DNS name binary encoding

[ntninja]

As mentioned in #22, /dns, /dns4, /dns6 & /dnsaddr do not currently have an official binary encoding defined.

In py-multiaddr we currently use Unicode for the text representation and IDNA-2008/Punnycode for the binary representation. Obviously this distinction is only relevant for domains with labels containing non-ASCII characters.

How is this handled in other implementations? Is the current behaviour something that could be standardized or is another behaviour more desirable?

[Stebalien]

I believe the current de-facto standard in go and js is to just encode the unicode to utf-8. IIRC, punnycode is only really needed for compatibility with the DNS protocol.

However, this may have canonization issues.

[ntninja]

However, this may have canonization issues.

Canonicalizing domain name strings is specified by UTS46. In particular that specification requires one to [ship a mapping table] and decode all domain name labels starting with xn--using Punnycode to validate & lower-case their contents. So while could store the end result of this process as UTF-8 rather then IDNA, we'd still have to implement/ship most IDNA stuff anyways to ensure canonicalization of the generated values.

[Stebalien]

(belatedly)

I believe most DNS resolvers will do this internally. I'm fine saying that any domain in a /dns string must have already been converted from punycode to utf-8.

(unless I'm missing something important)

[ntninja]

@Stebalien and me went for standardizing UTF-8 encoding with UTS-46 normalization/canonicalization in #101.