Closed kohenkatz closed 2 months ago
Sorry for the 13-month delay on this! I think it is ready for review now.
I just realized the coverage isnt running anymore...
Got codecov set back up but it probably wont run on this until we merge. I added @dylan-bourque to the review to see if he has any feedback, but the changes seem correct to me and consistent with the verbiage in the latest RFC. Thank you for taking the time to update this.
I gave this a quick scan and didn't see anything. I'm traveling today so won't have a chance to do a deeper review probably until Monday.
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 100.00%. Comparing base (
22c52c2
) to head (a3cbee1
). Report is 2 commits behind head on master.:exclamation: Current head a3cbee1 differs from pull request most recent head e2280ed. Consider uploading reports for the commit e2280ed to get more accurate results
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
~On 2022-10-17, the IETF uuidrev workgroup published a new draft that replaces peabody draft 04. On 2023-11-06, they released draft 14.~ On 2024-05-07, the IETF published RFC 9562, obsoleting RFC 4122.
This PR updates this project to implement the newest RFC. (fixes #118)
There are a few changes in documentation: mostly pointing to the new RFC URL, and adding comments as appropriate to explain the changes.
A new constant
VariantRFC9562
is added, with the oldVariantRFC4122
kept as an alias to it for backward compatibility.The one major code change is in the implementation of UUID version 6. Since draft-01, this version now
SHOULD
use new pseudo-random data for the final 62 bytes, instead of using aclock_seq
value for 14 of those bytes. (Although implementationsMAY
keep the oldclock_seq
behavior, this is discouraged, and is only allowed alongside keeping the MAC address as thenode
value. Since we are not keeping the MAC address, we also should not keep theclock_seq
.