libreOffice word 转 pdf

直接上结论

1、spire.doc.free 收费不考虑、前4页无水印 30s
2、docx4j 20s
3、liboffice 处理时间1.4s
4、documents4j较快只支持windows micro office

1 Windows 环境WORD 转换

dom4j 只能在windows环境调用microsoft office

引入jar包

      <dependency>
            <groupId>com.documents4j</groupId>
            <artifactId>documents4j-local</artifactId>
            <version>1.0.3</version>
        </dependency>

        <dependency>
            <groupId>com.documents4j</groupId>
            <artifactId>documents4j-transformer-msoffice-word</artifactId>
            <version>1.0.3</version>
        </dependency>

    public static void wordToPdf(MultipartFile file, String outFilePath) {
        File outputFile = new File(outFilePath);
        InputStream doc = null;
        OutputStream outputStream = null;
        try {
            doc = file.getInputStream();
            outputStream = Files.newOutputStream(outputFile.toPath(), StandardOpenOption.CREATE_NEW);
            IConverter converter = LocalConverter.builder().build();
            //转换docx=>pdf
            boolean flag = converter.convert(doc).as(DocumentType.DOC).to(outputStream).as(DocumentType.PDF).execute();
            if (flag) {
                converter.shutDown();
            }
            doc.close();
            outputStream.close();
            System.out.println("文件名：" + outFilePath + " 转换成功！");
        } catch (Exception e) {
            e.printStackTrace();
            throw new RuntimeException("文件名：" + outFilePath + " 转换失败！");
        }
    }

2 linux 环境WORD 转换

Dockerfile 配置

FROM openjdk:8-jdk-alpine

MAINTAINER auther<>
COPY start.sh /home/
COPY ./target/demo.jar /home/demo.jar

# 复制字体文件到容器的字体目录
COPY simhei.ttf /usr/share/fonts/

# 设置字体文件的权限
RUN chmod 655 /usr/share/fonts/simhei.ttf

ENV TZ=Asia/Shanghai
# 安装字体库，安装liboffice 
RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone
RUN echo -e 'https://mirrors.aliyun.com/alpine/v3.6/main/\nhttps://mirrors.aliyun.com/alpine/v3.6/community/' > /etc/apk/repositories \
 && apk update \
 && apk upgrade \
 && apk --no-cache add ttf-dejavu fontconfig libreoffice

# 刷新系统字体缓存
RUN fc-cache -fv


#无用处，liboffce调用的上面容器内的字体
ENV LANG en_US.UTF-8
ENV LANGUAGE en_US:en
ENV LC_ALL en_US.UTF-8

EXPOSE 8080
ENTRYPOINT ["/bin/sh", "/home/start.sh"]

调用方法

执行路径
libreOfficeCommand = "/usr/lib/libreoffice/program/soffice.bin"

   public static int poiWord2PDF(String source, String targetDir, String libreOfficeCommand) {
//   libreOfficeCommand Linux可执行文件的路径  source word完整路径  targeDir 则是PDF的存放目录，不带文件名称噢，文件名称会自动取source的文件名称作为名称
        String[] cmdArray = {
                libreOfficeCommand,
                "--headless",
                "--convert-to", "pdf:writer_pdf_Export",
                source,
                "--outdir", targetDir
        };
        int exitCode = 0;
        try {
            Process process = Runtime.getRuntime().exec(cmdArray);
            // simhei.ttf
            // 获取命令执行的输出
            InputStream inputStream = process.getInputStream();
            BufferedReader reader = new BufferedReader(new InputStreamReader(inputStream, StandardCharsets.UTF_8));
            String line;
            while ((line = reader.readLine()) != null) {
                System.out.println(line);
            }
            exitCode = process.waitFor();
            System.out.println(("转换码："+exitCode));
        } catch (Exception e) {
            System.out.println( "网络繁忙，请重试！");
        }
        return exitCode;
    }