Question

I put one row  with a jpg file stored to a column into hbase,but when
 I get the row from hbase,I find the file size is not the same with the
file that I put into hbase.why?

My hbase is in centos 64,my code and the jpg file are  in
macos.

lovedeMacBook-Pro:~ love$ ls -lh  /Users/love/Documents/vpn*

-rw-r-----@ 1 love  staff   8.5K  8 24 09:59 /Users/love/Documents/vpn.jpg

lovedeMacBook-Pro:~ love$ ls -lh 
 /Users/love/output.jpg

-rw-r--r--@ 1 love  staff   8.0K  8 24 13:41 /Users/love/output.jpg

我的代码是：

    public class Hadoop_test {
    //function to convert image file to bytes.
    public static byte[] extractBytes(String ImageName) throws IOException {

        File file = new File(ImageName);
        BufferedImage originalImage = ImageIO.read(file);
        ByteArrayOutputStream baos = new ByteArrayOutputStream();
        ImageIO.write(originalImage, "jpg", baos);
        byte[] imageInByte = baos.toByteArray();
        return imageInByte;
    }

    public static void main(String[] args) throws Exception{
        Configuration conf = HBaseConfiguration.create();
        String krbstr = Thread.currentThread().getContextClassLoader().getResource("krb.conf").getFile();
        System.setProperty("java.security.krb5.conf", krbstr);

        UserGroupInformation.setConfiguration(conf);

        String keytab = Thread.currentThread().getContextClassLoader().getResource("love.keytab").getFile();
        UserGroupInformation.loginUserFromKeytab("love/lovedeMacBook-Pro.local@for_hadoop", keytab);

        TableName table_name = null;
        Admin admin = null;
        Connection connection = null;

        connection = ConnectionFactory.createConnection(conf);
        admin = connection.getAdmin();
        table_name = TableName.valueOf("users");

        if(admin.tableExists(table_name)){
            System.out.println("table exist");
        }

        Put put = new Put("row3".getBytes());
        put.addColumn("info".getBytes(), "image".getBytes(), extractBytes("/Users/love/Documents/vpn.jpg"));

        connection.getTable(TableName.valueOf(table_name.getName())).put(put);

        //get image from hbase
        Get get = new Get(Bytes.toBytes("row3"));
        Result result = connection.getTable(TableName.valueOf(table_name.getName())).get(get);
        byte[] arr = result.getValue("info".getBytes(), "image".getBytes());
        OutputStream out = new BufferedOutputStream(new FileOutputStream(
                "/Users/love/output.jpg"));
        out.write(arr);
        out.close();
    }
}

Answer 1

尝试将方法extractBytes替换为：

import org.apache.commons.io.IOUtils;
...
public static byte[] extractBytes(String ImageName) throws IOException {
    return IOUtils.toByteArray(new FileInputStream(ImageName));
}

原始图像不会进行解码/编码，因此输出文件的大小应与输入的大小相同。

为什么我将相同的列放入或从hbase中获取，但获得不同的列数据大小

1 个答案: