- From: Willy Tarreau <w@1wt.eu>
- Date: Mon, 17 Oct 2016 22:58:11 +0200
- To: Poul-Henning Kamp <phk@phk.freebsd.dk>
- Cc: Matt Menke <mmenke@google.com>, Martin Thomson <martin.thomson@gmail.com>, HTTP Working Group <ietf-http-wg@w3.org>
On Mon, Oct 17, 2016 at 08:50:38PM +0000, Poul-Henning Kamp wrote: > -------- > In message <CAEK7mvoXNf_KdpQhj2kwiRLaHm-C1BtjduUgy1cwOnCQ265yzg@mail.gmail.com> > , Matt Menke writes: > > >I don't think it needs to be that complicated. If there's just a prefix > >before all types (Say '#' for numbers, '&' for times, and '!' for > >identifiers), we don't have to exclude the prefix characters from > >identifiers, do we? > > True, if we prefix all values we can avoid that problem. > > However, that would kill one of the major attraction of CS: That > 19 RFC723x headers already conform to it. This is actually a very good point that we constantly need to keep in mind and which is the best justification for not using prefixes on field names. Willy
Received on Monday, 17 October 2016 20:59:28 UTC