The following test program gives different results with gcj and Sun JDK:
// A.java
import java.nio.charset.Charset;
import java.nio.charset.CharsetEncoder;
public class A {
public static void main(String[] args) throws java.io.IOException
{
CharsetEncoder enc = Charset.forName("US-ASCII").newEncoder();
System.out.println(enc.canEncode('\u00e4'));
}
}
// end
$ javac A.java
$ /usr/lib/jvm/java-1.5.0-sun/bin/java A
false
$ /usr/lib/jvm/java-gcj/bin/java A
true
The Sun version seems to return something like (c > 31 && c < 127),
which makes sense, whereas gcj apparently always returns true.
This breaks lots of code, such as Apache JaxMe 2 version 0.51
(specifically the MarshallerTest in the test suite, which generated
invalid XML with gcj).
This was tested on Debian (testing, i386).
~$ gcj --version
gcj (GCC) 4.1.2 20060901 (prerelease) (Debian 4.1.1-13)
The bug was also reported to Debian's bug tracking system:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=388596
--
Summary: CharsetEncoder.canEncode() gives different results than
Sun version
Product: gcc
Version: 4.1.2
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: libgcj
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: marcus at better dot se
GCC host triplet: i486-linux-gnu
GCC target triplet: i486-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29178
