Name: skT45625 Date: 10/27/2000
java version "1.3.0"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.3.0-C)
Java HotSpot(TM) Server VM (build 2.0fcs-E, mixed mode)
The java.util.ResourceBundle class and subclasses, as well as the
java.util.Properties class are supposed to work with ISO-8859-1
encoded data (including UNICODE encoded using the \uxxxx format).
In practice, it encodes everything that is not printable US-ASCII
or a small set of control characters into the UNICODE \uxxxx
format.
The \uxxxx format is painful to work with, and next to impossible
to customize or localize. You'll notice this flies in the face
of the purpose of these classes.
Now, if the java.util.Properties class could be passed an
encoding to load the data with (as described in the java.lang
package.html), it would make it much easier for people to work
directly in UNICODE or other character encodings (espcially
UNICODE). Similarly, the java.util.ResourceBundle class, or
rather the java.util.PropertyResourceBundle class, should accept
an encoding for .properties files.
This means customization can be done in the native language by
natives in UNICODE. And the results further edited and saved as
UNICODE. All the actual functionality to do this is in
java.io.InputStreamReader and java.io.OutputStreamWriter. If
the java.util.Properties accepted a java.io.Reader or
java.io.Writer, you would be most of the way to
internationalization.
(Review ID: 109597)
======================================================================
- duplicates
-
JDK-4035543 Properties.load() doesn't work correctly against multi-byte character strings.
-
- Closed
-