Add a validation constructor for IPv6

[digy]

Implement an ipv6 validator

[baldram]

I will take this one.

[baldram]

Hey, I was working on the IPv6 validation and realised it's non-trivial. It's because of the different notations. There are many nooks and crannies of rules to explore. The implementation delivered by @uurl is the first good step for further extension.

What's missing? Example: ✅ The "compressed notation" works ok: 1762:0:0:0:0:B03:1:AF18 -> 1762::B03:1:AF18. 🟥 The "compressed mixed notation" does not: 1762:0:0:0:0:B03:127.32.67.15 -> 1762::B03:127.32.67.15, as only the first one works and the compressed form is rejected.

I will try to prepare more test cases to find various border cases.

There is also one thing to clarify. A test case in the current implementation expects a valid IPv6 address to be incorrect. I mean this test scenario: 2001:cdba:0000:0000:0000:0000:3257:9652. It seems to be a correct IPv6 address. (Leading zeros are optional and it does not have to be 2001:cdba:0:0:0:0:3257:9652). Would you please clarify @uurl? According to my research, I would fix it and move this scenario to valid ones.

I prepared the ground for extensions and fixes via some other PRs (merged), and I will try to continue here if you don't mind.

[uurl]

Hey, I was working on the IPv6 validation and realised it's non-trivial. It's because of the different notations. There are many nooks and crannies of rules to explore. The implementation delivered by @uurl is the first good step for further extension.

What's missing? Example: white_check_mark The "compressed notation" works ok: 1762:0:0:0:0:B03:1:AF18 -> 1762::B03:1:AF18. red_square The "compressed mixed notation" does not: 1762:0:0:0:0:B03:127.32.67.15 -> 1762::B03:127.32.67.15, as only the first one works and the compressed form is rejected.

I will try to prepare more test cases to find various border cases.

There is also one thing to clarify. A test case in the current implementation expects a valid IPv6 address to be incorrect. I mean this test scenario: 2001:cdba:0000:0000:0000:0000:3257:9652. It seems to be a correct IPv6 address. (Leading zeros are optional and it does not have to be 2001:cdba:0:0:0:0:3257:9652). Would you please clarify @uurl? According to my research, I would fix it and move this scenario to valid ones.

I prepared the ground for extensions and fixes via some other PRs (merged), and I will try to continue here if you don't mind.

Hello @baldram, go ahead, fix it, actually I based this first approach on http://jsfiddle.net/AJEzQ/ and http://jsfiddle.net/DanielD/8S4nq/ But honestly, I can't remember the code I made this morning :laughing: Regards!

[baldram]

A bugfix for leading zeros support is delivered. That was a quick win. However, this is still not everything.

As mentioned, when I started with the task, I discovered many incomplete regexes on the internet. Moreover, a current validator is very basic, and many cases still exist to implement.
It's about mentioned in https://github.com/zio/zio-schema/issues/258#issuecomment-1212374346 "compressed mixed notation" or valid addresses like below (that are not yet supported by zio-schema):

        ::0:0:0:0:0:0:0
        ::0:0:0:0:0:0
        ::0:0:0:0:0
        ::0:0:0:0
        ::0:0:0
        ::0:0
        ::0

I will book some time to prepare good test scenarios and fill gaps in the current implementation.