On Sun, 2005-09-18 at 12:13 +0200, Mark Wielaard wrote: > > The way I read it is if both conditions fail then our library is busted > > and we should throw a runtime exception, and not the expected > > NoSuchAlgorithmException. That would just mask a problem. > > Our library or the given Provider.
Our library. > I would still throw the > NoSuchAlgorithmException either with a chained exception cause or an > appropriate message. But I don't feel too strongly about it. The > NullPointerException was the worst thing we could throw since that > doesn't really give the user any information what went wrong. Even NullPointerException is preferable to NoSuchAlgorithmException. User applications will be written to expect NoSuchAlgorithmException so nobody will ever know that our library is broken. AG _______________________________________________ Classpath-patches mailing list Classpath-patches@gnu.org http://lists.gnu.org/mailman/listinfo/classpath-patches
