- (1) java.version=1.1 environment only.
- Use InputStreamReader to read text files with Japanese native code.
- (2) 1.1 development environment (JDK1.1.x etc.) and 1.0.2 displaying environment (Netscape, MSIE etc.).
- Use Native2Utf.java to convert Japanese native code to UTF-8 (compacted unicode).
This tool itself never analyzes encoding owing to InputStreamReader.In cases of (2) and (3), you'll have UTF-8 files.
- (3) 1.0.2 environment only.
- First, use J2Uc (by Yasushi Kuno) to convert Japanese native code to unicode escape. J2Uc is almost compatible with native2ascii included with JDK 1.1.x.
- Second, use Ue2Utf.java to convert unicode escape to UTF-8.
UTF-8 is a flexible length format of unicode (UCS). See more detail of UTF-8 in DataInputStream on JDK 1.1.x API documentation .
readUTF() method of DataInputStream class produces Strings of ordinary fixed length unicode.
test102.utf
generated by Ue2Utf.java.
test11x.utf
generated by Native2Utf.java
ReadTest.java / DataInputStream2.java
Appearances on Sparc and WinNT
The difference between these display results is comes from difference of handling backslashes.
There is a bug in DataInputStream#readUTF(). DataInputStream2 is a modified class. (update 30/Jun/1997)