This is the mail archive of the mauve-discuss@sourceware.cygnus.com mailing list for the Mauve project. See the Mauve home page for more information.


[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index] [Subject Index] [Author Index] [Thread Index]

Re: java.lang.Character testing



>> You probably know this, but Java's use of Unicode varies from
>> release to release.  So make sure to mark the tests appropriately
>> ...

Artur> Yes, I know, but I'm not sure if it is java standard or just
Artur> sun's choice.

That's an interesting point which is more general than just the
Character class.  I guess it's because the documentation isn't very
good that this confusion arises.

Artur> JDK x.y uses unicode standard z. When new unicode update
Artur> appears, new JDK uses it (1.2 vs euro for example). But I don't
Artur> think that java standard specify which unicode you should use -
Artur> and if it specify, it has to be set, not vary from version to
Artur> version.

Yeah.  However, JDK 1.0 explicitly uses Unicode 1.1.5 (this info is in
the Java Language Spec book).  So your tests should probably tagged
JDK1.1.

Artur> I'm goping to use latest possible unicode data. If some lib do
Artur> not implement it, then lib is out of date. Only thing that I
Artur> think could be done is to change some errors into warnings -
Artur> for example "'0xabcd' is defined since unicode z.y.z"

I guess this would be nice if it isn't too hard to do.  I'm not really
overly concerned with it.  We generate the data for Character directly
from the Unicode data tables.  For any given release of our package,
we'll probably similarly fix a version of the Mauve test suite.  So
when a new Unicode is released, we can just update them both at the
same time.

I'd imagine this is the case for all implementors.  Anybody do
something different?

Tom