Re: [Japi] More lazy ClassFile parsing checked in

[Stuart Ballard]

Last night's nightly runs produced a bunch of diffs that I don't have
a good explanation for - and don't seem correct - which makes me
suspicious that this change had some unintended side-effects. Could
you investigate whether this patch has caused any change to the japi
files that result from japizing the JDK or of the glibj-latest.zip or
glibj-generics-latest.zip on builder?

If so, could you either fix the problem or temporarily back out this
change until a release can be made?

Thanks,
Stuart.

On 11/13/06, Stuart Ballard <address@hidden> wrote:
> Fantastic! Thanks Jeroen.
>
> Sorry I didn't get to make the release when I intended to. Gonna have
> another try tonight - picking a night when I didn't have plans would
> have been a better choice in the first place, no? ;)
>
> Stuart.
>
> On 11/13/06, Jeroen Frijters <address@hidden> wrote:
> > Hi,
> >
> > I've checked in the first version of lazy class file parsing. This
> > splits the parsing in two. The first part reads just enough to ascertain
> > whether a class is public or not, the second part reads the rest. In the
> > future it may prove worthwhile to split it into more parts, but for the
> > time being this should at least solve the exceptions that we saw with
> > IBM's and BEA's libraries.
> >
> > Regards,
> > Jeroen
> >
> >
> > _______________________________________________
> > Japitools-list mailing list
> > address@hidden
> > http://lists.nongnu.org/mailman/listinfo/japitools-list
> >
>
>
> --
> http://sab39.netreach.com/


--
http://sab39.netreach.com/