feat: Word 文档提取时保留分页符

- 修改 WordTextExtractionService 遍历文档结构
- 检测 CTBr (page break) 和 SectPr (section break) 元素
- 在分页位置插入 \f 字符
- DocumentIndexService 可以正确识别分页并生成多页索引

backend/parse-service/src/main/java/com/lingyue/parse/service/WordTextExtractionService.java

@@ -4,8 +4,8 @@ import com.lingyue.common.exception.ServiceException;
 import lombok.extern.slf4j.Slf4j;
 import org.apache.poi.hwpf.HWPFDocument;
 import org.apache.poi.hwpf.extractor.WordExtractor;
-import org.apache.poi.xwpf.extractor.XWPFWordExtractor;
-import org.apache.poi.xwpf.usermodel.XWPFDocument;
+import org.apache.poi.xwpf.usermodel.*;
+import org.openxmlformats.schemas.wordprocessingml.x2006.main.CTR;
 import org.springframework.stereotype.Service;
 
 import java.io.File;
@@ -15,6 +15,7 @@ import java.io.IOException;
 /**
  * Word文档文本提取服务
  * 支持.docx和.doc格式
+ * 在分页符位置插入 \f 字符，便于后续识别页码
  * 
  * @author lingyue
  * @since 2026-01-14
@@ -23,11 +24,16 @@ import java.io.IOException;
 @Service
 public class WordTextExtractionService {
     
+    /**
+     * 分页符字符
+     */
+    private static final char PAGE_BREAK = '\f';
+    
     /**
      * 提取Word文档文本
      * 
      * @param wordFilePath Word文件路径
-     * @return 提取的文本内容
+     * @return 提取的文本内容（分页符位置插入 \f）
      */
     public String extractText(String wordFilePath) {
         File wordFile = new File(wordFilePath);
@@ -39,7 +45,7 @@ public class WordTextExtractionService {
         
         try {
             if (fileName.endsWith(".docx")) {
-                return extractFromDocx(wordFilePath);
+                return extractFromDocxWithPageBreaks(wordFilePath);
             } else if (fileName.endsWith(".doc")) {
                 return extractFromDoc(wordFilePath);
             } else {
@@ -52,23 +58,88 @@ public class WordTextExtractionService {
     }
     
     /**
-     * 从.docx文件提取文本
+     * 从.docx文件提取文本，保留分页符
      */
-    private String extractFromDocx(String filePath) throws IOException {
-        log.info("提取.docx文件文本: {}", filePath);
+    private String extractFromDocxWithPageBreaks(String filePath) throws IOException {
+        log.info("提取.docx文件文本（含分页符）: {}", filePath);
+        
+        StringBuilder sb = new StringBuilder();
+        int pageBreakCount = 0;
         
         try (FileInputStream fis = new FileInputStream(filePath);
-             XWPFDocument document = new XWPFDocument(fis);
-             XWPFWordExtractor extractor = new XWPFWordExtractor(document)) {
+             XWPFDocument document = new XWPFDocument(fis)) {
             
-            String text = extractor.getText();
-            log.debug("提取到文本长度: {}", text != null ? text.length() : 0);
-            return text != null ? text : "";
+            // 遍历文档主体的所有元素
+            for (IBodyElement element : document.getBodyElements()) {
+                if (element instanceof XWPFParagraph) {
+                    XWPFParagraph paragraph = (XWPFParagraph) element;
+                    
+                    // 检查段落中的分页符
+                    for (XWPFRun run : paragraph.getRuns()) {
+                        // 检查 run 中是否有分页符
+                        CTR ctr = run.getCTR();
+                        if (ctr != null) {
+                            // 检查是否有硬分页符 (page break)
+                            if (ctr.getBrList() != null) {
+                                for (org.openxmlformats.schemas.wordprocessingml.x2006.main.CTBr br : ctr.getBrList()) {
+                                    if (br.getType() == org.openxmlformats.schemas.wordprocessingml.x2006.main.STBrType.PAGE) {
+                                        sb.append(PAGE_BREAK);
+                                        pageBreakCount++;
+                                    }
+                                }
+                            }
+                        }
+                        
+                        // 添加文本内容
+                        String text = run.getText(0);
+                        if (text != null) {
+                            sb.append(text);
+                        }
+                    }
+                    
+                    // 检查段落后是否有分页符（通过段落属性）
+                    if (paragraph.getCTP() != null && paragraph.getCTP().getPPr() != null) {
+                        var pPr = paragraph.getCTP().getPPr();
+                        // 检查分节符带来的分页
+                        if (pPr.getSectPr() != null) {
+                            var sectPr = pPr.getSectPr();
+                            if (sectPr.getType() != null) {
+                                String type = sectPr.getType().getVal().toString();
+                                if ("nextPage".equals(type) || "oddPage".equals(type) || "evenPage".equals(type)) {
+                                    sb.append(PAGE_BREAK);
+                                    pageBreakCount++;
+                                }
+                            }
+                        }
+                    }
+                    
+                    sb.append("\n");
+                    
+                } else if (element instanceof XWPFTable) {
+                    XWPFTable table = (XWPFTable) element;
+                    for (XWPFTableRow row : table.getRows()) {
+                        for (XWPFTableCell cell : row.getTableCells()) {
+                            sb.append(cell.getText()).append("\t");
+                        }
+                        sb.append("\n");
+                    }
+                }
+            }
+            
+            // 检查文档末尾的分节符
+            if (document.getDocument().getBody().getSectPr() != null) {
+                // 文档末尾的分节符不需要额外处理
+            }
         }
+        
+        String result = sb.toString();
+        log.info("提取完成: 文本长度={}, 分页符数量={}", result.length(), pageBreakCount);
+        return result;
     }
     
     /**
      * 从.doc文件提取文本
+     * .doc 格式的分页符通常会被 WordExtractor 保留为 \f
      */
     private String extractFromDoc(String filePath) throws IOException {
         log.info("提取.doc文件文本: {}", filePath);
@@ -78,7 +149,14 @@ public class WordTextExtractionService {
              WordExtractor extractor = new WordExtractor(document)) {
             
             String text = extractor.getText();
-            log.debug("提取到文本长度: {}", text != null ? text.length() : 0);
+            int pageBreakCount = 0;
+            if (text != null) {
+                for (char c : text.toCharArray()) {
+                    if (c == PAGE_BREAK) pageBreakCount++;
+                }
+            }
+            log.info("提取完成: 文本长度={}, 分页符数量={}", 
+                    text != null ? text.length() : 0, pageBreakCount);
             return text != null ? text : "";
         }
     }