Description Yoshito Umaoka 2017-03-09 18:16:57 EST

When you run a code printing out multi-byte UTF-8 characters, the Unicode replacement character '�' (U+FFFD) is intermittently inserted in the console output.

Steps to reproduce this issue 

1. Creates a new Java project
2. Sets "Text file encoding" to UTF-8 in project Properties - Resource
3. Creates class - as below:

-----
package com.ibm.test;

public class TestConsole {
	public static void main(String... args) {
		StringBuilder buf = new StringBuilder();
		for (int j = 0; j < 50; j++) {
			buf.append((char)'\uFF11');
		}
		String s = buf.toString();

		for (int i = 0; i < 1000; i++) {
			System.out.println(s);
		}
	}
}
-----

The output prints out 50 FULLWIDTH DIGIT ONE '１'(U+FF11 / UTF-8: EF-BC-91) per line, repeats 1000 times.

The valid output is:

１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１

However, REPLACEMENT CHARACTER '�'(U+FFFD / UTF-8: EF-BF-BD) is inserted intermittently. For example:

１１１１１１１１１１１１１１１１１１１１１１１１１１��１１１１１１１１１１１１１１１１１１１１１１１
１���１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１１


This problem is reproduced on all platforms I tried, including Windows 7 and Ubuntu 16.04LTS.