add second fraction normalisation

[samuelcolvin]

fix #702

• [x] Have you added yourself and the change to the changelog? (Don't worry about adding the PR number)
• [x] If this pull request fixes a bug, does it add a test that verifies that we can't reintroduce it?

[samuelcolvin]

By the way, it might be worth allowing CI to run without manual approval, see https://github.com/encode/uvicorn/pull/1437#issuecomment-1086574265.

[samuelcolvin]

Humm, I'm not sure what to do about these tests - it seems like to support leap seconds, you explicitly want the functionality that this PR alters?

[djc]

Ah yeah, I don't have enough context on that -- I only became a maintainer of this crate recently.

[esheppa]

Hi @samuelcolvin - I've made some comments over at https://github.com/chronotope/chrono/issues/702 - please let me know your thoughts

[pitdicker]

@samuelcolvin It has been a year, sorry. Would you still be available to work on this?

• I don't know if it is a good idea to make a subtle change second() and nanosecond(). It would be easy to miss for library users when upgrading.
• Documentation needs to be updated.
• If normalization is added to the Timelike trait it can be removed from the formatting code, right?
• At least the with_second method should be adjusted to accept 60 so it can round-trip with second.

[pitdicker]

Closing due to inactivity and the concerns raised above. We still have https://github.com/chronotope/chrono/issues/702 to track this issue.