博客

Java中，String有很多个构造方法，其中有三个如下图所示：

人们习惯性喜欢使用没有字符集的构造方法，因为简单！而我也正是犯了同样的错误！

先贴出代码：

System.out.println("````````````````````````````````````" + new String(userQues.getBytes("GBK")));
System.out.println("````````````````````````````````````" + new String(Base64.decodeBase64(userQues.getBytes("GBK"))));
System.out.println("````````````````````````````````````" + Base64.encodeBase64String(new String(Base64.decodeBase64(userQues.getBytes("GBK"))).getBytes("GBK")));

这段代码在本地windows环境下运行正常，结果如图：

但是部署到linux环境后，出现结果却很奇怪：

这说明第一次new String()的时候，字符集就出现了问题，导致base64解码失败，进一步编码也就和初始编码字符串不一致了！

于是查看两个系统的编码集结果如下

Windows：



936代表的编码格式就是GBK

Linux：



linux的系统编码为UTF-8

问题查出来就好解决了！

对代码进一步优化，减少对系统编码的依赖：

System.out.println("````````````````````````````````````" + new String(userQues.getBytes("GBK")));
System.out.println("````````````````````````````````````" + new String(Base64.decodeBase64(userQues.getBytes("GBK")), "GBK"));
System.out.println("````````````````````````````````````" + Base64.encodeBase64String(new String(Base64.decodeBase64(userQues.getBytes("GBK")), "GBK").getBytes("GBK")));

问题完美解决 ！