On 20.05.2023 19:42, Oleg Kalnichevski wrote:
Arturo et al
HttpClient Cache module of the HttpComponent Client project has never
been truly reviewed for conformance with RFC73xx. There are still many
instances of references to RFC 2616 in comments and test cases and
there are several recommendations that the caching module tries to
fulfill no longer valid or present in the newer revisions of the HTTP
protocol.
However, even though HttpComponents core and client modules are still
at the RFC73xx level, I see no point upgrading the caching module to
conform to RFC 7234 when there is RFC 9111 that supersedes it.
If you have some spare cycles please do feel free to help me with the
review.
Oleg
+1
With my HTTP editor's hat on: when the IETF HTTP Working Group revised
the core specs, this was almost *entirely* about fixing bugs and
refactoring the structure (the latter mainly so that the various wire
formats - 1.1, 2, and 3 - can be defined on top of the core specs).
There are minor additions imported from other specs, such as new
cache-controls, but support for additions like these is truly optional.
When there's a normative change, it's usually because either the
previous revision was itself incorrect, or when research showed that
spec and the reality disagreed.
Each of the base specs (HTTP: 9110, Caching: 9111, plus 9112 and 9113
for HTTP/1.1 and HTTP/2) has a "Changes" section that is *supposed* to
summarize all non-editorial changes.
If you find issues with the specs, by all means report them to the
Working Group, either through the mailing list or through Github issues.
See <https://httpwg.org/> for more information.
Reminder: participation in IETF work is free and mostly happens on
mailing lists (and more recently issue trackers). See
<https://www.ietf.org/about/participate/tao/>.
Best regards, Julian
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@hc.apache.org
For additional commands, e-mail: dev-h...@hc.apache.org
