Should set user agent and language as per ACME spec when talking to LE

[webprofusion-chrisc]

https://tools.ietf.org/html/draft-ietf-acme-acme-07#section-6.1

ACME clients SHOULD send a User-Agent header in accordance with
[RFC7231], including the name and version of the ACME software in
addition to the name and version of the underlying HTTP client
software.

ACME clients SHOULD send an Accept-Language header in accordance with
[RFC7231] to enable localization of error messages.

[Marcus-L]

I think this is new for ACME v2 (draft 07). V1's spec is the draft 01 and does not have that language.

[cpu]

Hi @Marcus-L,

As a small point of clarity the "V1" API isn't strictly ACME draft 01. It's closer to somewhere between draft-03 and draft-07. We document the important divergences here. I believe the User Agent "SHOULD" portion of the spec has been present since ~draft-04.

[Marcus-L]

Ah, ok, thanks for the clarification. I didn't know it was such a moving target. I'd just finished implementing tls-sni-01 (from ACME v1) for Certify, which was in the draft-01. It started being called tls-sni-02 in draft-02, so I assumed that was the cutover.

[cpu]

@Marcus-L I can definitely understand your confusion, its a bit of a mess :-) The individual challenges are "versioned" based on the draft they were introduced in, but Boulder has evolved alongside the drafts and doesn't strictly conform to any one.

[webprofusion-chrisc]

@cpu just as an aside, rather than an rfc it would be cool to see a swagger style API doc or docs by example for each request/responds. I find the rfc a little verbose (and diverged from, as noted) when I'm just really after an API guide.

[cpu]

@webprofusion-chrisc I'm not familiar with Swagger but I will take a look. Thanks for the idea!

[Marcus-L]

I submitted PR ebekker/ACMESharp#300 to open up the UserAgent and Language.

[Marcus-L]

Closed via #180