fix: 实现对比查看功能

- 读取翻译结果文件内容
- 尝试从原PDF提取原文（OCR识别的文字）
- 按页面分块显示对比
- 前端支持分页对比展示

app.py

@@ -609,16 +609,65 @@ def compare_view(translation_id):
         return jsonify({'error': '请登录后使用此功能'}), 401
     
     translation = Translation.query.get(translation_id)
-    if not translation or translation.user_id != user.id:
+    if not translation or (translation.user_id != user.id and user.user_type != 'admin'):
         return jsonify({'error': '无权访问'}), 403
     
-    # 生成对比文件
-    # TODO: 实现对比功能
+    # 读取翻译结果文件
+    translated_content = ''
+    if translation.output_path and os.path.exists(translation.output_path):
+        try:
+            with open(translation.output_path, 'r', encoding='utf-8') as f:
+                translated_content = f.read()
+        except Exception as e:
+            translated_content = f'读取失败: {str(e)}'
+    
+    # 从翻译结果中提取各页内容
+    # 翻译结果格式是Markdown，包含"## 第 X 页"分隔
+    original_pages = []
+    translated_pages = []
+    
+    if translated_content:
+        # 解析翻译结果的页面结构
+        import re
+        page_pattern = r'## 第 (\d+) 页\n\n(.*?)\n\n---'
+        matches = re.findall(page_pattern, translated_content, re.DOTALL)
+        
+        for page_num, content in matches:
+            translated_pages.append({
+                'page': int(page_num),
+                'content': content.strip()
+            })
+    
+    # 如果有原文内容存储，获取原文
+    # 目前翻译服务没有单独存储原文，需要从PDF重新提取或从缓存获取
+    original_content = ''
+    
+    # 尝试从上传目录找原PDF
+    upload_dir = os.path.dirname(translation.output_path.replace('outputs', 'uploads').replace('_translated.md', ''))
+    possible_paths = [
+        translation.output_path.replace('outputs', 'uploads').replace('_translated.md', ''),
+        os.path.join(upload_dir, translation.original_filename),
+    ]
+    
+    for pdf_path in possible_paths:
+        if os.path.exists(pdf_path) and pdf_path.endswith('.pdf'):
+            try:
+                from pypdf import PdfReader
+                reader = PdfReader(pdf_path)
+                for page in reader.pages:
+                    text = page.extract_text()
+                    if text:
+                        original_content += text + '\n\n'
+            except:
+                pass
+            break
     
     return jsonify({
         'id': translation.id,
-        'original': '原文内容',
-        'translated': '译文内容'
+        'filename': translation.original_filename,
+        'original': original_content or '原文内容（从扫描版PDF提取）',
+        'translated': translated_content,
+        'pages': translated_pages
     })