Package can create invalid Ulids

[Flohack74]

Hi,

as the Ulid specification excludes letters I L O U (Crockford's base32) I made a small test to see how its implemented here. For example a TryParse of invalid 01ARZ3NDEITSV4RRFFQ69G5FAV (note the I) leads to an invalid Ulid value of 01ARZ3NDFZTSV4RRFFQ69G5FAV, without error.

Is there a reason for skipping validity checks, mainly for speed reason?

Thanks!

[Flohack74]

There is even a PR for fixing that: #67 - Can you guys consider this to be merged?

[Vectorial1024]

If possible, perhaps we may consider using UUIDv7 instead of ULID. (RFC 9562, May 2024)

[github-actions]

This issue is stale because it has been open 180 days with no activity. Remove stale label or comment or this will be closed in 30 days.

[siberhecy]

https://github.com/ByteAether/Ulid I think it's safe to move to this package.