1

I understand how UUIDs are pretty much unique if they are created using enough entropy. I read the documentation and trying to understand how they are generated (RFC 4122).

  • How the generation algorithm works?
  • What the different 5 groups (time_low, time_mid, time_hi_and_version, clock_seq_hi_and_res clock_seq_low, node) stand for? (answered here)

Edit: I don't think this is a duplicate of this because I am not asking only the format but also how it is implemented and how the UUID is calculated in Java.

Andrea Bergonzo
  • 3,983
  • 4
  • 19
  • 31
  • I don't think it is a duplicate of `UUID format: 8-4-4-4-12 - Why?` because I am not asking only what the format stands for but also how the Java implementation works. – Andrea Bergonzo Aug 18 '17 at 18:02
  • The Java implementation is open source. You can look it up yourself. Those "5 groups" are documented in the RFC and are not specific the Java. – Sotirios Delimanolis Aug 18 '17 at 18:05
  • With the "go read the implementation by yourself" attitude half of the stack overflow questions should be closed. Even the question this was marked duplicate of should be closed because "It is in the RFC, go to read that". – Andrea Bergonzo Aug 18 '17 at 18:08
  • 5
    Yes, I agree. They should. SO is not a replacement for documentation. If you have a specific question about the details of the implementation, that may be appropriate. – Sotirios Delimanolis Aug 18 '17 at 18:10
  • You seem to be unaware that UUIDs come in different versions. Only one version is based on random number requiring sufficient entropy. The other versions are not. Please read basic documentation such as [the Wikipedia page](https://en.m.wikipedia.org/wiki/UUID) before posting. – Basil Bourque Aug 18 '17 at 18:48
  • If this question was not closed as a duplicate, it would be closed as "Too Broad", since you are asking someone to explain how Java implements an entire algorithm. – Heretic Monkey Aug 18 '17 at 19:23

0 Answers0