[quicwg/base-drafts] HTTP SETTINGS: define setting content encoding (#1556)

[Luca Niccolini <[e-mail]>]

The current draft says:
> The payload of a SETTINGS frame consists of zero or more parameters,
   each consisting of an unsigned 16-bit setting identifier and a
   length-prefixed binary value.

This leaves the encoding of the setting content to the implementor and may be cause of ambiguity.
I believe we should define the encoding more explicitly for each setting, e.g. 
is `SETTINGS_NUM_PLACEHOLDERS` encoded as a fixed-length 4 or 2 (given its maximum value) bytes integer, or perhaps it can also be variable-length, e.g. can the value be encoded in just 1 byte if less than 256 ? 

Also, do we expect arbitrarily large setting contents ?
Implementors would likely enforce some sensible maximum values whenever the length is encoded as a quic-integer, do you think there is value in the RFC suggesting these maximum values ? 
This is probably a larger scope question, but in the settings frame seems more prominent

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/1556