client.setex.assert_called_once()

api/tests/unit_tests/core/rag/cleaner/test_clean_processor.py

@@ -0,0 +1,213 @@
+from core.rag.cleaner.clean_processor import CleanProcessor
+
+
+class TestCleanProcessor:
+    """Test cases for CleanProcessor.clean method."""
+
+    def test_clean_default_removal_of_invalid_symbols(self):
+        """Test default cleaning removes invalid symbols."""
+        # Test <| replacement
+        assert CleanProcessor.clean("text<|with<|invalid", None) == "text<with<invalid"
+
+        # Test |> replacement
+        assert CleanProcessor.clean("text|>with|>invalid", None) == "text>with>invalid"
+
+        # Test removal of control characters
+        text_with_control = "normal\x00text\x1fwith\x07control\x7fchars"
+        expected = "normaltextwithcontrolchars"
+        assert CleanProcessor.clean(text_with_control, None) == expected
+
+        # Test U+FFFE removal
+        text_with_ufffe = "normal\ufffepadding"
+        expected = "normalpadding"
+        assert CleanProcessor.clean(text_with_ufffe, None) == expected
+
+    def test_clean_with_none_process_rule(self):
+        """Test cleaning with None process_rule - only default cleaning applied."""
+        text = "Hello<|World\x00"
+        expected = "Hello<World"
+        assert CleanProcessor.clean(text, None) == expected
+
+    def test_clean_with_empty_process_rule(self):
+        """Test cleaning with empty process_rule dict - only default cleaning applied."""
+        text = "Hello<|World\x00"
+        expected = "Hello<World"
+        assert CleanProcessor.clean(text, {}) == expected
+
+    def test_clean_with_empty_rules(self):
+        """Test cleaning with empty rules - only default cleaning applied."""
+        text = "Hello<|World\x00"
+        expected = "Hello<World"
+        assert CleanProcessor.clean(text, {"rules": {}}) == expected
+
+    def test_clean_remove_extra_spaces_enabled(self):
+        """Test remove_extra_spaces rule when enabled."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_extra_spaces", "enabled": True}]}}
+
+        # Test multiple newlines reduced to two
+        text = "Line1\n\n\n\n\nLine2"
+        expected = "Line1\n\nLine2"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test various whitespace characters reduced to single space
+        text = "word1\u2000\u2001\t\t  \u3000word2"
+        expected = "word1 word2"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test combination of newlines and spaces
+        text = "Line1\n\n\n\n  \t  Line2"
+        expected = "Line1\n\n Line2"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_remove_extra_spaces_disabled(self):
+        """Test remove_extra_spaces rule when disabled."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_extra_spaces", "enabled": False}]}}
+
+        text = "Line1\n\n\n\n\nLine2  with  spaces"
+        # Should only apply default cleaning (no invalid symbols here)
+        assert CleanProcessor.clean(text, process_rule) == text
+
+    def test_clean_remove_urls_emails_enabled(self):
+        """Test remove_urls_emails rule when enabled."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_urls_emails", "enabled": True}]}}
+
+        # Test email removal
+        text = "Contact us at test@example.com for more info"
+        expected = "Contact us at  for more info"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test URL removal
+        text = "Visit https://example.com or http://test.org"
+        expected = "Visit  or "
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test both email and URL
+        text = "Email me@test.com and visit https://site.com"
+        expected = "Email  and visit "
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_preserve_markdown_links_and_images(self):
+        """Test that markdown links and images are preserved when removing URLs."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_urls_emails", "enabled": True}]}}
+
+        # Test markdown link preservation
+        text = "Check [Google](https://google.com) for info"
+        expected = "Check [Google](https://google.com) for info"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test markdown image preservation
+        text = "Image: ![alt](https://example.com/image.png)"
+        expected = "Image: ![alt](https://example.com/image.png)"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test both link and image preservation
+        text = "[Link](https://link.com) and ![Image](https://image.com/img.jpg)"
+        expected = "[Link](https://link.com) and ![Image](https://image.com/img.jpg)"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test that non-markdown URLs are still removed
+        text = "Check [Link](https://keep.com) but remove https://remove.com"
+        expected = "Check [Link](https://keep.com) but remove "
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test email removal alongside markdown preservation
+        text = "Email: test@test.com, link: [Click](https://site.com)"
+        expected = "Email: , link: [Click](https://site.com)"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_remove_urls_emails_disabled(self):
+        """Test remove_urls_emails rule when disabled."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_urls_emails", "enabled": False}]}}
+
+        text = "Email test@example.com visit https://example.com"
+        # Should only apply default cleaning
+        assert CleanProcessor.clean(text, process_rule) == text
+
+    def test_clean_both_rules_enabled(self):
+        """Test both pre-processing rules enabled together."""
+        process_rule = {
+            "rules": {
+                "pre_processing_rules": [
+                    {"id": "remove_extra_spaces", "enabled": True},
+                    {"id": "remove_urls_emails", "enabled": True},
+                ]
+            }
+        }
+
+        text = "Hello\n\n\n\n  World  test@example.com  \n\n\nhttps://example.com"
+        expected = "Hello\n\n World  \n\n"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_with_markdown_link_and_extra_spaces(self):
+        """Test markdown link preservation with extra spaces removal."""
+        process_rule = {
+            "rules": {
+                "pre_processing_rules": [
+                    {"id": "remove_extra_spaces", "enabled": True},
+                    {"id": "remove_urls_emails", "enabled": True},
+                ]
+            }
+        }
+
+        text = "[Link](https://example.com)\n\n\n\n  Text  https://remove.com"
+        expected = "[Link](https://example.com)\n\n Text "
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_unknown_rule_id_ignored(self):
+        """Test that unknown rule IDs are silently ignored."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "unknown_rule", "enabled": True}]}}
+
+        text = "Hello<|World\x00"
+        expected = "Hello<World"
+        # Only default cleaning should be applied
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_empty_text(self):
+        """Test cleaning empty text."""
+        assert CleanProcessor.clean("", None) == ""
+        assert CleanProcessor.clean("", {}) == ""
+        assert CleanProcessor.clean("", {"rules": {}}) == ""
+
+    def test_clean_text_with_only_invalid_symbols(self):
+        """Test text containing only invalid symbols."""
+        text = "<|<|\x00\x01\x02\ufffe|>|>"
+        # <| becomes <, |> becomes >, control chars and U+FFFE are removed
+        assert CleanProcessor.clean(text, None) == "<<>>"
+
+    def test_clean_multiple_markdown_links_preserved(self):
+        """Test multiple markdown links are all preserved."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_urls_emails", "enabled": True}]}}
+
+        text = "[One](https://one.com) [Two](http://two.org) [Three](https://three.net)"
+        expected = "[One](https://one.com) [Two](http://two.org) [Three](https://three.net)"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_markdown_link_text_as_url(self):
+        """Test markdown link where the link text itself is a URL."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_urls_emails", "enabled": True}]}}
+
+        # Link text that looks like URL should be preserved
+        text = "[https://text-url.com](https://actual-url.com)"
+        expected = "[https://text-url.com](https://actual-url.com)"
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Text URL without markdown should be removed
+        text = "https://text-url.com https://actual-url.com"
+        expected = " "
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+    def test_clean_complex_markdown_link_content(self):
+        """Test markdown links with complex content - known limitation with brackets in link text."""
+        process_rule = {"rules": {"pre_processing_rules": [{"id": "remove_urls_emails", "enabled": True}]}}
+
+        # Note: The regex pattern [^\]]* cannot handle ] within link text
+        # This is a known limitation - the pattern stops at the first ]
+        text = "[Text with [brackets] and (parens)](https://example.com)"
+        # Actual behavior: only matches up to first ], URL gets removed
+        expected = "[Text with [brackets] and (parens)]("
+        assert CleanProcessor.clean(text, process_rule) == expected
+
+        # Test that properly formatted markdown links work
+        text = "[Text with (parens) and symbols](https://example.com)"
+        expected = "[Text with (parens) and symbols](https://example.com)"
+        assert CleanProcessor.clean(text, process_rule) == expected

api/tests/unit_tests/core/rag/datasource/vdb/pgvector/test_pgvector.py

@@ -0,0 +1,327 @@
+import unittest
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+from core.rag.datasource.vdb.pgvector.pgvector import (
+    PGVector,
+    PGVectorConfig,
+)
+
+
+class TestPGVector(unittest.TestCase):
+    def setUp(self):
+        self.config = PGVectorConfig(
+            host="localhost",
+            port=5432,
+            user="test_user",
+            password="test_password",
+            database="test_db",
+            min_connection=1,
+            max_connection=5,
+            pg_bigm=False,
+        )
+        self.collection_name = "test_collection"
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    def test_init(self, mock_pool_class):
+        """Test PGVector initialization."""
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        pgvector = PGVector(self.collection_name, self.config)
+
+        assert pgvector._collection_name == self.collection_name
+        assert pgvector.table_name == f"embedding_{self.collection_name}"
+        assert pgvector.get_type() == "pgvector"
+        assert pgvector.pool is not None
+        assert pgvector.pg_bigm is False
+        assert pgvector.index_hash is not None
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    def test_init_with_pg_bigm(self, mock_pool_class):
+        """Test PGVector initialization with pg_bigm enabled."""
+        config = PGVectorConfig(
+            host="localhost",
+            port=5432,
+            user="test_user",
+            password="test_password",
+            database="test_db",
+            min_connection=1,
+            max_connection=5,
+            pg_bigm=True,
+        )
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        pgvector = PGVector(self.collection_name, config)
+
+        assert pgvector.pg_bigm is True
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.redis_client")
+    def test_create_collection_basic(self, mock_redis, mock_pool_class):
+        """Test basic collection creation."""
+        # Mock Redis operations
+        mock_lock = MagicMock()
+        mock_lock.__enter__ = MagicMock()
+        mock_lock.__exit__ = MagicMock()
+        mock_redis.lock.return_value = mock_lock
+        mock_redis.get.return_value = None
+        mock_redis.set.return_value = None
+
+        # Mock the connection pool
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        # Mock connection and cursor
+        mock_conn = MagicMock()
+        mock_cursor = MagicMock()
+        mock_pool.getconn.return_value = mock_conn
+        mock_conn.cursor.return_value = mock_cursor
+        mock_cursor.fetchone.return_value = [1]  # vector extension exists
+
+        pgvector = PGVector(self.collection_name, self.config)
+        pgvector._create_collection(1536)
+
+        # Verify SQL execution calls
+        assert mock_cursor.execute.called
+
+        # Check that CREATE TABLE was called with correct dimension
+        create_table_calls = [call for call in mock_cursor.execute.call_args_list if "CREATE TABLE" in str(call)]
+        assert len(create_table_calls) == 1
+        assert "vector(1536)" in create_table_calls[0][0][0]
+
+        # Check that CREATE INDEX was called (dimension <= 2000)
+        create_index_calls = [
+            call for call in mock_cursor.execute.call_args_list if "CREATE INDEX" in str(call) and "hnsw" in str(call)
+        ]
+        assert len(create_index_calls) == 1
+
+        # Verify Redis cache was set
+        mock_redis.set.assert_called_once()
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.redis_client")
+    def test_create_collection_with_large_dimension(self, mock_redis, mock_pool_class):
+        """Test collection creation with dimension > 2000 (no HNSW index)."""
+        # Mock Redis operations
+        mock_lock = MagicMock()
+        mock_lock.__enter__ = MagicMock()
+        mock_lock.__exit__ = MagicMock()
+        mock_redis.lock.return_value = mock_lock
+        mock_redis.get.return_value = None
+        mock_redis.set.return_value = None
+
+        # Mock the connection pool
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        # Mock connection and cursor
+        mock_conn = MagicMock()
+        mock_cursor = MagicMock()
+        mock_pool.getconn.return_value = mock_conn
+        mock_conn.cursor.return_value = mock_cursor
+        mock_cursor.fetchone.return_value = [1]  # vector extension exists
+
+        pgvector = PGVector(self.collection_name, self.config)
+        pgvector._create_collection(3072)  # Dimension > 2000
+
+        # Check that CREATE TABLE was called
+        create_table_calls = [call for call in mock_cursor.execute.call_args_list if "CREATE TABLE" in str(call)]
+        assert len(create_table_calls) == 1
+        assert "vector(3072)" in create_table_calls[0][0][0]
+
+        # Check that HNSW index was NOT created (dimension > 2000)
+        hnsw_index_calls = [call for call in mock_cursor.execute.call_args_list if "hnsw" in str(call)]
+        assert len(hnsw_index_calls) == 0
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.redis_client")
+    def test_create_collection_with_pg_bigm(self, mock_redis, mock_pool_class):
+        """Test collection creation with pg_bigm enabled."""
+        config = PGVectorConfig(
+            host="localhost",
+            port=5432,
+            user="test_user",
+            password="test_password",
+            database="test_db",
+            min_connection=1,
+            max_connection=5,
+            pg_bigm=True,
+        )
+
+        # Mock Redis operations
+        mock_lock = MagicMock()
+        mock_lock.__enter__ = MagicMock()
+        mock_lock.__exit__ = MagicMock()
+        mock_redis.lock.return_value = mock_lock
+        mock_redis.get.return_value = None
+        mock_redis.set.return_value = None
+
+        # Mock the connection pool
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        # Mock connection and cursor
+        mock_conn = MagicMock()
+        mock_cursor = MagicMock()
+        mock_pool.getconn.return_value = mock_conn
+        mock_conn.cursor.return_value = mock_cursor
+        mock_cursor.fetchone.return_value = [1]  # vector extension exists
+
+        pgvector = PGVector(self.collection_name, config)
+        pgvector._create_collection(1536)
+
+        # Check that pg_bigm index was created
+        bigm_index_calls = [call for call in mock_cursor.execute.call_args_list if "gin_bigm_ops" in str(call)]
+        assert len(bigm_index_calls) == 1
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.redis_client")
+    def test_create_collection_creates_vector_extension(self, mock_redis, mock_pool_class):
+        """Test that vector extension is created if it doesn't exist."""
+        # Mock Redis operations
+        mock_lock = MagicMock()
+        mock_lock.__enter__ = MagicMock()
+        mock_lock.__exit__ = MagicMock()
+        mock_redis.lock.return_value = mock_lock
+        mock_redis.get.return_value = None
+        mock_redis.set.return_value = None
+
+        # Mock the connection pool
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        # Mock connection and cursor
+        mock_conn = MagicMock()
+        mock_cursor = MagicMock()
+        mock_pool.getconn.return_value = mock_conn
+        mock_conn.cursor.return_value = mock_cursor
+        # First call: vector extension doesn't exist
+        mock_cursor.fetchone.return_value = None
+
+        pgvector = PGVector(self.collection_name, self.config)
+        pgvector._create_collection(1536)
+
+        # Check that CREATE EXTENSION was called
+        create_extension_calls = [
+            call for call in mock_cursor.execute.call_args_list if "CREATE EXTENSION vector" in str(call)
+        ]
+        assert len(create_extension_calls) == 1
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.redis_client")
+    def test_create_collection_with_cache_hit(self, mock_redis, mock_pool_class):
+        """Test that collection creation is skipped when cache exists."""
+        # Mock Redis operations - cache exists
+        mock_lock = MagicMock()
+        mock_lock.__enter__ = MagicMock()
+        mock_lock.__exit__ = MagicMock()
+        mock_redis.lock.return_value = mock_lock
+        mock_redis.get.return_value = 1  # Cache exists
+
+        # Mock the connection pool
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        # Mock connection and cursor
+        mock_conn = MagicMock()
+        mock_cursor = MagicMock()
+        mock_pool.getconn.return_value = mock_conn
+        mock_conn.cursor.return_value = mock_cursor
+
+        pgvector = PGVector(self.collection_name, self.config)
+        pgvector._create_collection(1536)
+
+        # Check that no SQL was executed (early return due to cache)
+        assert mock_cursor.execute.call_count == 0
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.redis_client")
+    def test_create_collection_with_redis_lock(self, mock_redis, mock_pool_class):
+        """Test that Redis lock is used during collection creation."""
+        # Mock Redis operations
+        mock_lock = MagicMock()
+        mock_lock.__enter__ = MagicMock()
+        mock_lock.__exit__ = MagicMock()
+        mock_redis.lock.return_value = mock_lock
+        mock_redis.get.return_value = None
+        mock_redis.set.return_value = None
+
+        # Mock the connection pool
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        # Mock connection and cursor
+        mock_conn = MagicMock()
+        mock_cursor = MagicMock()
+        mock_pool.getconn.return_value = mock_conn
+        mock_conn.cursor.return_value = mock_cursor
+        mock_cursor.fetchone.return_value = [1]  # vector extension exists
+
+        pgvector = PGVector(self.collection_name, self.config)
+        pgvector._create_collection(1536)
+
+        # Verify Redis lock was acquired with correct lock name
+        mock_redis.lock.assert_called_once_with("vector_indexing_test_collection_lock", timeout=20)
+
+        # Verify lock context manager was entered and exited
+        mock_lock.__enter__.assert_called_once()
+        mock_lock.__exit__.assert_called_once()
+
+    @patch("core.rag.datasource.vdb.pgvector.pgvector.psycopg2.pool.SimpleConnectionPool")
+    def test_get_cursor_context_manager(self, mock_pool_class):
+        """Test that _get_cursor properly manages connection lifecycle."""
+        mock_pool = MagicMock()
+        mock_pool_class.return_value = mock_pool
+
+        mock_conn = MagicMock()
+        mock_cursor = MagicMock()
+        mock_pool.getconn.return_value = mock_conn
+        mock_conn.cursor.return_value = mock_cursor
+
+        pgvector = PGVector(self.collection_name, self.config)
+
+        with pgvector._get_cursor() as cur:
+            assert cur == mock_cursor
+
+        # Verify connection lifecycle methods were called
+        mock_pool.getconn.assert_called_once()
+        mock_cursor.close.assert_called_once()
+        mock_conn.commit.assert_called_once()
+        mock_pool.putconn.assert_called_once_with(mock_conn)
+
+
+@pytest.mark.parametrize(
+    "invalid_config_override",
+    [
+        {"host": ""},  # Test empty host
+        {"port": 0},  # Test invalid port
+        {"user": ""},  # Test empty user
+        {"password": ""},  # Test empty password
+        {"database": ""},  # Test empty database
+        {"min_connection": 0},  # Test invalid min_connection
+        {"max_connection": 0},  # Test invalid max_connection
+        {"min_connection": 10, "max_connection": 5},  # Test min > max
+    ],
+)
+def test_config_validation_parametrized(invalid_config_override):
+    """Test configuration validation for various invalid inputs using parametrize."""
+    config = {
+        "host": "localhost",
+        "port": 5432,
+        "user": "test_user",
+        "password": "test_password",
+        "database": "test_db",
+        "min_connection": 1,
+        "max_connection": 5,
+    }
+    config.update(invalid_config_override)
+
+    with pytest.raises(ValueError):
+        PGVectorConfig(**config)
+
+
+if __name__ == "__main__":
+    unittest.main()

api/tests/unit_tests/core/rag/extractor/test_pdf_extractor.py

@@ -0,0 +1,186 @@
+from types import SimpleNamespace
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+import core.rag.extractor.pdf_extractor as pe
+
+
+@pytest.fixture
+def mock_dependencies(monkeypatch):
+    # Mock storage
+    saves = []
+
+    def save(key, data):
+        saves.append((key, data))
+
+    monkeypatch.setattr(pe, "storage", SimpleNamespace(save=save))
+
+    # Mock db
+    class DummySession:
+        def __init__(self):
+            self.added = []
+            self.committed = False
+
+        def add(self, obj):
+            self.added.append(obj)
+
+        def add_all(self, objs):
+            self.added.extend(objs)
+
+        def commit(self):
+            self.committed = True
+
+    db_stub = SimpleNamespace(session=DummySession())
+    monkeypatch.setattr(pe, "db", db_stub)
+
+    # Mock UploadFile
+    class FakeUploadFile:
+        DEFAULT_ID = "test_file_id"
+
+        def __init__(self, **kwargs):
+            # Assign id from DEFAULT_ID, allow override via kwargs if needed
+            self.id = self.DEFAULT_ID
+            for k, v in kwargs.items():
+                setattr(self, k, v)
+
+    monkeypatch.setattr(pe, "UploadFile", FakeUploadFile)
+
+    # Mock config
+    monkeypatch.setattr(pe.dify_config, "FILES_URL", "http://files.local")
+    monkeypatch.setattr(pe.dify_config, "INTERNAL_FILES_URL", None)
+    monkeypatch.setattr(pe.dify_config, "STORAGE_TYPE", "local")
+
+    return SimpleNamespace(saves=saves, db=db_stub, UploadFile=FakeUploadFile)
+
+
+@pytest.mark.parametrize(
+    ("image_bytes", "expected_mime", "expected_ext", "file_id"),
+    [
+        (b"\xff\xd8\xff some jpeg", "image/jpeg", "jpg", "test_file_id_jpeg"),
+        (b"\x89PNG\r\n\x1a\n some png", "image/png", "png", "test_file_id_png"),
+    ],
+)
+def test_extract_images_formats(mock_dependencies, monkeypatch, image_bytes, expected_mime, expected_ext, file_id):
+    saves = mock_dependencies.saves
+    db_stub = mock_dependencies.db
+
+    # Customize FakeUploadFile id for this test case.
+    # Using monkeypatch ensures the class attribute is reset between parameter sets.
+    monkeypatch.setattr(mock_dependencies.UploadFile, "DEFAULT_ID", file_id)
+
+    # Mock page and image objects
+    mock_page = MagicMock()
+    mock_image_obj = MagicMock()
+
+    def mock_extract(buf, fb_format=None):
+        buf.write(image_bytes)
+
+    mock_image_obj.extract.side_effect = mock_extract
+
+    mock_page.get_objects.return_value = [mock_image_obj]
+
+    extractor = pe.PdfExtractor(file_path="test.pdf", tenant_id="t1", user_id="u1")
+
+    # We need to handle the import inside _extract_images
+    with patch("pypdfium2.raw") as mock_raw:
+        mock_raw.FPDF_PAGEOBJ_IMAGE = 1
+        result = extractor._extract_images(mock_page)
+
+    assert f"![image](http://files.local/files/{file_id}/file-preview)" in result
+    assert len(saves) == 1
+    assert saves[0][1] == image_bytes
+    assert len(db_stub.session.added) == 1
+    assert db_stub.session.added[0].tenant_id == "t1"
+    assert db_stub.session.added[0].size == len(image_bytes)
+    assert db_stub.session.added[0].mime_type == expected_mime
+    assert db_stub.session.added[0].extension == expected_ext
+    assert db_stub.session.committed is True
+
+
+@pytest.mark.parametrize(
+    ("get_objects_side_effect", "get_objects_return_value"),
+    [
+        (None, []),  # Empty list
+        (None, None),  # None returned
+        (Exception("Failed to get objects"), None),  # Exception raised
+    ],
+)
+def test_extract_images_get_objects_scenarios(mock_dependencies, get_objects_side_effect, get_objects_return_value):
+    mock_page = MagicMock()
+    if get_objects_side_effect:
+        mock_page.get_objects.side_effect = get_objects_side_effect
+    else:
+        mock_page.get_objects.return_value = get_objects_return_value
+
+    extractor = pe.PdfExtractor(file_path="test.pdf", tenant_id="t1", user_id="u1")
+
+    with patch("pypdfium2.raw") as mock_raw:
+        mock_raw.FPDF_PAGEOBJ_IMAGE = 1
+        result = extractor._extract_images(mock_page)
+
+    assert result == ""
+
+
+def test_extract_calls_extract_images(mock_dependencies, monkeypatch):
+    # Mock pypdfium2
+    mock_pdf_doc = MagicMock()
+    mock_page = MagicMock()
+    mock_pdf_doc.__iter__.return_value = [mock_page]
+
+    # Mock text extraction
+    mock_text_page = MagicMock()
+    mock_text_page.get_text_range.return_value = "Page text content"
+    mock_page.get_textpage.return_value = mock_text_page
+
+    with patch("pypdfium2.PdfDocument", return_value=mock_pdf_doc):
+        # Mock Blob
+        mock_blob = MagicMock()
+        mock_blob.source = "test.pdf"
+        with patch("core.rag.extractor.pdf_extractor.Blob.from_path", return_value=mock_blob):
+            extractor = pe.PdfExtractor(file_path="test.pdf", tenant_id="t1", user_id="u1")
+
+            # Mock _extract_images to return a known string
+            monkeypatch.setattr(extractor, "_extract_images", lambda p: "![image](img_url)")
+
+            documents = list(extractor.extract())
+
+            assert len(documents) == 1
+            assert "Page text content" in documents[0].page_content
+            assert "![image](img_url)" in documents[0].page_content
+            assert documents[0].metadata["page"] == 0
+
+
+def test_extract_images_failures(mock_dependencies):
+    saves = mock_dependencies.saves
+    db_stub = mock_dependencies.db
+
+    # Mock page and image objects
+    mock_page = MagicMock()
+    mock_image_obj_fail = MagicMock()
+    mock_image_obj_ok = MagicMock()
+
+    # First image raises exception
+    mock_image_obj_fail.extract.side_effect = Exception("Extraction failure")
+
+    # Second image is OK (JPEG)
+    jpeg_bytes = b"\xff\xd8\xff some image data"
+
+    def mock_extract(buf, fb_format=None):
+        buf.write(jpeg_bytes)
+
+    mock_image_obj_ok.extract.side_effect = mock_extract
+
+    mock_page.get_objects.return_value = [mock_image_obj_fail, mock_image_obj_ok]
+
+    extractor = pe.PdfExtractor(file_path="test.pdf", tenant_id="t1", user_id="u1")
+
+    with patch("pypdfium2.raw") as mock_raw:
+        mock_raw.FPDF_PAGEOBJ_IMAGE = 1
+        result = extractor._extract_images(mock_page)
+
+    # Should have one success
+    assert "![image](http://files.local/files/test_file_id/file-preview)" in result
+    assert len(saves) == 1
+    assert saves[0][1] == jpeg_bytes
+    assert db_stub.session.committed is True

api/tests/unit_tests/core/rag/retrieval/test_dataset_retrieval.py

@@ -421,7 +421,18 @@ class TestRetrievalService:
             # In real code, this waits for all futures to complete
             # In tests, futures complete immediately, so wait is a no-op
             with patch("core.rag.datasource.retrieval_service.concurrent.futures.wait"):
-                yield mock_executor
+                # Mock concurrent.futures.as_completed for early error propagation
+                # In real code, this yields futures as they complete
+                # In tests, we yield all futures immediately since they're already done
+                def mock_as_completed(futures_list, timeout=None):
+                    """Mock as_completed that yields futures immediately."""
+                    yield from futures_list
+
+                with patch(
+                    "core.rag.datasource.retrieval_service.concurrent.futures.as_completed",
+                    side_effect=mock_as_completed,
+                ):
+                    yield mock_executor
 
     # ==================== Vector Search Tests ====================
 

api/tests/unit_tests/core/rag/retrieval/test_dataset_retrieval_metadata_filter.py

@@ -0,0 +1,873 @@
+"""
+Unit tests for DatasetRetrieval.process_metadata_filter_func.
+
+This module provides comprehensive test coverage for the process_metadata_filter_func
+method in the DatasetRetrieval class, which is responsible for building SQLAlchemy
+filter expressions based on metadata filtering conditions.
+
+Conditions Tested:
+==================
+1. **String Conditions**: contains, not contains, start with, end with
+2. **Equality Conditions**: is / =, is not / ≠
+3. **Null Conditions**: empty, not empty
+4. **Numeric Comparisons**: before / <, after / >, ≤ / <=, ≥ / >=
+5. **List Conditions**: in
+6. **Edge Cases**: None values, different data types (str, int, float)
+
+Test Architecture:
+==================
+- Direct instantiation of DatasetRetrieval
+- Mocking of DatasetDocument model attributes
+- Verification of SQLAlchemy filter expressions
+- Follows Arrange-Act-Assert (AAA) pattern
+
+Running Tests:
+==============
+    # Run all tests in this module
+    uv run --project api pytest \
+        api/tests/unit_tests/core/rag/retrieval/test_dataset_retrieval_metadata_filter.py -v
+
+    # Run a specific test
+    uv run --project api pytest \
+        api/tests/unit_tests/core/rag/retrieval/test_dataset_retrieval_metadata_filter.py::\
+TestProcessMetadataFilterFunc::test_contains_condition -v
+"""
+
+from unittest.mock import MagicMock
+
+import pytest
+
+from core.rag.retrieval.dataset_retrieval import DatasetRetrieval
+
+
+class TestProcessMetadataFilterFunc:
+    """
+    Comprehensive test suite for process_metadata_filter_func method.
+
+    This test class validates all metadata filtering conditions supported by
+    the DatasetRetrieval class, including string operations, numeric comparisons,
+    null checks, and list operations.
+
+    Method Signature:
+    ==================
+    def process_metadata_filter_func(
+        self, sequence: int, condition: str, metadata_name: str, value: Any | None, filters: list
+    ) -> list:
+
+    The method builds SQLAlchemy filter expressions by:
+    1. Validating value is not None (except for empty/not empty conditions)
+    2. Using DatasetDocument.doc_metadata JSON field operations
+    3. Adding appropriate SQLAlchemy expressions to the filters list
+    4. Returning the updated filters list
+
+    Mocking Strategy:
+    ==================
+    - Mock DatasetDocument.doc_metadata to avoid database dependencies
+    - Verify filter expressions are created correctly
+    - Test with various data types (str, int, float, list)
+    """
+
+    @pytest.fixture
+    def retrieval(self):
+        """
+        Create a DatasetRetrieval instance for testing.
+
+        Returns:
+            DatasetRetrieval: Instance to test process_metadata_filter_func
+        """
+        return DatasetRetrieval()
+
+    @pytest.fixture
+    def mock_doc_metadata(self):
+        """
+        Mock the DatasetDocument.doc_metadata JSON field.
+
+        The method uses DatasetDocument.doc_metadata[metadata_name] to access
+        JSON fields. We mock this to avoid database dependencies.
+
+        Returns:
+            Mock: Mocked doc_metadata attribute
+        """
+        mock_metadata_field = MagicMock()
+
+        # Create mock for string access
+        mock_string_access = MagicMock()
+        mock_string_access.like = MagicMock()
+        mock_string_access.notlike = MagicMock()
+        mock_string_access.__eq__ = MagicMock(return_value=MagicMock())
+        mock_string_access.__ne__ = MagicMock(return_value=MagicMock())
+        mock_string_access.in_ = MagicMock(return_value=MagicMock())
+
+        # Create mock for float access (for numeric comparisons)
+        mock_float_access = MagicMock()
+        mock_float_access.__eq__ = MagicMock(return_value=MagicMock())
+        mock_float_access.__ne__ = MagicMock(return_value=MagicMock())
+        mock_float_access.__lt__ = MagicMock(return_value=MagicMock())
+        mock_float_access.__gt__ = MagicMock(return_value=MagicMock())
+        mock_float_access.__le__ = MagicMock(return_value=MagicMock())
+        mock_float_access.__ge__ = MagicMock(return_value=MagicMock())
+
+        # Create mock for null checks
+        mock_null_access = MagicMock()
+        mock_null_access.is_ = MagicMock(return_value=MagicMock())
+        mock_null_access.isnot = MagicMock(return_value=MagicMock())
+
+        # Setup __getitem__ to return appropriate mock based on usage
+        def getitem_side_effect(name):
+            if name in ["author", "title", "category"]:
+                return mock_string_access
+            elif name in ["year", "price", "rating"]:
+                return mock_float_access
+            else:
+                return mock_string_access
+
+        mock_metadata_field.__getitem__ = MagicMock(side_effect=getitem_side_effect)
+        mock_metadata_field.as_string.return_value = mock_string_access
+        mock_metadata_field.as_float.return_value = mock_float_access
+        mock_metadata_field[metadata_name:str].is_ = mock_null_access.is_
+        mock_metadata_field[metadata_name:str].isnot = mock_null_access.isnot
+
+        return mock_metadata_field
+
+    # ==================== String Condition Tests ====================
+
+    def test_contains_condition_string_value(self, retrieval):
+        """
+        Test 'contains' condition with string value.
+
+        Verifies:
+        - Filters list is populated with LIKE expression
+        - Pattern matching uses %value% syntax
+        """
+        filters = []
+        sequence = 0
+        condition = "contains"
+        metadata_name = "author"
+        value = "John"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_not_contains_condition(self, retrieval):
+        """
+        Test 'not contains' condition.
+
+        Verifies:
+        - Filters list is populated with NOT LIKE expression
+        - Pattern matching uses %value% syntax with negation
+        """
+        filters = []
+        sequence = 0
+        condition = "not contains"
+        metadata_name = "title"
+        value = "banned"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_start_with_condition(self, retrieval):
+        """
+        Test 'start with' condition.
+
+        Verifies:
+        - Filters list is populated with LIKE expression
+        - Pattern matching uses value% syntax
+        """
+        filters = []
+        sequence = 0
+        condition = "start with"
+        metadata_name = "category"
+        value = "tech"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_end_with_condition(self, retrieval):
+        """
+        Test 'end with' condition.
+
+        Verifies:
+        - Filters list is populated with LIKE expression
+        - Pattern matching uses %value syntax
+        """
+        filters = []
+        sequence = 0
+        condition = "end with"
+        metadata_name = "filename"
+        value = ".pdf"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    # ==================== Equality Condition Tests ====================
+
+    def test_is_condition_with_string_value(self, retrieval):
+        """
+        Test 'is' (=) condition with string value.
+
+        Verifies:
+        - Filters list is populated with equality expression
+        - String comparison is used
+        """
+        filters = []
+        sequence = 0
+        condition = "is"
+        metadata_name = "author"
+        value = "Jane Doe"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_equals_condition_with_string_value(self, retrieval):
+        """
+        Test '=' condition with string value.
+
+        Verifies:
+        - Same behavior as 'is' condition
+        - String comparison is used
+        """
+        filters = []
+        sequence = 0
+        condition = "="
+        metadata_name = "category"
+        value = "technology"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_is_condition_with_int_value(self, retrieval):
+        """
+        Test 'is' condition with integer value.
+
+        Verifies:
+        - Numeric comparison is used
+        - as_float() is called on the metadata field
+        """
+        filters = []
+        sequence = 0
+        condition = "is"
+        metadata_name = "year"
+        value = 2023
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_is_condition_with_float_value(self, retrieval):
+        """
+        Test 'is' condition with float value.
+
+        Verifies:
+        - Numeric comparison is used
+        - as_float() is called on the metadata field
+        """
+        filters = []
+        sequence = 0
+        condition = "is"
+        metadata_name = "price"
+        value = 19.99
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_is_not_condition_with_string_value(self, retrieval):
+        """
+        Test 'is not' (≠) condition with string value.
+
+        Verifies:
+        - Filters list is populated with inequality expression
+        - String comparison is used
+        """
+        filters = []
+        sequence = 0
+        condition = "is not"
+        metadata_name = "author"
+        value = "Unknown"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_not_equals_condition(self, retrieval):
+        """
+        Test '≠' condition with string value.
+
+        Verifies:
+        - Same behavior as 'is not' condition
+        - Inequality expression is used
+        """
+        filters = []
+        sequence = 0
+        condition = "≠"
+        metadata_name = "category"
+        value = "archived"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_is_not_condition_with_numeric_value(self, retrieval):
+        """
+        Test 'is not' condition with numeric value.
+
+        Verifies:
+        - Numeric inequality comparison is used
+        - as_float() is called on the metadata field
+        """
+        filters = []
+        sequence = 0
+        condition = "is not"
+        metadata_name = "year"
+        value = 2000
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    # ==================== Null Condition Tests ====================
+
+    def test_empty_condition(self, retrieval):
+        """
+        Test 'empty' condition (null check).
+
+        Verifies:
+        - Filters list is populated with IS NULL expression
+        - Value can be None for this condition
+        """
+        filters = []
+        sequence = 0
+        condition = "empty"
+        metadata_name = "author"
+        value = None
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_not_empty_condition(self, retrieval):
+        """
+        Test 'not empty' condition (not null check).
+
+        Verifies:
+        - Filters list is populated with IS NOT NULL expression
+        - Value can be None for this condition
+        """
+        filters = []
+        sequence = 0
+        condition = "not empty"
+        metadata_name = "description"
+        value = None
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    # ==================== Numeric Comparison Tests ====================
+
+    def test_before_condition(self, retrieval):
+        """
+        Test 'before' (<) condition.
+
+        Verifies:
+        - Filters list is populated with less than expression
+        - Numeric comparison is used
+        """
+        filters = []
+        sequence = 0
+        condition = "before"
+        metadata_name = "year"
+        value = 2020
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_less_than_condition(self, retrieval):
+        """
+        Test '<' condition.
+
+        Verifies:
+        - Same behavior as 'before' condition
+        - Less than expression is used
+        """
+        filters = []
+        sequence = 0
+        condition = "<"
+        metadata_name = "price"
+        value = 100.0
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_after_condition(self, retrieval):
+        """
+        Test 'after' (>) condition.
+
+        Verifies:
+        - Filters list is populated with greater than expression
+        - Numeric comparison is used
+        """
+        filters = []
+        sequence = 0
+        condition = "after"
+        metadata_name = "year"
+        value = 2020
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_greater_than_condition(self, retrieval):
+        """
+        Test '>' condition.
+
+        Verifies:
+        - Same behavior as 'after' condition
+        - Greater than expression is used
+        """
+        filters = []
+        sequence = 0
+        condition = ">"
+        metadata_name = "rating"
+        value = 4.5
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_less_than_or_equal_condition_unicode(self, retrieval):
+        """
+        Test '≤' condition.
+
+        Verifies:
+        - Filters list is populated with less than or equal expression
+        - Numeric comparison is used
+        """
+        filters = []
+        sequence = 0
+        condition = "≤"
+        metadata_name = "price"
+        value = 50.0
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_less_than_or_equal_condition_ascii(self, retrieval):
+        """
+        Test '<=' condition.
+
+        Verifies:
+        - Same behavior as '≤' condition
+        - Less than or equal expression is used
+        """
+        filters = []
+        sequence = 0
+        condition = "<="
+        metadata_name = "year"
+        value = 2023
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_greater_than_or_equal_condition_unicode(self, retrieval):
+        """
+        Test '≥' condition.
+
+        Verifies:
+        - Filters list is populated with greater than or equal expression
+        - Numeric comparison is used
+        """
+        filters = []
+        sequence = 0
+        condition = "≥"
+        metadata_name = "rating"
+        value = 3.5
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_greater_than_or_equal_condition_ascii(self, retrieval):
+        """
+        Test '>=' condition.
+
+        Verifies:
+        - Same behavior as '≥' condition
+        - Greater than or equal expression is used
+        """
+        filters = []
+        sequence = 0
+        condition = ">="
+        metadata_name = "year"
+        value = 2000
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    # ==================== List/In Condition Tests ====================
+
+    def test_in_condition_with_comma_separated_string(self, retrieval):
+        """
+        Test 'in' condition with comma-separated string value.
+
+        Verifies:
+        - String is split into list
+        - Whitespace is trimmed from each value
+        - IN expression is created
+        """
+        filters = []
+        sequence = 0
+        condition = "in"
+        metadata_name = "category"
+        value = "tech, science,  AI  "
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_in_condition_with_list_value(self, retrieval):
+        """
+        Test 'in' condition with list value.
+
+        Verifies:
+        - List is processed correctly
+        - None values are filtered out
+        - IN expression is created with valid values
+        """
+        filters = []
+        sequence = 0
+        condition = "in"
+        metadata_name = "tags"
+        value = ["python", "javascript", None, "golang"]
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_in_condition_with_tuple_value(self, retrieval):
+        """
+        Test 'in' condition with tuple value.
+
+        Verifies:
+        - Tuple is processed like a list
+        - IN expression is created
+        """
+        filters = []
+        sequence = 0
+        condition = "in"
+        metadata_name = "category"
+        value = ("tech", "science", "ai")
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_in_condition_with_empty_string(self, retrieval):
+        """
+        Test 'in' condition with empty string value.
+
+        Verifies:
+        - Empty string results in literal(False) filter
+        - No valid values to match
+        """
+        filters = []
+        sequence = 0
+        condition = "in"
+        metadata_name = "category"
+        value = ""
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+        # Verify it's a literal(False) expression
+        # This is a bit tricky to test without access to the actual expression
+
+    def test_in_condition_with_only_whitespace(self, retrieval):
+        """
+        Test 'in' condition with whitespace-only string value.
+
+        Verifies:
+        - Whitespace-only string results in literal(False) filter
+        - All values are stripped and filtered out
+        """
+        filters = []
+        sequence = 0
+        condition = "in"
+        metadata_name = "category"
+        value = "   ,   ,   "
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_in_condition_with_single_string(self, retrieval):
+        """
+        Test 'in' condition with single non-comma string.
+
+        Verifies:
+        - Single string is treated as single-item list
+        - IN expression is created with one value
+        """
+        filters = []
+        sequence = 0
+        condition = "in"
+        metadata_name = "category"
+        value = "technology"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    # ==================== Edge Case Tests ====================
+
+    def test_none_value_with_non_empty_condition(self, retrieval):
+        """
+        Test None value with conditions that require value.
+
+        Verifies:
+        - Original filters list is returned unchanged
+        - No filter is added for None values (except empty/not empty)
+        """
+        filters = []
+        sequence = 0
+        condition = "contains"
+        metadata_name = "author"
+        value = None
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 0  # No filter added
+
+    def test_none_value_with_equals_condition(self, retrieval):
+        """
+        Test None value with 'is' (=) condition.
+
+        Verifies:
+        - Original filters list is returned unchanged
+        - No filter is added for None values
+        """
+        filters = []
+        sequence = 0
+        condition = "is"
+        metadata_name = "author"
+        value = None
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 0
+
+    def test_none_value_with_numeric_condition(self, retrieval):
+        """
+        Test None value with numeric comparison condition.
+
+        Verifies:
+        - Original filters list is returned unchanged
+        - No filter is added for None values
+        """
+        filters = []
+        sequence = 0
+        condition = ">"
+        metadata_name = "year"
+        value = None
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 0
+
+    def test_existing_filters_preserved(self, retrieval):
+        """
+        Test that existing filters are preserved.
+
+        Verifies:
+        - Existing filters in the list are not removed
+        - New filters are appended to the list
+        """
+        existing_filter = MagicMock()
+        filters = [existing_filter]
+        sequence = 0
+        condition = "contains"
+        metadata_name = "author"
+        value = "test"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 2
+        assert filters[0] == existing_filter
+
+    def test_multiple_filters_accumulated(self, retrieval):
+        """
+        Test multiple calls to accumulate filters.
+
+        Verifies:
+        - Each call adds a new filter to the list
+        - All filters are preserved across calls
+        """
+        filters = []
+
+        # First filter
+        retrieval.process_metadata_filter_func(0, "contains", "author", "John", filters)
+        assert len(filters) == 1
+
+        # Second filter
+        retrieval.process_metadata_filter_func(1, ">", "year", 2020, filters)
+        assert len(filters) == 2
+
+        # Third filter
+        retrieval.process_metadata_filter_func(2, "is", "category", "tech", filters)
+        assert len(filters) == 3
+
+    def test_unknown_condition(self, retrieval):
+        """
+        Test unknown/unsupported condition.
+
+        Verifies:
+        - Original filters list is returned unchanged
+        - No filter is added for unknown conditions
+        """
+        filters = []
+        sequence = 0
+        condition = "unknown_condition"
+        metadata_name = "author"
+        value = "test"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 0
+
+    def test_empty_string_value_with_contains(self, retrieval):
+        """
+        Test empty string value with 'contains' condition.
+
+        Verifies:
+        - Filter is added even with empty string
+        - LIKE expression is created
+        """
+        filters = []
+        sequence = 0
+        condition = "contains"
+        metadata_name = "author"
+        value = ""
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_special_characters_in_value(self, retrieval):
+        """
+        Test special characters in value string.
+
+        Verifies:
+        - Special characters are handled in value
+        - LIKE expression is created correctly
+        """
+        filters = []
+        sequence = 0
+        condition = "contains"
+        metadata_name = "title"
+        value = "C++ & Python's features"
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_zero_value_with_numeric_condition(self, retrieval):
+        """
+        Test zero value with numeric comparison condition.
+
+        Verifies:
+        - Zero is treated as valid value
+        - Numeric comparison is performed
+        """
+        filters = []
+        sequence = 0
+        condition = ">"
+        metadata_name = "price"
+        value = 0
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_negative_value_with_numeric_condition(self, retrieval):
+        """
+        Test negative value with numeric comparison condition.
+
+        Verifies:
+        - Negative numbers are handled correctly
+        - Numeric comparison is performed
+        """
+        filters = []
+        sequence = 0
+        condition = "<"
+        metadata_name = "temperature"
+        value = -10.5
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1
+
+    def test_float_value_with_integer_comparison(self, retrieval):
+        """
+        Test float value with numeric comparison condition.
+
+        Verifies:
+        - Float values work correctly
+        - Numeric comparison is performed
+        """
+        filters = []
+        sequence = 0
+        condition = ">="
+        metadata_name = "rating"
+        value = 4.5
+
+        result = retrieval.process_metadata_filter_func(sequence, condition, metadata_name, value, filters)
+
+        assert result == filters
+        assert len(filters) == 1

api/tests/unit_tests/core/workflow/graph_engine/graph_traversal/__init__.py

@@ -0,0 +1 @@
+"""Tests for graph traversal components."""

api/tests/unit_tests/core/workflow/graph_engine/graph_traversal/test_skip_propagator.py

@@ -0,0 +1,307 @@
+"""Unit tests for skip propagator."""
+
+from unittest.mock import MagicMock, create_autospec
+
+from core.workflow.graph import Edge, Graph
+from core.workflow.graph_engine.graph_state_manager import GraphStateManager
+from core.workflow.graph_engine.graph_traversal.skip_propagator import SkipPropagator
+
+
+class TestSkipPropagator:
+    """Test suite for SkipPropagator."""
+
+    def test_propagate_skip_from_edge_with_unknown_edges_stops_processing(self) -> None:
+        """When there are unknown incoming edges, propagation should stop."""
+        # Arrange
+        mock_graph = create_autospec(Graph)
+        mock_state_manager = create_autospec(GraphStateManager)
+
+        # Create a mock edge
+        mock_edge = MagicMock(spec=Edge)
+        mock_edge.id = "edge_1"
+        mock_edge.head = "node_2"
+
+        # Setup graph edges dict
+        mock_graph.edges = {"edge_1": mock_edge}
+
+        # Setup incoming edges
+        incoming_edges = [MagicMock(spec=Edge), MagicMock(spec=Edge)]
+        mock_graph.get_incoming_edges.return_value = incoming_edges
+
+        # Setup state manager to return has_unknown=True
+        mock_state_manager.analyze_edge_states.return_value = {
+            "has_unknown": True,
+            "has_taken": False,
+            "all_skipped": False,
+        }
+
+        propagator = SkipPropagator(mock_graph, mock_state_manager)
+
+        # Act
+        propagator.propagate_skip_from_edge("edge_1")
+
+        # Assert
+        mock_graph.get_incoming_edges.assert_called_once_with("node_2")
+        mock_state_manager.analyze_edge_states.assert_called_once_with(incoming_edges)
+        # Should not call any other state manager methods
+        mock_state_manager.enqueue_node.assert_not_called()
+        mock_state_manager.start_execution.assert_not_called()
+        mock_state_manager.mark_node_skipped.assert_not_called()
+
+    def test_propagate_skip_from_edge_with_taken_edge_enqueues_node(self) -> None:
+        """When there is at least one taken edge, node should be enqueued."""
+        # Arrange
+        mock_graph = create_autospec(Graph)
+        mock_state_manager = create_autospec(GraphStateManager)
+
+        # Create a mock edge
+        mock_edge = MagicMock(spec=Edge)
+        mock_edge.id = "edge_1"
+        mock_edge.head = "node_2"
+
+        mock_graph.edges = {"edge_1": mock_edge}
+        incoming_edges = [MagicMock(spec=Edge)]
+        mock_graph.get_incoming_edges.return_value = incoming_edges
+
+        # Setup state manager to return has_taken=True
+        mock_state_manager.analyze_edge_states.return_value = {
+            "has_unknown": False,
+            "has_taken": True,
+            "all_skipped": False,
+        }
+
+        propagator = SkipPropagator(mock_graph, mock_state_manager)
+
+        # Act
+        propagator.propagate_skip_from_edge("edge_1")
+
+        # Assert
+        mock_state_manager.enqueue_node.assert_called_once_with("node_2")
+        mock_state_manager.start_execution.assert_called_once_with("node_2")
+        mock_state_manager.mark_node_skipped.assert_not_called()
+
+    def test_propagate_skip_from_edge_with_all_skipped_propagates_to_node(self) -> None:
+        """When all incoming edges are skipped, should propagate skip to node."""
+        # Arrange
+        mock_graph = create_autospec(Graph)
+        mock_state_manager = create_autospec(GraphStateManager)
+
+        # Create a mock edge
+        mock_edge = MagicMock(spec=Edge)
+        mock_edge.id = "edge_1"
+        mock_edge.head = "node_2"
+
+        mock_graph.edges = {"edge_1": mock_edge}
+        incoming_edges = [MagicMock(spec=Edge)]
+        mock_graph.get_incoming_edges.return_value = incoming_edges
+
+        # Setup state manager to return all_skipped=True
+        mock_state_manager.analyze_edge_states.return_value = {
+            "has_unknown": False,
+            "has_taken": False,
+            "all_skipped": True,
+        }
+
+        propagator = SkipPropagator(mock_graph, mock_state_manager)
+
+        # Act
+        propagator.propagate_skip_from_edge("edge_1")
+
+        # Assert
+        mock_state_manager.mark_node_skipped.assert_called_once_with("node_2")
+        mock_state_manager.enqueue_node.assert_not_called()
+        mock_state_manager.start_execution.assert_not_called()
+
+    def test_propagate_skip_to_node_marks_node_and_outgoing_edges_skipped(self) -> None:
+        """_propagate_skip_to_node should mark node and all outgoing edges as skipped."""
+        # Arrange
+        mock_graph = create_autospec(Graph)
+        mock_state_manager = create_autospec(GraphStateManager)
+
+        # Create outgoing edges
+        edge1 = MagicMock(spec=Edge)
+        edge1.id = "edge_2"
+        edge1.head = "node_downstream_1"  # Set head for propagate_skip_from_edge
+
+        edge2 = MagicMock(spec=Edge)
+        edge2.id = "edge_3"
+        edge2.head = "node_downstream_2"
+
+        # Setup graph edges dict for propagate_skip_from_edge
+        mock_graph.edges = {"edge_2": edge1, "edge_3": edge2}
+        mock_graph.get_outgoing_edges.return_value = [edge1, edge2]
+
+        # Setup get_incoming_edges to return empty list to stop recursion
+        mock_graph.get_incoming_edges.return_value = []
+
+        propagator = SkipPropagator(mock_graph, mock_state_manager)
+
+        # Use mock to call private method
+        # Act
+        propagator._propagate_skip_to_node("node_1")
+
+        # Assert
+        mock_state_manager.mark_node_skipped.assert_called_once_with("node_1")
+        mock_state_manager.mark_edge_skipped.assert_any_call("edge_2")
+        mock_state_manager.mark_edge_skipped.assert_any_call("edge_3")
+        assert mock_state_manager.mark_edge_skipped.call_count == 2
+        # Should recursively propagate from each edge
+        # Since propagate_skip_from_edge is called, we need to verify it was called
+        # But we can't directly verify due to recursion. We'll trust the logic.
+
+    def test_skip_branch_paths_marks_unselected_edges_and_propagates(self) -> None:
+        """skip_branch_paths should mark all unselected edges as skipped and propagate."""
+        # Arrange
+        mock_graph = create_autospec(Graph)
+        mock_state_manager = create_autospec(GraphStateManager)
+
+        # Create unselected edges
+        edge1 = MagicMock(spec=Edge)
+        edge1.id = "edge_1"
+        edge1.head = "node_downstream_1"
+
+        edge2 = MagicMock(spec=Edge)
+        edge2.id = "edge_2"
+        edge2.head = "node_downstream_2"
+
+        unselected_edges = [edge1, edge2]
+
+        # Setup graph edges dict
+        mock_graph.edges = {"edge_1": edge1, "edge_2": edge2}
+        # Setup get_incoming_edges to return empty list to stop recursion
+        mock_graph.get_incoming_edges.return_value = []
+
+        propagator = SkipPropagator(mock_graph, mock_state_manager)
+
+        # Act
+        propagator.skip_branch_paths(unselected_edges)
+
+        # Assert
+        mock_state_manager.mark_edge_skipped.assert_any_call("edge_1")
+        mock_state_manager.mark_edge_skipped.assert_any_call("edge_2")
+        assert mock_state_manager.mark_edge_skipped.call_count == 2
+        # propagate_skip_from_edge should be called for each edge
+        # We can't directly verify due to the mock, but the logic is covered
+
+    def test_propagate_skip_from_edge_recursively_propagates_through_graph(self) -> None:
+        """Skip propagation should recursively propagate through the graph."""
+        # Arrange
+        mock_graph = create_autospec(Graph)
+        mock_state_manager = create_autospec(GraphStateManager)
+
+        # Create edge chain: edge_1 -> node_2 -> edge_3 -> node_4
+        edge1 = MagicMock(spec=Edge)
+        edge1.id = "edge_1"
+        edge1.head = "node_2"
+
+        edge3 = MagicMock(spec=Edge)
+        edge3.id = "edge_3"
+        edge3.head = "node_4"
+
+        mock_graph.edges = {"edge_1": edge1, "edge_3": edge3}
+
+        # Setup get_incoming_edges to return different values based on node
+        def get_incoming_edges_side_effect(node_id):
+            if node_id == "node_2":
+                return [edge1]
+            elif node_id == "node_4":
+                return [edge3]
+            return []
+
+        mock_graph.get_incoming_edges.side_effect = get_incoming_edges_side_effect
+
+        # Setup get_outgoing_edges to return different values based on node
+        def get_outgoing_edges_side_effect(node_id):
+            if node_id == "node_2":
+                return [edge3]
+            elif node_id == "node_4":
+                return []  # No outgoing edges, stops recursion
+            return []
+
+        mock_graph.get_outgoing_edges.side_effect = get_outgoing_edges_side_effect
+
+        # Setup state manager to return all_skipped for both nodes
+        mock_state_manager.analyze_edge_states.return_value = {
+            "has_unknown": False,
+            "has_taken": False,
+            "all_skipped": True,
+        }
+
+        propagator = SkipPropagator(mock_graph, mock_state_manager)
+
+        # Act
+        propagator.propagate_skip_from_edge("edge_1")
+
+        # Assert
+        # Should mark node_2 as skipped
+        mock_state_manager.mark_node_skipped.assert_any_call("node_2")
+        # Should mark edge_3 as skipped
+        mock_state_manager.mark_edge_skipped.assert_any_call("edge_3")
+        # Should propagate to node_4
+        mock_state_manager.mark_node_skipped.assert_any_call("node_4")
+        assert mock_state_manager.mark_node_skipped.call_count == 2
+
+    def test_propagate_skip_from_edge_with_mixed_edge_states_handles_correctly(self) -> None:
+        """Test with mixed edge states (some unknown, some taken, some skipped)."""
+        # Arrange
+        mock_graph = create_autospec(Graph)
+        mock_state_manager = create_autospec(GraphStateManager)
+
+        mock_edge = MagicMock(spec=Edge)
+        mock_edge.id = "edge_1"
+        mock_edge.head = "node_2"
+
+        mock_graph.edges = {"edge_1": mock_edge}
+        incoming_edges = [MagicMock(spec=Edge), MagicMock(spec=Edge), MagicMock(spec=Edge)]
+        mock_graph.get_incoming_edges.return_value = incoming_edges
+
+        # Test 1: has_unknown=True, has_taken=False, all_skipped=False
+        mock_state_manager.analyze_edge_states.return_value = {
+            "has_unknown": True,
+            "has_taken": False,
+            "all_skipped": False,
+        }
+
+        propagator = SkipPropagator(mock_graph, mock_state_manager)
+
+        # Act
+        propagator.propagate_skip_from_edge("edge_1")
+
+        # Assert - should stop processing
+        mock_state_manager.enqueue_node.assert_not_called()
+        mock_state_manager.mark_node_skipped.assert_not_called()
+
+        # Reset mocks for next test
+        mock_state_manager.reset_mock()
+        mock_graph.reset_mock()
+
+        # Test 2: has_unknown=False, has_taken=True, all_skipped=False
+        mock_state_manager.analyze_edge_states.return_value = {
+            "has_unknown": False,
+            "has_taken": True,
+            "all_skipped": False,
+        }
+
+        # Act
+        propagator.propagate_skip_from_edge("edge_1")
+
+        # Assert - should enqueue node
+        mock_state_manager.enqueue_node.assert_called_once_with("node_2")
+        mock_state_manager.start_execution.assert_called_once_with("node_2")
+
+        # Reset mocks for next test
+        mock_state_manager.reset_mock()
+        mock_graph.reset_mock()
+
+        # Test 3: has_unknown=False, has_taken=False, all_skipped=True
+        mock_state_manager.analyze_edge_states.return_value = {
+            "has_unknown": False,
+            "has_taken": False,
+            "all_skipped": True,
+        }
+
+        # Act
+        propagator.propagate_skip_from_edge("edge_1")
+
+        # Assert - should propagate skip
+        mock_state_manager.mark_node_skipped.assert_called_once_with("node_2")

api/tests/unit_tests/extensions/test_celery_ssl.py

@@ -8,11 +8,12 @@ class TestCelerySSLConfiguration:
     """Test suite for Celery SSL configuration."""
 
     def test_get_celery_ssl_options_when_ssl_disabled(self):
-        """Test SSL options when REDIS_USE_SSL is False."""
-        mock_config = MagicMock()
-        mock_config.REDIS_USE_SSL = False
+        """Test SSL options when BROKER_USE_SSL is False."""
+        from configs import DifyConfig
 
-        with patch("extensions.ext_celery.dify_config", mock_config):
+        dify_config = DifyConfig(CELERY_BROKER_URL="redis://localhost:6379/0")
+
+        with patch("extensions.ext_celery.dify_config", dify_config):
             from extensions.ext_celery import _get_celery_ssl_options
 
             result = _get_celery_ssl_options()
@@ -21,7 +22,6 @@ class TestCelerySSLConfiguration:
     def test_get_celery_ssl_options_when_broker_not_redis(self):
         """Test SSL options when broker is not Redis."""
         mock_config = MagicMock()
-        mock_config.REDIS_USE_SSL = True
         mock_config.CELERY_BROKER_URL = "amqp://localhost:5672"
 
         with patch("extensions.ext_celery.dify_config", mock_config):
@@ -33,7 +33,6 @@ class TestCelerySSLConfiguration:
     def test_get_celery_ssl_options_with_cert_none(self):
         """Test SSL options with CERT_NONE requirement."""
         mock_config = MagicMock()
-        mock_config.REDIS_USE_SSL = True
         mock_config.CELERY_BROKER_URL = "redis://localhost:6379/0"
         mock_config.REDIS_SSL_CERT_REQS = "CERT_NONE"
         mock_config.REDIS_SSL_CA_CERTS = None
@@ -53,7 +52,6 @@ class TestCelerySSLConfiguration:
     def test_get_celery_ssl_options_with_cert_required(self):
         """Test SSL options with CERT_REQUIRED and certificates."""
         mock_config = MagicMock()
-        mock_config.REDIS_USE_SSL = True
         mock_config.CELERY_BROKER_URL = "rediss://localhost:6380/0"
         mock_config.REDIS_SSL_CERT_REQS = "CERT_REQUIRED"
         mock_config.REDIS_SSL_CA_CERTS = "/path/to/ca.crt"
@@ -73,7 +71,6 @@ class TestCelerySSLConfiguration:
     def test_get_celery_ssl_options_with_cert_optional(self):
         """Test SSL options with CERT_OPTIONAL requirement."""
         mock_config = MagicMock()
-        mock_config.REDIS_USE_SSL = True
         mock_config.CELERY_BROKER_URL = "redis://localhost:6379/0"
         mock_config.REDIS_SSL_CERT_REQS = "CERT_OPTIONAL"
         mock_config.REDIS_SSL_CA_CERTS = "/path/to/ca.crt"
@@ -91,7 +88,6 @@ class TestCelerySSLConfiguration:
     def test_get_celery_ssl_options_with_invalid_cert_reqs(self):
         """Test SSL options with invalid cert requirement defaults to CERT_NONE."""
         mock_config = MagicMock()
-        mock_config.REDIS_USE_SSL = True
         mock_config.CELERY_BROKER_URL = "redis://localhost:6379/0"
         mock_config.REDIS_SSL_CERT_REQS = "INVALID_VALUE"
         mock_config.REDIS_SSL_CA_CERTS = None
@@ -108,7 +104,6 @@ class TestCelerySSLConfiguration:
     def test_celery_init_applies_ssl_to_broker_and_backend(self):
         """Test that SSL options are applied to both broker and backend when using Redis."""
         mock_config = MagicMock()
-        mock_config.REDIS_USE_SSL = True
         mock_config.CELERY_BROKER_URL = "redis://localhost:6379/0"
         mock_config.CELERY_BACKEND = "redis"
         mock_config.CELERY_RESULT_BACKEND = "redis://localhost:6379/0"

api/tests/unit_tests/libs/test_archive_storage.py

@@ -0,0 +1,272 @@
+import base64
+import hashlib
+from datetime import datetime
+from unittest.mock import ANY, MagicMock
+
+import pytest
+from botocore.exceptions import ClientError
+
+from libs import archive_storage as storage_module
+from libs.archive_storage import (
+    ArchiveStorage,
+    ArchiveStorageError,
+    ArchiveStorageNotConfiguredError,
+)
+
+BUCKET_NAME = "archive-bucket"
+
+
+def _configure_storage(monkeypatch, **overrides):
+    defaults = {
+        "ARCHIVE_STORAGE_ENABLED": True,
+        "ARCHIVE_STORAGE_ENDPOINT": "https://storage.example.com",
+        "ARCHIVE_STORAGE_ARCHIVE_BUCKET": BUCKET_NAME,
+        "ARCHIVE_STORAGE_ACCESS_KEY": "access",
+        "ARCHIVE_STORAGE_SECRET_KEY": "secret",
+        "ARCHIVE_STORAGE_REGION": "auto",
+    }
+    defaults.update(overrides)
+    for key, value in defaults.items():
+        monkeypatch.setattr(storage_module.dify_config, key, value, raising=False)
+
+
+def _client_error(code: str) -> ClientError:
+    return ClientError({"Error": {"Code": code}}, "Operation")
+
+
+def _mock_client(monkeypatch):
+    client = MagicMock()
+    client.head_bucket.return_value = None
+    boto_client = MagicMock(return_value=client)
+    monkeypatch.setattr(storage_module.boto3, "client", boto_client)
+    return client, boto_client
+
+
+def test_init_disabled(monkeypatch):
+    _configure_storage(monkeypatch, ARCHIVE_STORAGE_ENABLED=False)
+    with pytest.raises(ArchiveStorageNotConfiguredError, match="not enabled"):
+        ArchiveStorage(bucket=BUCKET_NAME)
+
+
+def test_init_missing_config(monkeypatch):
+    _configure_storage(monkeypatch, ARCHIVE_STORAGE_ENDPOINT=None)
+    with pytest.raises(ArchiveStorageNotConfiguredError, match="incomplete"):
+        ArchiveStorage(bucket=BUCKET_NAME)
+
+
+def test_init_bucket_not_found(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.head_bucket.side_effect = _client_error("404")
+
+    with pytest.raises(ArchiveStorageNotConfiguredError, match="does not exist"):
+        ArchiveStorage(bucket=BUCKET_NAME)
+
+
+def test_init_bucket_access_denied(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.head_bucket.side_effect = _client_error("403")
+
+    with pytest.raises(ArchiveStorageNotConfiguredError, match="Access denied"):
+        ArchiveStorage(bucket=BUCKET_NAME)
+
+
+def test_init_bucket_other_error(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.head_bucket.side_effect = _client_error("500")
+
+    with pytest.raises(ArchiveStorageError, match="Failed to access archive bucket"):
+        ArchiveStorage(bucket=BUCKET_NAME)
+
+
+def test_init_sets_client(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, boto_client = _mock_client(monkeypatch)
+
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    boto_client.assert_called_once_with(
+        "s3",
+        endpoint_url="https://storage.example.com",
+        aws_access_key_id="access",
+        aws_secret_access_key="secret",
+        region_name="auto",
+        config=ANY,
+    )
+    assert storage.client is client
+    assert storage.bucket == BUCKET_NAME
+
+
+def test_put_object_returns_checksum(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    data = b"hello"
+    checksum = storage.put_object("key", data)
+
+    expected_md5 = hashlib.md5(data).hexdigest()
+    expected_content_md5 = base64.b64encode(hashlib.md5(data).digest()).decode()
+    client.put_object.assert_called_once_with(
+        Bucket="archive-bucket",
+        Key="key",
+        Body=data,
+        ContentMD5=expected_content_md5,
+    )
+    assert checksum == expected_md5
+
+
+def test_put_object_raises_on_error(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+    client.put_object.side_effect = _client_error("500")
+
+    with pytest.raises(ArchiveStorageError, match="Failed to upload object"):
+        storage.put_object("key", b"data")
+
+
+def test_get_object_returns_bytes(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    body = MagicMock()
+    body.read.return_value = b"payload"
+    client.get_object.return_value = {"Body": body}
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    assert storage.get_object("key") == b"payload"
+
+
+def test_get_object_missing(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.get_object.side_effect = _client_error("NoSuchKey")
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    with pytest.raises(FileNotFoundError, match="Archive object not found"):
+        storage.get_object("missing")
+
+
+def test_get_object_stream(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    body = MagicMock()
+    body.iter_chunks.return_value = [b"a", b"b"]
+    client.get_object.return_value = {"Body": body}
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    assert list(storage.get_object_stream("key")) == [b"a", b"b"]
+
+
+def test_get_object_stream_missing(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.get_object.side_effect = _client_error("NoSuchKey")
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    with pytest.raises(FileNotFoundError, match="Archive object not found"):
+        list(storage.get_object_stream("missing"))
+
+
+def test_object_exists(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    assert storage.object_exists("key") is True
+    client.head_object.side_effect = _client_error("404")
+    assert storage.object_exists("missing") is False
+
+
+def test_delete_object_error(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.delete_object.side_effect = _client_error("500")
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    with pytest.raises(ArchiveStorageError, match="Failed to delete object"):
+        storage.delete_object("key")
+
+
+def test_list_objects(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    paginator = MagicMock()
+    paginator.paginate.return_value = [
+        {"Contents": [{"Key": "a"}, {"Key": "b"}]},
+        {"Contents": [{"Key": "c"}]},
+    ]
+    client.get_paginator.return_value = paginator
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    assert storage.list_objects("prefix") == ["a", "b", "c"]
+    paginator.paginate.assert_called_once_with(Bucket="archive-bucket", Prefix="prefix")
+
+
+def test_list_objects_error(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    paginator = MagicMock()
+    paginator.paginate.side_effect = _client_error("500")
+    client.get_paginator.return_value = paginator
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    with pytest.raises(ArchiveStorageError, match="Failed to list objects"):
+        storage.list_objects("prefix")
+
+
+def test_generate_presigned_url(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.generate_presigned_url.return_value = "http://signed-url"
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    url = storage.generate_presigned_url("key", expires_in=123)
+
+    client.generate_presigned_url.assert_called_once_with(
+        ClientMethod="get_object",
+        Params={"Bucket": "archive-bucket", "Key": "key"},
+        ExpiresIn=123,
+    )
+    assert url == "http://signed-url"
+
+
+def test_generate_presigned_url_error(monkeypatch):
+    _configure_storage(monkeypatch)
+    client, _ = _mock_client(monkeypatch)
+    client.generate_presigned_url.side_effect = _client_error("500")
+    storage = ArchiveStorage(bucket=BUCKET_NAME)
+
+    with pytest.raises(ArchiveStorageError, match="Failed to generate pre-signed URL"):
+        storage.generate_presigned_url("key")
+
+
+def test_serialization_roundtrip():
+    records = [
+        {
+            "id": "1",
+            "created_at": datetime(2024, 1, 1, 12, 0, 0),
+            "payload": {"nested": "value"},
+            "items": [{"name": "a"}],
+        },
+        {"id": "2", "value": 123},
+    ]
+
+    data = ArchiveStorage.serialize_to_jsonl_gz(records)
+    decoded = ArchiveStorage.deserialize_from_jsonl_gz(data)
+
+    assert decoded[0]["id"] == "1"
+    assert decoded[0]["payload"]["nested"] == "value"
+    assert decoded[0]["items"][0]["name"] == "a"
+    assert "2024-01-01T12:00:00" in decoded[0]["created_at"]
+    assert decoded[1]["value"] == 123
+
+
+def test_content_md5_matches_checksum():
+    data = b"checksum"
+    expected = base64.b64encode(hashlib.md5(data).digest()).decode()
+
+    assert ArchiveStorage._content_md5(data) == expected
+    assert ArchiveStorage.compute_checksum(data) == hashlib.md5(data).hexdigest()

api/tests/unit_tests/oss/tencent_cos/test_tencent_cos.py

@@ -1,4 +1,4 @@
-from unittest.mock import patch
+from unittest.mock import MagicMock, patch
 
 import pytest
 from qcloud_cos import CosConfig
@@ -18,3 +18,72 @@ class TestTencentCos(BaseStorageTest):
         with patch.object(CosConfig, "__init__", return_value=None):
             self.storage = TencentCosStorage()
         self.storage.bucket_name = get_example_bucket()
+
+
+class TestTencentCosConfiguration:
+    """Tests for TencentCosStorage initialization with different configurations."""
+
+    def test_init_with_custom_domain(self):
+        """Test initialization with custom domain configured."""
+        # Mock dify_config to return custom domain configuration
+        mock_dify_config = MagicMock()
+        mock_dify_config.TENCENT_COS_CUSTOM_DOMAIN = "cos.example.com"
+        mock_dify_config.TENCENT_COS_SECRET_ID = "test-secret-id"
+        mock_dify_config.TENCENT_COS_SECRET_KEY = "test-secret-key"
+        mock_dify_config.TENCENT_COS_SCHEME = "https"
+
+        # Mock CosConfig and CosS3Client
+        mock_config_instance = MagicMock()
+        mock_client = MagicMock()
+
+        with (
+            patch("extensions.storage.tencent_cos_storage.dify_config", mock_dify_config),
+            patch(
+                "extensions.storage.tencent_cos_storage.CosConfig", return_value=mock_config_instance
+            ) as mock_cos_config,
+            patch("extensions.storage.tencent_cos_storage.CosS3Client", return_value=mock_client),
+        ):
+            TencentCosStorage()
+
+            # Verify CosConfig was called with Domain parameter (not Region)
+            mock_cos_config.assert_called_once()
+            call_kwargs = mock_cos_config.call_args[1]
+            assert "Domain" in call_kwargs
+            assert call_kwargs["Domain"] == "cos.example.com"
+            assert "Region" not in call_kwargs
+            assert call_kwargs["SecretId"] == "test-secret-id"
+            assert call_kwargs["SecretKey"] == "test-secret-key"
+            assert call_kwargs["Scheme"] == "https"
+
+    def test_init_with_region(self):
+        """Test initialization with region configured (no custom domain)."""
+        # Mock dify_config to return region configuration
+        mock_dify_config = MagicMock()
+        mock_dify_config.TENCENT_COS_CUSTOM_DOMAIN = None
+        mock_dify_config.TENCENT_COS_REGION = "ap-guangzhou"
+        mock_dify_config.TENCENT_COS_SECRET_ID = "test-secret-id"
+        mock_dify_config.TENCENT_COS_SECRET_KEY = "test-secret-key"
+        mock_dify_config.TENCENT_COS_SCHEME = "https"
+
+        # Mock CosConfig and CosS3Client
+        mock_config_instance = MagicMock()
+        mock_client = MagicMock()
+
+        with (
+            patch("extensions.storage.tencent_cos_storage.dify_config", mock_dify_config),
+            patch(
+                "extensions.storage.tencent_cos_storage.CosConfig", return_value=mock_config_instance
+            ) as mock_cos_config,
+            patch("extensions.storage.tencent_cos_storage.CosS3Client", return_value=mock_client),
+        ):
+            TencentCosStorage()
+
+            # Verify CosConfig was called with Region parameter (not Domain)
+            mock_cos_config.assert_called_once()
+            call_kwargs = mock_cos_config.call_args[1]
+            assert "Region" in call_kwargs
+            assert call_kwargs["Region"] == "ap-guangzhou"
+            assert "Domain" not in call_kwargs
+            assert call_kwargs["SecretId"] == "test-secret-id"
+            assert call_kwargs["SecretKey"] == "test-secret-key"
+            assert call_kwargs["Scheme"] == "https"

api/tests/unit_tests/services/test_billing_service.py

@@ -1294,6 +1294,42 @@ class TestBillingServiceSubscriptionOperations:
         # Assert
         assert result == {}
 
+    def test_get_plan_bulk_with_invalid_tenant_plan_skipped(self, mock_send_request):
+        """Test bulk plan retrieval when one tenant has invalid plan data (should skip that tenant)."""
+        # Arrange
+        tenant_ids = ["tenant-valid-1", "tenant-invalid", "tenant-valid-2"]
+
+        # Response with one invalid tenant plan (missing expiration_date) and two valid ones
+        mock_send_request.return_value = {
+            "data": {
+                "tenant-valid-1": {"plan": "sandbox", "expiration_date": 1735689600},
+                "tenant-invalid": {"plan": "professional"},  # Missing expiration_date field
+                "tenant-valid-2": {"plan": "team", "expiration_date": 1767225600},
+            }
+        }
+
+        # Act
+        with patch("services.billing_service.logger") as mock_logger:
+            result = BillingService.get_plan_bulk(tenant_ids)
+
+        # Assert - should only contain valid tenants
+        assert len(result) == 2
+        assert "tenant-valid-1" in result
+        assert "tenant-valid-2" in result
+        assert "tenant-invalid" not in result
+
+        # Verify valid tenants have correct data
+        assert result["tenant-valid-1"]["plan"] == "sandbox"
+        assert result["tenant-valid-1"]["expiration_date"] == 1735689600
+        assert result["tenant-valid-2"]["plan"] == "team"
+        assert result["tenant-valid-2"]["expiration_date"] == 1767225600
+
+        # Verify exception was logged for the invalid tenant
+        mock_logger.exception.assert_called_once()
+        log_call_args = mock_logger.exception.call_args[0]
+        assert "get_plan_bulk: failed to validate subscription plan for tenant" in log_call_args[0]
+        assert "tenant-invalid" in log_call_args[1]
+
     def test_get_expired_subscription_cleanup_whitelist_success(self, mock_send_request):
         """Test successful retrieval of expired subscription cleanup whitelist."""
         # Arrange

api/tests/unit_tests/tools/test_mcp_tool.py

@@ -0,0 +1,122 @@
+import base64
+from unittest.mock import Mock, patch
+
+import pytest
+
+from core.mcp.types import (
+    AudioContent,
+    BlobResourceContents,
+    CallToolResult,
+    EmbeddedResource,
+    ImageContent,
+    TextResourceContents,
+)
+from core.tools.__base.tool_runtime import ToolRuntime
+from core.tools.entities.common_entities import I18nObject
+from core.tools.entities.tool_entities import ToolEntity, ToolIdentity, ToolInvokeMessage
+from core.tools.mcp_tool.tool import MCPTool
+
+
+def _make_mcp_tool(output_schema: dict | None = None) -> MCPTool:
+    identity = ToolIdentity(
+        author="test",
+        name="test_mcp_tool",
+        label=I18nObject(en_US="Test MCP Tool", zh_Hans="测试MCP工具"),
+        provider="test_provider",
+    )
+    entity = ToolEntity(identity=identity, output_schema=output_schema or {})
+    runtime = Mock(spec=ToolRuntime)
+    runtime.credentials = {}
+    return MCPTool(
+        entity=entity,
+        runtime=runtime,
+        tenant_id="test_tenant",
+        icon="",
+        server_url="https://server.invalid",
+        provider_id="provider_1",
+        headers={},
+    )
+
+
+class TestMCPToolInvoke:
+    @pytest.mark.parametrize(
+        ("content_factory", "mime_type"),
+        [
+            (
+                lambda b64, mt: ImageContent(type="image", data=b64, mimeType=mt),
+                "image/png",
+            ),
+            (
+                lambda b64, mt: AudioContent(type="audio", data=b64, mimeType=mt),
+                "audio/mpeg",
+            ),
+        ],
+    )
+    def test_invoke_image_or_audio_yields_blob(self, content_factory, mime_type) -> None:
+        tool = _make_mcp_tool()
+        raw = b"\x00\x01test-bytes\x02"
+        b64 = base64.b64encode(raw).decode()
+        content = content_factory(b64, mime_type)
+        result = CallToolResult(content=[content])
+
+        with patch.object(tool, "invoke_remote_mcp_tool", return_value=result):
+            messages = list(tool._invoke(user_id="test_user", tool_parameters={}))
+
+        assert len(messages) == 1
+        msg = messages[0]
+        assert msg.type == ToolInvokeMessage.MessageType.BLOB
+        assert isinstance(msg.message, ToolInvokeMessage.BlobMessage)
+        assert msg.message.blob == raw
+        assert msg.meta == {"mime_type": mime_type}
+
+    def test_invoke_embedded_text_resource_yields_text(self) -> None:
+        tool = _make_mcp_tool()
+        text_resource = TextResourceContents(uri="file://test.txt", mimeType="text/plain", text="hello world")
+        content = EmbeddedResource(type="resource", resource=text_resource)
+        result = CallToolResult(content=[content])
+
+        with patch.object(tool, "invoke_remote_mcp_tool", return_value=result):
+            messages = list(tool._invoke(user_id="test_user", tool_parameters={}))
+
+        assert len(messages) == 1
+        msg = messages[0]
+        assert msg.type == ToolInvokeMessage.MessageType.TEXT
+        assert isinstance(msg.message, ToolInvokeMessage.TextMessage)
+        assert msg.message.text == "hello world"
+
+    @pytest.mark.parametrize(
+        ("mime_type", "expected_mime"),
+        [("application/pdf", "application/pdf"), (None, "application/octet-stream")],
+    )
+    def test_invoke_embedded_blob_resource_yields_blob(self, mime_type, expected_mime) -> None:
+        tool = _make_mcp_tool()
+        raw = b"binary-data"
+        b64 = base64.b64encode(raw).decode()
+        blob_resource = BlobResourceContents(uri="file://doc.bin", mimeType=mime_type, blob=b64)
+        content = EmbeddedResource(type="resource", resource=blob_resource)
+        result = CallToolResult(content=[content])
+
+        with patch.object(tool, "invoke_remote_mcp_tool", return_value=result):
+            messages = list(tool._invoke(user_id="test_user", tool_parameters={}))
+
+        assert len(messages) == 1
+        msg = messages[0]
+        assert msg.type == ToolInvokeMessage.MessageType.BLOB
+        assert isinstance(msg.message, ToolInvokeMessage.BlobMessage)
+        assert msg.message.blob == raw
+        assert msg.meta == {"mime_type": expected_mime}
+
+    def test_invoke_yields_variables_when_structured_content_and_schema(self) -> None:
+        tool = _make_mcp_tool(output_schema={"type": "object"})
+        result = CallToolResult(content=[], structuredContent={"a": 1, "b": "x"})
+
+        with patch.object(tool, "invoke_remote_mcp_tool", return_value=result):
+            messages = list(tool._invoke(user_id="test_user", tool_parameters={}))
+
+        # Expect two variable messages corresponding to keys a and b
+        assert len(messages) == 2
+        var_msgs = [m for m in messages if isinstance(m.message, ToolInvokeMessage.VariableMessage)]
+        assert {m.message.variable_name for m in var_msgs} == {"a", "b"}
+        # Validate values
+        values = {m.message.variable_name: m.message.variable_value for m in var_msgs}
+        assert values == {"a": 1, "b": "x"}

api/tests/unit_tests/utils/test_text_processing.py

@@ -15,6 +15,11 @@ from core.tools.utils.text_processing_utils import remove_leading_symbols
         ("", ""),
         ("   ", "   "),
         ("【测试】", "【测试】"),
+        # Markdown link preservation - should be preserved if text starts with a markdown link
+        ("[Google](https://google.com) is a search engine", "[Google](https://google.com) is a search engine"),
+        ("[Example](http://example.com) some text", "[Example](http://example.com) some text"),
+        # Leading symbols before markdown link are removed, including the opening bracket [
+        ("@[Test](https://example.com)", "Test](https://example.com)"),
     ],
 )
 def test_remove_leading_symbols(input_text, expected_output):

api/uv.lock

@@ -3072,11 +3072,11 @@ wheels = [
 
 [[package]]
 name = "json-repair"
-version = "0.54.1"
+version = "0.54.3"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/00/46/d3a4d9a3dad39bb4a2ad16b8adb9fe2e8611b20b71197fe33daa6768e85d/json_repair-0.54.1.tar.gz", hash = "sha256:d010bc31f1fc66e7c36dc33bff5f8902674498ae5cb8e801ad455a53b455ad1d", size = 38555, upload-time = "2025-11-19T14:55:24.265Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/b5/86/48b12ac02032f121ac7e5f11a32143edca6c1e3d19ffc54d6fb9ca0aafd0/json_repair-0.54.3.tar.gz", hash = "sha256:e50feec9725e52ac91f12184609754684ac1656119dfbd31de09bdaf9a1d8bf6", size = 38626, upload-time = "2025-12-15T09:41:58.594Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/db/96/c9aad7ee949cc1bf15df91f347fbc2d3bd10b30b80c7df689ce6fe9332b5/json_repair-0.54.1-py3-none-any.whl", hash = "sha256:016160c5db5d5fe443164927bb58d2dfbba5f43ad85719fa9bc51c713a443ab1", size = 29311, upload-time = "2025-11-19T14:55:22.886Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/08/abe317237add63c3e62f18a981bccf92112b431835b43d844aedaf61f4a0/json_repair-0.54.3-py3-none-any.whl", hash = "sha256:4cdc132ee27d4780576f71bf27a113877046224a808bfc17392e079cb344fb81", size = 29357, upload-time = "2025-12-15T09:41:57.436Z" },
 ]
 
 [[package]]

docker/.env.example

@@ -399,6 +399,7 @@ CONSOLE_CORS_ALLOW_ORIGINS=*
 COOKIE_DOMAIN=
 # When the frontend and backend run on different subdomains, set NEXT_PUBLIC_COOKIE_DOMAIN=1.
 NEXT_PUBLIC_COOKIE_DOMAIN=
+NEXT_PUBLIC_BATCH_CONCURRENCY=5
 
 # ------------------------------
 # File Storage Configuration
@@ -446,6 +447,15 @@ S3_SECRET_KEY=
 # If set to false, the access key and secret key must be provided.
 S3_USE_AWS_MANAGED_IAM=false
 
+# Workflow run and Conversation archive storage (S3-compatible)
+ARCHIVE_STORAGE_ENABLED=false
+ARCHIVE_STORAGE_ENDPOINT=
+ARCHIVE_STORAGE_ARCHIVE_BUCKET=
+ARCHIVE_STORAGE_EXPORT_BUCKET=
+ARCHIVE_STORAGE_ACCESS_KEY=
+ARCHIVE_STORAGE_SECRET_KEY=
+ARCHIVE_STORAGE_REGION=auto
+
 # Azure Blob Configuration
 #
 AZURE_BLOB_ACCOUNT_NAME=difyai
@@ -477,6 +487,7 @@ TENCENT_COS_SECRET_KEY=your-secret-key
 TENCENT_COS_SECRET_ID=your-secret-id
 TENCENT_COS_REGION=your-region
 TENCENT_COS_SCHEME=your-scheme
+TENCENT_COS_CUSTOM_DOMAIN=your-custom-domain
 
 # Oracle Storage Configuration
 #

docker/docker-compose.yaml

@@ -108,6 +108,7 @@ x-shared-env: &shared-api-worker-env
   CONSOLE_CORS_ALLOW_ORIGINS: ${CONSOLE_CORS_ALLOW_ORIGINS:-*}
   COOKIE_DOMAIN: ${COOKIE_DOMAIN:-}
   NEXT_PUBLIC_COOKIE_DOMAIN: ${NEXT_PUBLIC_COOKIE_DOMAIN:-}
+  NEXT_PUBLIC_BATCH_CONCURRENCY: ${NEXT_PUBLIC_BATCH_CONCURRENCY:-5}
   STORAGE_TYPE: ${STORAGE_TYPE:-opendal}
   OPENDAL_SCHEME: ${OPENDAL_SCHEME:-fs}
   OPENDAL_FS_ROOT: ${OPENDAL_FS_ROOT:-storage}
@@ -121,6 +122,13 @@ x-shared-env: &shared-api-worker-env
   S3_ACCESS_KEY: ${S3_ACCESS_KEY:-}
   S3_SECRET_KEY: ${S3_SECRET_KEY:-}
   S3_USE_AWS_MANAGED_IAM: ${S3_USE_AWS_MANAGED_IAM:-false}
+  ARCHIVE_STORAGE_ENABLED: ${ARCHIVE_STORAGE_ENABLED:-false}
+  ARCHIVE_STORAGE_ENDPOINT: ${ARCHIVE_STORAGE_ENDPOINT:-}
+  ARCHIVE_STORAGE_ARCHIVE_BUCKET: ${ARCHIVE_STORAGE_ARCHIVE_BUCKET:-}
+  ARCHIVE_STORAGE_EXPORT_BUCKET: ${ARCHIVE_STORAGE_EXPORT_BUCKET:-}
+  ARCHIVE_STORAGE_ACCESS_KEY: ${ARCHIVE_STORAGE_ACCESS_KEY:-}
+  ARCHIVE_STORAGE_SECRET_KEY: ${ARCHIVE_STORAGE_SECRET_KEY:-}
+  ARCHIVE_STORAGE_REGION: ${ARCHIVE_STORAGE_REGION:-auto}
   AZURE_BLOB_ACCOUNT_NAME: ${AZURE_BLOB_ACCOUNT_NAME:-difyai}
   AZURE_BLOB_ACCOUNT_KEY: ${AZURE_BLOB_ACCOUNT_KEY:-difyai}
   AZURE_BLOB_CONTAINER_NAME: ${AZURE_BLOB_CONTAINER_NAME:-difyai-container}
@@ -140,6 +148,7 @@ x-shared-env: &shared-api-worker-env
   TENCENT_COS_SECRET_ID: ${TENCENT_COS_SECRET_ID:-your-secret-id}
   TENCENT_COS_REGION: ${TENCENT_COS_REGION:-your-region}
   TENCENT_COS_SCHEME: ${TENCENT_COS_SCHEME:-your-scheme}
+  TENCENT_COS_CUSTOM_DOMAIN: ${TENCENT_COS_CUSTOM_DOMAIN:-your-custom-domain}
   OCI_ENDPOINT: ${OCI_ENDPOINT:-https://your-object-storage-namespace.compat.objectstorage.us-ashburn-1.oraclecloud.com}
   OCI_BUCKET_NAME: ${OCI_BUCKET_NAME:-your-bucket-name}
   OCI_ACCESS_KEY: ${OCI_ACCESS_KEY:-your-access-key}

sdks/nodejs-client/package.json

@@ -54,17 +54,17 @@
     "publish:npm": "./scripts/publish.sh"
   },
   "dependencies": {
-    "axios": "^1.3.5"
+    "axios": "^1.13.2"
   },
   "devDependencies": {
-    "@eslint/js": "^9.2.0",
-    "@types/node": "^20.11.30",
+    "@eslint/js": "^9.39.2",
+    "@types/node": "^25.0.3",
     "@typescript-eslint/eslint-plugin": "^8.50.1",
     "@typescript-eslint/parser": "^8.50.1",
-    "@vitest/coverage-v8": "1.6.1",
-    "eslint": "^9.2.0",
+    "@vitest/coverage-v8": "4.0.16",
+    "eslint": "^9.39.2",
     "tsup": "^8.5.1",
-    "typescript": "^5.4.5",
-    "vitest": "^1.5.0"
+    "typescript": "^5.9.3",
+    "vitest": "^4.0.16"
   }
 }