Re: classpath ./ChangeLog ./THANKYOU ./configure.in...

[Brian Jones]

"Raif S. Naffah" <address@hidden> writes:

> On Saturday 29 March 2003 18:27, Brian Jones wrote:
> > Tom Tromey <address@hidden> writes:
> > > >>>>> "Brian" == Brian Jones <address@hidden> writes:
> > >
> > > Brian> There are still some getInstance problems with some of the
> > > Brian> things I recently checked in though a check through my Mauve
> > > Brian> results appears to indicate nothing terribly new here.
> > >
> > > Are these getInstance problems revealed by Mauve?
> > > Or did you find them some other way?
> > > I'd like to put all this java.security stuff into libgcj, but I'd
> > > also prefer to wait if there is a regression that will be ironed
> > > out soon...
> >
> > I looked at the Mauve results before and after and essentially the
> > same getInstance problems and I don't know much about it other than
> > that.  It could just be because I don't have gnu-crypto configured
> > with the mauve testing.  Need to investigate.
> 
> the Mauve tests for the getInstance() methods (at least those i added 
> recently) do not rely/depend on GNU Crypto.  all the testcases, install 
> their own, temporary, providers.
> 
> to test the patch i submitted earlier, i used my local copy of gcc, to 
> which i applied the same patch.  all the security related Mauve tests 
> passed.

It's broken under Kissme/Classpath.

Brian
-- 
Brian Jones <address@hidden>