Re: h2#404 requiring gzip and/or deflate

On Mar 31, 2014, at 22:36, "mcmanus@ducksong.com<mailto:mcmanus@ducksong.com>" <mcmanus@ducksong.com<mailto:mcmanus@ducksong.com>> wrote:

Martin, I don't think you should take that as editorial as it conflicts with some of the implicit-gzip use cases.

CE gzip was made a fixed part of the protocol, in part, because intercepting intermediaries (or antivirus libraries as a subset of that) were explicitly stripping negotiation for it for their own convenience. ...

I don't doubt it. On the other hand, enterprising intermediaries are always going to do something they aren't supposed to do. (e.g. In this case they could decompress and remove the CE gzip if it's more convenient for them.)

I would prefer to see the proposed text plus a requirement that intermediaries MUST NOT change the Accept-Encoding header of a request.

Regardless, if it's decided to stick with forced acceptance of gzip no matter what, the http/2 spec section 9.3 needs to be updated to explicitly disallow "gzip;q=0".

Section 9.3 currently reads "This effectively changes the implicit value of the Accept-Encoding header field ([HTTP-p2]<http://http2.github.io/http2-spec/#HTTP-p2>, Section 5.3.4<https://svn.tools.ietf.org/svn/wg/httpbis/draft-ietf-httpbis/26/p2-semantics.html#header.accept-encoding>) from "identity" to "identity, gzip".

If I read this and the reference to [HTTP-p2]<http://http2.github.io/http2-spec/#HTTP-p2>, Section 5.3.4<https://svn.tools.ietf.org/svn/wg/httpbis/draft-ietf-httpbis/26/p2-semantics.html#header.accept-encoding> then I would assume that you can disable gzip with "gzip;q=0" because that's what you can do with the implicit identity.

BTW, the text in [HTTP-p2]<http://http2.github.io/http2-spec/#HTTP-p2>, Section 5.3.4<https://svn.tools.ietf.org/svn/wg/httpbis/draft-ietf-httpbis/26/p2-semantics.html#header.accept-encoding> strikes me as ambiguous. It doesn't say if you can mix entries with q values with entries without q values. It also doesn't say what the q value is for the implicit identity entry (and by extension the implicit gzip entry in h2) - this would be important if a client specifies multiple entries and wants to give preference over gzip to a new compression algorithm e.g. superawesomezip.



This email message is intended only for the use of the named recipient. Information contained in this email message and its attachments may be privileged, confidential and protected from disclosure. If you are not the intended recipient, please do not read, copy, use or disclose this communication to others. Also please notify the sender by replying to this message and then delete it from your system.

Received on Monday, 31 March 2014 21:46:39 UTC