- From: Martin J. Dürst <duerst@it.aoyama.ac.jp>
- Date: Wed, 14 Aug 2013 22:23:40 +0900
- To: James M Snell <jasnell@gmail.com>
- CC: "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
For header names, I think ASCII is enough. Please note that in e-mail, headers are ASCII but can contain virtually any kind of symbol character (see http://tools.ietf.org/html/rfc5322#section-3.6.8). But as this is more limited in HTTP 1.1, this is fine. Regards, Martin. On 2013/08/14 7:08, James M Snell wrote: > https://github.com/http2/http2-spec/issues/231 > > The current header compression spec allows for UTF-8 encoded header > names without any character restrictions. The main HTTP/2 spec, > however, states that header names are "strings of ASCII characters" > (also without specifying any character restrictions). We need to be > clearer. > > Recommend that we specify in both the HTTP/2 and Header Compression > spec that header names MUST conform to: > > LOWERALPHA = %x61-7A > header-name = "!" / "#" / "$" / "%" / "&" / "'" / > "*" / "+" / "-" / "." / "^" / "_" / > "`" / "|" / "~" / DIGIT / LOWERALPHA > > Which is the all-lower-case equivalent to the header-name definition > currently in httpbis. > >
Received on Wednesday, 14 August 2013 13:24:28 UTC