Background:
After one customer upgraded from 5u17, garbled text occurred on their application where "UNICODE" was used.
Then they found it was caused by the fix of JDK-6292322.

The behavior of String#getBytes("UNICODE") was changed by the fix.
before: return UTF-16 of Litle Endian with BOM
after : return UTF-16 of Big Endian with BOM


This is an enhancement request that requires Sustaining resolution.

Exception is not raised when an invalid name "UNICODE" is used in
java.lang.String#getBytes. However, "UNICODE" is not supported
according to the page below:

http://docs.oracle.com/javase/1.5.0/docs/guide/intl/encoding.doc.html

The test case Test.java

class Test{
  public static void main(String[] args) throws
  java.io.UnsupportedEncodingException{
  System.out.println("Hello".getBytes("UNICODE"));
  }
}
 
To reproduce:
1. jdk8u25/bin/javac Test.java
2. jdk8u25/bin/java Test
 
Suggested Fix:
  Does strict check on API unsupported parameters
  Make noticeable announcement if incompatibility change is made