Conversation
|
Also see #927 - though I think you go further here! |
|
Sorry, totally missed #927. And |
|
@konstin
Agreed. |
|
@konstin |
|
Since this PR is almost ready, I think it's convenient to use this branch for migrating to stable. Here's a TODO list:
|
|
Is |
Though I like slice_pattern and it simplifies our codes, I think we can avoid it.
I'm not sure it's the only blocker, but happy to try to down the version 🙂 |
|
I think we can support 1.39.0, but cannot support 1.38 because of const_vec_new. |
|
Now the only TODO is documenting |
|
I suggest adding to |
|
@davidhewitt |
|
👍 I'll do my best to find some time to read this all tonight |
davidhewitt
left a comment
There was a problem hiding this comment.
Looks really brilliant! I've made some small suggestions to wording of documentation - pick the suggestions you like 😄
Cargo.toml
Outdated
| # For CI | ||
| all-stable = ["default", "num-bigint", "num-complex"] |
There was a problem hiding this comment.
If this feature is just for convenience in CI scripts, maybe it's better to code it into the CI scripts instead of making this user-facing feature?
| [](https://github.com/PyO3/pyo3/actions) | ||
| [](https://codecov.io/gh/PyO3/pyo3) | ||
| [](https://crates.io/crates/pyo3) | ||
| [](https://rust-lang.github.io/rfcs/2495-min-rust-version.html) |
|
Thank you for the reviews! |
|
Congrats, this is a huge step! |
Fixes #5 🚀
Fixes #210
Fixed #697
Based on #961, I wanted to find out what's still missing and it turns out it's not much!
I've moved the performance improvements through specialization behind a
unstablefeature gate. There still two tests failing, which afaik test the behaviour of the specialized functions I deactivated.Once #961 is ready, you should be able to cherry pick 97f6aa6.