使用 POI 读取 Word 中的书签替换书签内容（doc和docx）

Posted 2022-04-19 catoop

通过操作书签可以实现 word 模板替换变量的功能场景，本文一下代码内容，直接可以复制使用正常编译运行。

添加 maven 依赖

    <dependencies>
        <dependency>
            <groupId>org.apache.poi</groupId>
            <artifactId>poi-ooxml</artifactId>
            <version>5.2.2</version>
        </dependency>
        <dependency>
            <groupId>org.apache.poi</groupId>
            <artifactId>poi-examples</artifactId>
            <version>5.2.2</version>
        </dependency>
        <dependency>
            <groupId>org.apache.poi</groupId>
            <artifactId>poi-scratchpad</artifactId>
            <version>5.2.2</version>
        </dependency>
    </dependencies>

操作书签的代码

package org.example;

import org.apache.poi.hwpf.HWPFDocument;
import org.apache.poi.hwpf.usermodel.Bookmark;
import org.apache.poi.hwpf.usermodel.Bookmarks;
import org.apache.poi.hwpf.usermodel.Range;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.usermodel.XWPFParagraph;
import org.apache.poi.xwpf.usermodel.XWPFRun;
import org.openxmlformats.schemas.wordprocessingml.x2006.main.CTBookmark;
import org.openxmlformats.schemas.wordprocessingml.x2006.main.CTP;
import org.w3c.dom.Node;

import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.nio.file.Files;
import java.nio.file.Paths;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

public class BookmarkDemo 

    /**
     * 因为docx为xml格式的结构，一下为docx中定义的部分常量引用
     **/
    public static final String RUN_NODE_NAME = "w:r";
    public static final String TEXT_NODE_NAME = "w:t";
    public static final String BOOKMARK_START_TAG = "w:bookmarkStart";
    public static final String BOOKMARK_END_TAG = "w:bookmarkEnd";
    public static final String BOOKMARK_ID_ATTR_NAME = "w:id";
    public static final String STYLE_NODE_NAME = "w:rPr";


    public static void main(String[] args) throws IOException 
        // 读取 doc 文件中的所有书签
        InputStream inputStream = Files.newInputStream(Paths.get("D:\\\\Desktop\\\\Welcome2.doc"));
        getBookmarksByDoc(inputStream);

        // 读取 doc 文件中的所有书签
        InputStream inputStream2 = Files.newInputStream(Paths.get("D:\\\\Desktop\\\\Welcome.docx"));
        getBookmarksByDocx(inputStream2);

        // 替换文件中的 bookmark 内容
        InputStream inputStream3 = Files.newInputStream(Paths.get("D:\\\\Desktop\\\\Welcome.docx"));
        Map<String, String> dataMap = new HashMap<>();
        dataMap.put("strong", "单红宇");
        dataMap.put("footnotes", "李小雨");
        replaceBookmarksByDocx(inputStream3, Files.newOutputStream(Paths.get("D:\\\\Desktop\\\\Welcome3.docx")), dataMap);
    


    /**
     * 读取 doc 文件中的所有书签
     *
     * @param inputStream
     * @throws IOException
     */
    public static void getBookmarksByDoc(InputStream inputStream) throws IOException 
        HWPFDocument wordDoc = new HWPFDocument(inputStream);

        Bookmarks bookmarks = wordDoc.getBookmarks();

        for (int b = 0; b < bookmarks.getBookmarksCount(); b++) 
            Bookmark bookmark = bookmarks.getBookmark(b);
            String bookMarkText = new Range(bookmark.getStart(), bookmark.getEnd(), wordDoc).text();

            System.out.println("[" + bookmark.getStart() + "; "
                    + bookmark.getEnd() + "]: " + bookmark.getName() + " = " + bookMarkText);
        

    

    /**
     * 读取 docx 文件中的所有书签
     *
     * @param inputStream
     * @throws IOException
     */
    public static void getBookmarksByDocx(InputStream inputStream) throws IOException 
        XWPFDocument docx = new XWPFDocument(inputStream);

        List<XWPFParagraph> paragraphList = docx.getParagraphs();
        for (XWPFParagraph xwpfParagraph : paragraphList) 
            CTP ctp = xwpfParagraph.getCTP();

            for (int dwI = 0; dwI < ctp.sizeOfBookmarkStartArray(); dwI++) 
                CTBookmark bookmark = ctp.getBookmarkStartArray(dwI);
                String bookmarkName = bookmark.getName();

                // 因为 docx 的结构是xml格式，它不像 doc 文档那样有具体的start和end值来定位一个bookmark的值范围，
                // 所有如果你想要读取 docx 文档中bookmark的内容，你需要从 bookmarkStart 节点开始逐级逐层依次解析xml文
                // 件的 nodeValue 值并进行拼接，直至读取到下一个名为 bookmarkEnd 的节点为止。
                // 下面这个方法 getBookmarkTextContent() 是一个没有实现的方法，如有需要请自行实现
                // String bookmarkTextContent = getBookmarkTextContent(bookmark);

                System.out.println(bookmarkName);
            
        
    


    /**
     * docx 文件中书签的替换
     *
     * @param inputStream
     * @param outputStream
     * @param dataMap
     * @throws IOException
     */
    public static void replaceBookmarksByDocx(InputStream inputStream, OutputStream outputStream, Map<String, String> dataMap) throws IOException 
        XWPFDocument document = new XWPFDocument(inputStream).getXWPFDocument();
        List<XWPFParagraph> paragraphList = document.getParagraphs();
        for (XWPFParagraph xwpfParagraph : paragraphList) 
            CTP ctp = xwpfParagraph.getCTP();

            for (int dwI = 0; dwI < ctp.sizeOfBookmarkStartArray(); dwI++) 
                CTBookmark bookmark = ctp.getBookmarkStartArray(dwI);
                if (dataMap.containsKey(bookmark.getName())) 

                    XWPFRun run = xwpfParagraph.createRun();
                    run.setText(dataMap.get(bookmark.getName()));

                    Node firstNode = bookmark.getDomNode();
                    Node nextNode = firstNode.getNextSibling();
                    while (nextNode != null) 
                        // 循环查找结束符
                        String nodeName = nextNode.getNodeName();
                        if (nodeName.equals(BOOKMARK_END_TAG)) 
                            break;
                        

                        // 删除中间的非结束节点，即删除原书签内容
                        Node delNode = nextNode;
                        nextNode = nextNode.getNextSibling();

                        ctp.getDomNode().removeChild(delNode);
                    

                    if (nextNode == null) 
                        // 始终找不到结束标识的，就在书签前面添加
                        ctp.getDomNode().insertBefore(run.getCTR().getDomNode(), firstNode);
                     else 
                        // 找到结束符，将新内容添加到结束符之前，即内容写入bookmark中间
                        ctp.getDomNode().insertBefore(run.getCTR().getDomNode(), nextNode);
                    
                
            
        

        document.write(outputStream);
        document.close();
    


    /**
     * doc 文件中书签的替换
     *
     * @param inputStream
     * @param outputStream
     * @param dataMap
     * @throws IOException
     */
    public static void replaceBookmarksByDoc(InputStream inputStream, OutputStream outputStream,
                                             Map<String, String> dataMap) throws IOException 
        HWPFDocument document = new HWPFDocument(inputStream);
        Bookmarks bookmarks = document.getBookmarks();
        for (int dwI = 0; dwI < bookmarks.getBookmarksCount(); dwI++) 
            Bookmark bookmark = bookmarks.getBookmark(dwI);
            if (dataMap.containsKey(bookmark.getName())) 
                Range range = new Range(bookmark.getStart(), bookmark.getEnd(), document);
                range.replaceText(dataMap.get(bookmark.getName()), false);
            
        
        document.write(outputStream);
    

---

（END）