Add TypeID community implementation for Erlang

[eproxus]

Summary

Adds keysmith, a new TypeID Erlang community implementation.

How was it tested?

Markdown has been manually reviewed.

[loreto]

@eproxus Thanks for submitting!

Could you point me to the tests validating conformance with the typeid spec? Basically the tests that test these cases:

• https://github.com/jetify-com/opensource/blob/main/typeid/typeid/spec/invalid.yml
• https://github.com/jetify-com/opensource/blob/main/typeid/typeid/spec/valid.yml

[eproxus]

@loreto 👍 I'm working on adding those. I'll let you know when it is done.

[eproxus]

@loreto Looking into the test data, it seems the test data does not follow the specification. The specification says:

3. A UUID suffix: a 128-bit UUIDv7 encoded as a 26-character string in base32.

But the test data contains (among others) the UUID 00000000-0000-0000-0000-000000000000 which is not a valid UUID version 7 (the valid version with all zeroes would be 00000000-0000-7000-8000-000000000000).

Since my implementation does strict parsing and generation of UUIDs it cannot use this test data for automated tests.

[eproxus]

I realize now that any custom UUID is technically allowed by the spec. I've added validation for the invalid IDs for now, and will work on adding support for validating the valid IDs as well.

[loreto]

I realize now that any custom UUID is technically allowed by the spec. I've added validation for the invalid IDs for now, and will work on adding support for validating the valid IDs as well.

Thanks! Yeah the way I see it is:

• By default a library should always generate new TypeIDs using UUIDv7
• However, if the user of the library so chooses, they can encode any UUID as a TypeID

[eproxus]

I've now added support for a more general UUID format so both the valid and invalid test files now pass. You can find the test code here: https://github.com/eproxus/keysmith/blob/main/test/keysmith_tests.erl#L347-L383

[loreto]

LGTM