使用POI解析doc/docx格式文件并转化HTML，替换其中的图片

一、简介

“Apache POI”是一个用于操作Microsoft Office格式文件（如Word文档、Excel表格、PowerPoint演示文稿）的开源Java库。POI允许开发人员读取和写入这些Office文件，使得Java应用程序能够与Microsoft Office文件进行交互。
这里介绍一下POI解析doc与docx文件的demo。

二、doc文档解析

private String parseDoc(String fileUrl, String savePath, String encode) throws Exception{
        ByteArrayOutputStream out = null;
        try {
            HWPFDocument wordDocument = new HWPFDocument(new FileInputStream(savePath));
            WordToHtmlConverter wordToHtmlConverter = new WordToHtmlConverter(
                    DocumentBuilderFactory.newInstance().newDocumentBuilder()
                            .newDocument());
            wordToHtmlConverter.setPicturesManager( new PicturesManager() {
                // 此处返回的图片的路径是html中图片的访问路径
                @Override
                public String savePicture(byte[] content,
                                          PictureType pictureType, String suggestedName,
                                          float widthInches, float heightInches )
                {
                    //  此处将图片保存到本地，用于后续上传到nos中
                    return PIC_SAVE_PATH + "/" + suggestedName;
                }
            });
            wordToHtmlConverter.processDocument(wordDocument);
            List pics=wordDocument.getPicturesTable().getAllPictures();
            if(pics!=null){
                for(int i=0;i<pics.size();i++){
                    Picture pic = (Picture)pics.get(i);
                    pic.writeImageContent(new FileOutputStream(PIC_SAVE_PATH  + "/" + pic.suggestFullFileName()));
                }
            }
            Document htmlDocument = wordToHtmlConverter.getDocument();
            out = new ByteArrayOutputStream();
            DOMSource domSource = new DOMSource(htmlDocument);
            StreamResult streamResult = new StreamResult(out);

            TransformerFactory tf = TransformerFactory.newInstance();
            Transformer serializer = tf.newTransformer();
            serializer.setOutputProperty(OutputKeys.ENCODING, encode);
            serializer.setOutputProperty(OutputKeys.INDENT, "yes");
            serializer.setOutputProperty(OutputKeys.METHOD, "html");
            serializer.transform(domSource, streamResult);
            // 替换图片路径
            String htmlContent = replacedPicContent(new String(out.toByteArray()));
            return htmlContent;
        } catch (Exception e) {
            logger.info("traverse Doc fail. fileUrl:{}, savePath:{}, stackTrace:{}", fileUrl, savePath, e.getStackTrace());
            return "";
        } finally {
            IOUtils.close(out);
        }
    }

其中的PIC_SAVE_PATH是我们自己定义的本地图片保存路径。

三、docx文档解析

private String parseDocx(String fileUrl, String savePath, String encode) throws Exception {
        String redisKey = DigestUtils.md5Hex(fileUrl);
        String parsingPath = FILE_PARSING_RES_PATH + "/" + redisKey + ".html";
        FileInputStream fis = null;
        OutputStream htmlOutputStream = null;
        try {
            fis = new FileInputStream(new File(savePath));
            XWPFDocument document = new XWPFDocument(fis);

            // 提取并保存图片
            for (XWPFPictureData picture : document.getAllPictures()) {
                byte[] pictureData = picture.getData();
                String imageFileName = picture.getFileName();
                FileOutputStream imageOutFile = new FileOutputStream(PIC_SAVE_PATH + "/" + imageFileName);
                imageOutFile.write(pictureData);
                imageOutFile.close();
            }

            // 将文档内容转换为HTML
            File htmlOutputFile = new File(parsingPath);
            htmlOutputStream = new FileOutputStream(htmlOutputFile);
            XHTMLConverter.getInstance().convert(document, htmlOutputStream, null);

            // 修改图片路径，默认会存在/word/media/下，我们将其修改为固定路径
            String htmlContent = readFileToString(htmlOutputFile, encode);
            htmlContent = htmlContent.replaceAll("src=\"word/media", "src=\"" + PIC_SAVE_PATH);
            String resContent = replacedPicContent(htmlContent);
            return resContent;
        } catch (Exception e) {
            logger.info("traverse Docx fail. fileUrl:{}, savePath:{}, stackTrace:{}", fileUrl, savePath, e.getStackTrace());
            return "";
        } finally {
            IOUtils.close(htmlOutputStream);
            IOUtils.close(fis);
        }
    }

四、图片路径替换

这里由于业务需要，将存在本地的图片需要上传到NOS中，因此这里案例是替换了nos的地址，具体的需求具体分析。

// 需要html中图片的正则
public static final String PIC_REG= "<img\\s+src=\"([^\"]+)\"";

private static String replacedPicContent(String orginHtmlContent) {
        // 使用正则表达式提取img标签中的src属性值
        Pattern pattern = Pattern.compile(PIC_REG);
        Matcher matcher = pattern.matcher(orginHtmlContent);
        StringBuffer result = new StringBuffer();
        // 每找到一张图片，则上传到nos，并替换html中的图片路径为nos路径地址
        while (matcher.find()){
            String fileName = matcher.group(1);
            // 获取本地图片路径
            File file = new File(fileName);
            String nosUrl = "";
            try {
                // 图片上传后返回的nos地址，用于替换html中的图片路径
                nosUrl = uploadPic(fileName, file);
            } catch (Exception e) {
                throw new RuntimeException(e);
            }
            matcher.appendReplacement(result, "<img src=\"" + nosUrl + "\"");
        }
        // 将替换后的内容追加到result中
        matcher.appendTail(result);

        orginHtmlContent = result.toString();
        return orginHtmlContent;
    }

以上就解决了doc/docx文件转化为HTML的需求。具体效果如下：

可以看到基本的格式还都是保留的。效果还不错～