Something went wrong. Try again.
An HTTP library for Zig 0.17, server and client, HTTP/1.0 through HTTP/3, with the protocol in a module that owns no socket.
Something went wrong. Try again.
123456789101112131415161718192021222324252627282930313233343536// SPDX-FileCopyrightText: © 2026 Jeffrey C. Ollie <jeff@ocjtech.us>// SPDX-License-Identifier: MIT
//! The parts of a field section that are not one version's.//!//! HPACK (RFC 7541, for HTTP/2) and QPACK (RFC 9204, for HTTP/3) are//! different compressors with different tables and different ways of getting//! a dynamic table from one end to the other. What they are not different//! about is the Huffman code: §5 of RFC 9204 does not define one, it points//! at RFC 7541 Appendix B and stops.//!//! So the table lives here rather than in either driver. `http2` and `http3`//! do not know about each other -- that is the arrangement `root` describes//! and it is worth keeping -- and a table that both of them need is evidence//! that the table belongs to neither. The alternative was to transcribe two//! hundred and fifty-seven codes a second time, which is the kind of//! duplication that stays right until the day somebody fixes one copy.//!//! The same is true of what a field section *means*. RFC 9113 §8.3 and RFC//! 9114 §4.3 put the same rules on pseudo-headers -- which ones there are,//! that they come first, that a field name is lower case, that a//! connection-specific field has no meaning in a protocol with no//! connection-specific anything -- because HTTP/3 inherited the design rather//! than reinventing it. So `request` is here too, with the version it is//! building for as a parameter, and the one thing that differs between them//! is the number written into `Request.version`.
const std = @import("std");
pub const huffman = @import("field/huffman.zig");pub const request = @import("field/request.zig");
test { std.testing.refAllDecls(@This());}