[Mailman-Developers] Parsing and Rendering rfc8222

[Barry Warsaw]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Jul 5, 2006, at 12:30 AM, Mark Sapiro wrote:

> Also there is a related issue if A posts, B replies, A replies off  
> list
> to B, and B replies on list. If threading relies solely on References:
> or In-Reply-To:, and either A's or B's MUA generates only In-Reply-To,
> this thread is broken at the 'missing' post. I don't have any really
> good suggestions for alternative threading algorithms however. I think
> there was something on this not too long ago on mailman-users or maybe
> mailman-developers - I looked and found what I think I remember. The
> relevant post is at
> <http://mail.python.org/pipermail/mailman-developers/2005-January/ 
> 017660.html>
> and points to a description of an algorithm at
> <http://www.jwz.org/doc/threading.html>.

I haven't read JWZ's article in a while, but IIRC, it lays out an  
algorithm that does about as good as you can do in these cases.

BTW, my earlier post re: X-List-Message-ID isn't about threading,  
it's about message identification and trying to make that as robust  
as possible.

- -Barry

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (Darwin)

iQCVAwUBRKxM3XEjvBPtnXfVAQJUhQP/ZXeVZhPYuXMecIMTO6QiPncRuHMZiSAA
rpGeYNspQkszVbsLB27nkppfShkJQC8iSMY8WPEssGUJ4guqROtCdJ7hUURUve5D
4uvAizTrl6Kriv82SZPCmUvN12XnfkT5+ENIkwpWEnh5xf4qoQXNqXfAqWX83Y/Y
ldDwY+zkltY=
=Nhwt
-----END PGP SIGNATURE-----