java解析word文档

依赖

      <dependency>
            <groupId>org.apache.poi</groupId>
            <artifactId>poi</artifactId>
            <version>3.16</version>
        </dependency>

        <dependency>
            <groupId>org.apache.poi</groupId>
            <artifactId>poi-ooxml</artifactId>
            <version>3.16</version>
        </dependency>

        <dependency>
            <groupId>org.apache.poi</groupId>
            <artifactId>poi-scratchpad</artifactId>
            <version>3.16</version>
        </dependency>

        <dependency>
            <groupId>org.apache.xmlbeans</groupId>
            <artifactId>xmlbeans</artifactId>
            <version>2.6.0</version>
        </dependency>

        <dependency>
            <groupId>org.apache.commons</groupId>
            <artifactId>commons-collections4</artifactId>
            <version>4.1</version>
        </dependency>

只需要前面两个依赖就好了

实现

/**
 * @author : xujp0818
 * @date : 2021/7/3 20:49
 */
@RestController
public class WordAction {
    @Autowired
    private WordContextReadService wordContextReadService;
    /**
     * word文档上传解析
     *
     * @return
     */
    @PostMapping("/wordUpload")
    public String wordUpload(@RequestParam("wordUploader") MultipartFile mFile) {
        OutputStream os = null;
        try {
            // 段落
            int word = 0;
            // 总段落字数
            int wordNum = 0;
            // 图片
            int picture = 0;
            // 视频
            int video = 0;
            String buffer = "";
            XWPFDocument xwpfDocument = new XWPFDocument(mFile.getInputStream());
            POIXMLTextExtractor ex = new XWPFWordExtractor(xwpfDocument);
            buffer = ex.getText();
            ex.close();
            List<String> stringList = DocUtil.pharaNum(buffer);
            if (!CollectionUtils.isEmpty(stringList)) {
                word = stringList.size();
            }
            String wordString = String.join("", stringList);
            wordNum = wordString.length();
            List<XWPFParagraph> paragraphList = xwpfDocument.getParagraphs();

            List<String> picList = new ArrayList<>();
            // 统计素材元素个数
            for (int i = 0; i < paragraphList.size(); i++) {
                List<XWPFRun> runList = paragraphList.get(i).getRuns();
                for (XWPFRun run : runList) {
                    CTR ctr = run.getCTR();
                    String params = ctr.toString();
                    if (params.contains("a:graphicData")) {
                        picture++;
                        List<String> pictureIds = wordContextReadService
                                .readImageInParagraph(paragraphList.get(i));
                        if (CollectionUtils
                                .isNotEmpty(pictureIds)) {
                            for (String pictureId : pictureIds) {
                                picList.add(pictureId);
                            }
                        }
                    } else if (params.contains("v:imagedata")) {
                        video++;

                    }
                }
            }

            int count = 0;

            int aa = 0;
            for (String s : stringList) {
                if (StringUtils.isNotBlank(s)) {
                    if (s.contains("视频地址:")) {
                        System.out.println("视频："+s);
                    }
                } else {
                    String picId = picList.get(aa);
                    XWPFPictureData pictureData = xwpfDocument
                            .getPictureDataByID(picId);
                    byte[] data = pictureData.getData();
                    aa++;
                   //将该图片的字节数组保存到本地的一张图片
                    wordContextReadService.write(data,"e:/test/"+System.currentTimeMillis()+".png");
                }
            }
            System.out.println("文字字数："+wordNum+",图片："+picture+",视频："+video);
        } catch (Exception e) {
            e.printStackTrace();
            return "上传失败";
        }

        return "上传成功";
    }
}