- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Thu, 21 May 2020 11:01:12 +0200
- To: Magnus Westerlund <magnus.westerlund@ericsson.com>, "mnot@mnot.net" <mnot@mnot.net>
- Cc: "httpbis-chairs@ietf.org" <httpbis-chairs@ietf.org>, "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>, "draft-ietf-httpbis-header-structure@ietf.org" <draft-ietf-httpbis-header-structure@ietf.org>, "iesg@ietf.org" <iesg@ietf.org>, "tpauly@apple.com" <tpauly@apple.com>
On 21.05.2020 10:42, Magnus Westerlund wrote: > Hi Mark, > > To be clear do what ever the you think are best. From my perspective restricting > this to space is a more strict and gets things done. However, calling space only > white space would be inconsistent which is why I raised this. > > Establish consensus and when it is done and a new version is availble I am happy > to clear my discuss. > > Regarding my comment. I do understand that there are praxis here. And it is > surprising how some of that praxis has been diverging between the protocols > being based on the same original specification practice. > > Cheers > > Magnus Westerlund > ... FWIW, Mark is right that we need to deal with HTAB on the wire, because the HTTP spec allows it to be inserted. The alternative solution would be to normalize HTAB before parsing, but that would be tricky as we really want to allow it between list elements. Best regards, Julian
Received on Thursday, 21 May 2020 09:01:40 UTC