fix: prevent preview infinite loop and add document type filtering

- Remove onAutoConfigReceived callback that caused state update loop
- Add document analysis to check if file needs OCR track
- Only show preprocessing options for OCR-eligible files (images, scanned PDFs)
- Show informative message for editable PDFs that use direct text extraction
- Display text coverage percentage for editable documents

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

frontend/src/components/PreprocessingPreview.tsx

@@ -16,7 +16,6 @@ interface PreprocessingPreviewProps {
   config: PreprocessingConfig
   page?: number
   className?: string
-  onAutoConfigReceived?: (config: PreprocessingConfig) => void
 }
 
 export default function PreprocessingPreview({
@@ -25,7 +24,6 @@ export default function PreprocessingPreview({
   config,
   page = 1,
   className,
-  onAutoConfigReceived,
 }: PreprocessingPreviewProps) {
   const { t } = useTranslation()
   const [isLoading, setIsLoading] = useState(false)
@@ -81,18 +79,13 @@ export default function PreprocessingPreview({
 
       setOriginalImageUrl(URL.createObjectURL(originalBlob))
       setPreprocessedImageUrl(URL.createObjectURL(preprocessedBlob))
-
-      // Pass auto config to parent if available
-      if (response.auto_config && onAutoConfigReceived) {
-        onAutoConfigReceived(response.auto_config)
-      }
     } catch (err: any) {
       console.error('Preview fetch error:', err)
       setError(err.response?.data?.detail || err.message || 'Failed to load preview')
     } finally {
       setIsLoading(false)
     }
-  }, [taskId, debouncedMode, debouncedConfig, page, onAutoConfigReceived])
+  }, [taskId, debouncedMode, debouncedConfig, page])
 
   useEffect(() => {
     fetchPreview()

frontend/src/pages/ProcessingPage.tsx

@@ -1,20 +1,20 @@
 import { useState, useEffect } from 'react'
 import { useNavigate } from 'react-router-dom'
 import { useTranslation } from 'react-i18next'
-import { useMutation } from '@tanstack/react-query'
+import { useMutation, useQuery } from '@tanstack/react-query'
 import { Card, CardContent, CardHeader, CardTitle } from '@/components/ui/card'
 import { Progress } from '@/components/ui/progress'
 import { Button } from '@/components/ui/button'
 import { Badge } from '@/components/ui/badge'
 import { useToast } from '@/components/ui/toast'
 import { apiClientV2 } from '@/services/apiV2'
-import { Play, CheckCircle, FileText, AlertCircle, Clock, Activity, Loader2 } from 'lucide-react'
+import { Play, CheckCircle, FileText, AlertCircle, Clock, Activity, Loader2, Info } from 'lucide-react'
 import LayoutModelSelector from '@/components/LayoutModelSelector'
 import PreprocessingSettings from '@/components/PreprocessingSettings'
 import PreprocessingPreview from '@/components/PreprocessingPreview'
 import TaskNotFound from '@/components/TaskNotFound'
 import { useTaskValidation } from '@/hooks/useTaskValidation'
-import type { LayoutModel, ProcessingOptions, PreprocessingMode, PreprocessingConfig } from '@/types/apiV2'
+import type { LayoutModel, ProcessingOptions, PreprocessingMode, PreprocessingConfig, DocumentAnalysisResponse } from '@/types/apiV2'
 
 export default function ProcessingPage() {
   const { t } = useTranslation()
@@ -47,6 +47,20 @@ export default function ProcessingPage() {
   })
   const [showPreview, setShowPreview] = useState(false)
 
+  // Analyze document to determine if OCR is needed (only for pending tasks)
+  const { data: documentAnalysis, isLoading: isAnalyzing } = useQuery({
+    queryKey: ['documentAnalysis', taskId],
+    queryFn: () => apiClientV2.analyzeDocument(taskId!),
+    enabled: !!taskId && taskDetail?.status === 'pending',
+    staleTime: Infinity, // Cache indefinitely since document doesn't change
+  })
+
+  // Determine if preprocessing options should be shown
+  // Only show for OCR track files (images and non-editable PDFs)
+  const needsOcrTrack = documentAnalysis?.recommended_track === 'ocr' ||
+    documentAnalysis?.recommended_track === 'hybrid' ||
+    !documentAnalysis // Show by default while analyzing
+
   // Start OCR processing
   const processOCRMutation = useMutation({
     mutationFn: () => {
@@ -360,36 +374,66 @@ export default function ProcessingPage() {
       {/* Processing Options (only show when task is pending) */}
       {isPending && (
         <div className="space-y-6">
-          {/* Layout Model Selection */}
+          {/* Document Analysis Loading */}
-          <LayoutModelSelector
+          {isAnalyzing && (
-            value={layoutModel}
+            <div className="flex items-center gap-2 p-4 bg-muted/30 rounded-lg border">
-            onChange={setLayoutModel}
+              <Loader2 className="w-4 h-4 animate-spin text-primary" />
-            disabled={processOCRMutation.isPending}
+              <span className="text-sm text-muted-foreground">分析文件類型中...</span>
-          />
+            </div>
+          )}
 
-          {/* Preprocessing Settings */}
+          {/* Direct Track Notice - Show when document is editable PDF */}
-          <PreprocessingSettings
+          {documentAnalysis && documentAnalysis.recommended_track === 'direct' && (
-            mode={preprocessingMode}
+            <Card className="border-blue-200 bg-blue-50">
-            config={preprocessingConfig}
+              <CardContent className="pt-4">
-            onModeChange={setPreprocessingMode}
+                <div className="flex items-start gap-3">
-            onConfigChange={setPreprocessingConfig}
+                  <Info className="w-5 h-5 text-blue-600 flex-shrink-0 mt-0.5" />
-            onPreview={() => setShowPreview(!showPreview)}
+                  <div>
-            disabled={processOCRMutation.isPending}
+                    <p className="text-sm font-medium text-blue-800">此文件為可編輯 PDF</p>
-          />
+                    <p className="text-sm text-blue-700 mt-1">
+                      系統偵測到此 PDF 包含文字圖層，將使用直接文字提取方式處理。
+                      版面偵測和影像前處理設定不適用於此類文件。
+                    </p>
+                    {documentAnalysis.text_coverage && (
+                      <p className="text-xs text-blue-600 mt-2">
+                        文字覆蓋率: {(documentAnalysis.text_coverage * 100).toFixed(1)}%
+                      </p>
+                    )}
+                  </div>
+                </div>
+              </CardContent>
+            </Card>
+          )}
 
-          {/* Preprocessing Preview */}
+          {/* OCR Track Options - Only show when document needs OCR */}
-          {showPreview && taskId && (
+          {needsOcrTrack && !isAnalyzing && (
-            <PreprocessingPreview
+            <>
-              taskId={taskId}
+              {/* Layout Model Selection */}
-              mode={preprocessingMode}
+              <LayoutModelSelector
-              config={preprocessingConfig}
+                value={layoutModel}
-              onAutoConfigReceived={(autoConfig) => {
+                onChange={setLayoutModel}
-                // Only update if user hasn't switched to manual mode
+                disabled={processOCRMutation.isPending}
-                if (preprocessingMode === 'auto') {
+              />
-                  setPreprocessingConfig(autoConfig)
+
-                }
+              {/* Preprocessing Settings */}
-              }}
+              <PreprocessingSettings
-            />
+                mode={preprocessingMode}
+                config={preprocessingConfig}
+                onModeChange={setPreprocessingMode}
+                onConfigChange={setPreprocessingConfig}
+                onPreview={() => setShowPreview(!showPreview)}
+                disabled={processOCRMutation.isPending}
+              />
+
+              {/* Preprocessing Preview */}
+              {showPreview && taskId && (
+                <PreprocessingPreview
+                  taskId={taskId}
+                  mode={preprocessingMode}
+                  config={preprocessingConfig}
+                />
+              )}
+            </>
           )}
         </div>
       )}