[17436] in Kerberos_V5_Development
Re: Future ASN.1 support
daemon@ATHENA.MIT.EDU (Nico Williams)
Fri Dec 9 16:14:15 2011
MIME-Version: 1.0
In-Reply-To: <1323454139.25965.7.camel@localhost>
Date: Fri, 9 Dec 2011 13:58:36 -0600
Message-ID: <CAK3OfOhLzRgj+XbWmRsfEsFPGzheXJqQAp8L7yQxkX4fie=wDA@mail.gmail.com>
From: Nico Williams <nico@cryptonector.com>
To: Nathaniel McCallum <npmccallum@redhat.com>
Cc: krbdev@mit.edu
Content-Type: text/plain; charset="utf-8"
Errors-To: krbdev-bounces@mit.edu
Content-Transfer-Encoding: 8bit
On Fri, Dec 9, 2011 at 12:08 PM, Nathaniel McCallum
<npmccallum@redhat.com> wrote:
>> > By "migrating data types" I meant basically option 1. I don't like
>> > option 2 as it seems to imply a fork of heimdal's libasn1. I'd be fine
>> > with that option so long as heimdal agreed to spin it off as a separate
>> > project.
>>
>> Why would it be a fork? Why would changes not be contributed and accepted?
>
> I would hope they would be.
I can't speak for the upstream, but I don't see why they wouldn't be,
particularly if there's no impact on footprint nor performance for
built Heimdal itself (and i do think that can be arranged).
I would like to see Heimdal eventually support the ASN.1 compiler as a
separate project. That's slightly complicated by the compiler's and
run-time's dependency on libroken and any future dependency on, say,
libheimbase (but I doubt this). The dependency on libroken is
probably not a big deal at all (since libroken mostly has
implementations of functions that the OS should provide). It'd
probably be easier to keep the ASN.1 compiler in the Heimdal tree and
just offer an option to build/install only that, or for MIT to package
it separately from Heimdal, or for MIT to have a process to copy
Heimdal's compiler sources into the MIT tree (to keep it up to date
yet separate but without having to effect the separation in the
upstream tree).
The best argument for a new compiler, IMO, is that having more open
source ASN.1 compilers is a good thing. But improving an existing
open source compiler would be at least equally good.
Nico
--
_______________________________________________
krbdev mailing list krbdev@mit.edu
https://mailman.mit.edu/mailman/listinfo/krbdev
