Thanks a lot ! Lucky for us IPv6 is not that common yet.

I don't think this change was correct based on what I'm seeing in the specification. If you take a look at the host part you'll notice the mention of IP Literals you'll notice the following text:

A host identified by an Internet Protocol literal address, version 6
[RFC3513] or later, is distinguished by enclosing the IP literal
within square brackets ("[" and "]").

I see no mention around there of the brackets being optional. Could you point me to some documentation on this?

Let me clarify, I think this change was correct for the IPv6 validation, but it in turn messed up validation for URIs. So now URIs with IP literals that don't have the brackets would also be valid. I think some refactoring of this change just needs to occur so we have the correct setup for IP literals within a URI and also a single IP address.

Okay. Did some more looking at this, I'm being an idiot and misunderstood the change that was made here. At the surface I thought this was removing the brackets around the IP Literals, but I was entirely incorrect.

This is merely dealing with correcting the translation of the ABNF format for IPv6 here:

IPv6address = 6( h16 ":" ) ls32
/ "::" 5( h16 ":" ) ls32
/ [ h16 ] "::" 4( h16 ":" ) ls32
/[ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32
/ [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32
/ [ *3( h16 ":" ) h16 ] "::" h16 ":" ls32
/ [ *4( h16 ":" ) h16 ] "::" ls32
/ [ *5( h16 ":" ) h16 ] "::" h16
/ [ *6( h16 ":" ) h16 ] "::"

Specifically the lines 3-9 which have [ h16 ] etc. meaning it is optional. Please disregard.

Just to clarify for anyone who wonders: [ , ] is for options while literal square brackets are written "[" , "]" in ABNF.

This thread has been automatically locked due to inactivity. Please open a new issue for related bugs or questions following the new issue template instructions.