前言

最近做的一个项目有个word导入的功能，一开始做的是导入纯文本，然后显示在前端页面，客户提出了建议，能不能改成和word文件里面格式一样的。

推荐

1、poi

2、open office

3、libreoffice

poi

pom.xml

org.apache.poi

poi-scratchpad

3.14

org.apache.poi

poi-ooxml

3.14

fr.opensagres.xdocreport

xdocreport

1.0.6

org.apache.poi

poi-ooxml-schemas

3.14

org.apache.poi

ooxml-schemas

1.3

word读取

/**

* 将word2003转换为html文件

*

* @param wordPath word文件路径

* @param wordName word文件名称无后缀

* @param suffix word文件后缀

* @param htmlPath html存储地址

* @throws IOException

* @throws TransformerException

* @throws ParserConfigurationException

*/

public static String Word2003ToHtml(String wordPath, String wordName, String suffix, String htmlPath)

throws IOException, TransformerException, ParserConfigurationException {

String htmlName = wordName + ".html";

final String imagePath = htmlPath + "image" + File.separator;

// 判断html文件是否存在

File htmlFile = new File(htmlPath + htmlName);

if (htmlFile.exists()) {

return htmlFile.getAbsolutePath();

}

// 原word文档

final String file = wordPath + File.separator + wordName + suffix;

InputStream input = new FileInputStream(new File(file));

HWPFDocument wordDocument = new HWPFDocument(input);

WordToHtmlConverter wordToHtmlConverter = new WordToHtmlConverter(

DocumentBuilderFactory.newInstance().newDocumentBuilder().newDocument());

// 设置图片存放的位置

wordToHtmlConverter.setPicturesManager(new PicturesManager() {

public String savePicture(byte[] content, PictureType pictureType, String suggestedName, float widthInches,

float heightInches) {

File imgPath = new File(imagePath);

// 图片目录不存在则创建

if (!imgPath.exists()) {

imgPath.mkdirs();

}

File file = new File(imagePath + suggestedName);

try {

OutputStream os = new FileOutputStream(file);

os.write(content);

os.close();

} catch (FileNotFoundException e) {

e.printStackTrace();

} catch (IOException e) {

e.printStackTrace();

}

// 图片在html文件上的路径 相对路径

return "image/" + suggestedName;

}

});

// 解析word文档

wordToHtmlConverter.processDocument(wordDocument);

Document htmlDocument = wordToHtmlConverter.getDocument();

// 生成html文件上级文件夹

File folder = new File(htmlPath);

if (!folder.exists()) {

folder.mkdirs();

}

OutputStream outStream = new FileOutputStream(htmlFile);

DOMSource domSource = new DOMSource(htmlDocument);

StreamResult streamResult = new StreamResult(outStream);

TransformerFactory factory = TransformerFactory.newInstance();

Transformer serializer = factory.newTransformer();

serializer.setOutputProperty(OutputKeys.ENCODING, "GB2312");

serializer.setOutputProperty(OutputKeys.INDENT, "yes");

serializer.setOutputProperty(OutputKeys.METHOD, "html");

serializer.transform(domSource, streamResult);

return htmlFile.getAbsolutePath();

}

/**

*

* 2007版本word转换成html

*

* @param wordPath word文件路径

* @param wordName word文件名称无后缀

* @param suffix word文件后缀

* @param htmlPath html存储地址

* @return

* @throws IOException

*/

public static String word2007ToHtml(String wordPath, String wordName, String suffix, String htmlPath)

throws IOException {

String htmlName = wordName + ".html";

String imagePath = htmlPath + "image" + File.separator;

// 判断html文件是否存在

File htmlFile = new File(htmlPath + htmlName);

if (htmlFile.exists()) {

return htmlFile.getAbsolutePath();

}

// word文件

File wordFile = new File(wordPath + File.separator + wordName + suffix);

// 1) 加载word文档生成 XWPFDocument对象

InputStream in = new FileInputStream(wordFile);

XWPFDocument document = new XWPFDocument(in);

// 2) 解析 XHTML配置 (这里设置IURIResolver来设置图片存放的目录)

File imgFolder = new File(imagePath);

XHTMLOptions options = XHTMLOptions.create();

options.setExtractor(new FileImageExtractor(imgFolder));

// html中图片的路径 相对路径

options.URIResolver(new BasicURIResolver("image"));

options.setIgnoreStylesIfUnused(false);

options.setFragment(true);

// 3) 将 XWPFDocument转换成XHTML

// 生成html文件上级文件夹

File folder = new File(htmlPath);

if (!folder.exists()) {

folder.mkdirs();

}

OutputStream out = new FileOutputStream(htmlFile);

XHTMLConverter.getInstance().convert(document, out, options);

return htmlFile.getAbsolutePath();

}

open office

linux 一键安装脚本

cd /tmp

install_redhat() {

wget https://kkfileview.keking.cn/Apache_OpenOffice_4.1.6_Linux_x86-64_install-rpm_zh-CN.tar.gz -cO openoffice_rpm.tar.gz && tar zxf /tmp/openoffice_rpm.tar.gz && cd /tmp/zh-CN/RPMS

if [ $? -eq 0 ];then

yum install -y libXext.x86_64

yum groupinstall -y "X Window System"

rpm -Uvih *.rpm

echo 'install desktop service ...'

rpm -Uvih desktop-integration/openoffice4.1.6-redhat-menus-4.1.6-9790.noarch.rpm

echo 'install finshed...'

else

echo 'download package error...'

fi

}

install_ubuntu() {

wget https://kkfileview.keking.cn/Apache_OpenOffice_4.1.6_Linux_x86-64_install-deb_zh-CN.tar.gz -cO openoffice_deb.tar.gz && tar zxf /tmp/openoffice_deb.tar.gz && cd /tmp/zh-CN/DEBS

echo $?

if [ $? -eq 0 ];then

apt-get install -y libxrender1

apt-get install -y libxt6

apt-get install -y libxext-dev

apt-get install -y libfreetype6-dev

dpkg -i *.deb

echo 'install desktop service ...'

dpkg -i desktop-integration/openoffice4.1-debian-menus_4.1.6-9790_all.deb

echo 'install finshed...'

else

echo 'download package error...'

fi

}

if [ -f "/etc/redhat-release" ]; then

yum install -y wget

install_redhat

else

apt-get install -y wget

install_ubuntu

安装OpenOffice，安装结束后，调用命令行，启动OpenOffice的一项服务：

soffice -headless -accept="socket,port=8100;urp;"

测试 pom.xml

com.artofsolving

jodconverter

2.2.1

log4j

log4j

1.2.17

org.slf4j

slf4j-api

1.7.25

org.slf4j

slf4j-log4j12

1.7.25

test

org.slf4j

slf4j-simple

1.7.25

test

public class OfficeConverter {

public static void main(String[] args) {

File inputFile = new File("C:/test/yy.doc");

File outputFile = new File("C:/test/yy.html");

OpenOfficeConnection con = new SocketOpenOfficeConnection(8100);

try {

con.connect();

} catch (ConnectException e) {

System.err.println("文件转换出错，请检查OpenOffice服务是否启动。");

e.printStackTrace();

}

DocumentConverter converter = new OpenOfficeDocumentConverter(con);

converter.convert(inputFile, outputFile);

con.disconnect();

}

}

libreoffice

安装

yum install libreoffice libreoffice-headless

word 转 html

soffice --headless --convert-to html:HTML test.docx

poi、open office、libreoffice的区别

poi纯java转换,转换比较慢 openoffice小文件转换速度快，libreoffice转换较慢，大文件libreoffice转换快， openoffice只支持java代码操作，libreoffice支持命令行和java代码， 最重要一点openoffice支持队列转换，libreoffice不支持，同时转换一个文件会出现soffice卡死，

参考

来源：oschina

链接：https://my.oschina.net/u/3568600/blog/4411071