2975: RFC: Add GILProtected synchronization primitive and use it for LazyTypeObjectInner. r=davidhewitt a=adamreichold
I would also like to use that type in rust-numpy and it seems we can avoid ~~both a manual unsafe impl and~~ a full blown mutex if we apply it to `LazyTypeObjectInner`.
One downside might be that it ties us closer to the GIL when we want to enable nogil experimentation, but on the other hand, it may also help by reifying the GIL usage. (This is currently limited to comments in unsafe code in rust-numpy for example.)
3022: Fix function name shadowing r=davidhewitt a=mejrs
Fixes https://github.com/PyO3/pyo3/issues/3017
3023: Emit a better error for bad argument names r=davidhewitt a=mejrs
This will emit a better error for code like
```rust
#[pyfunction]
fn output([a,b,c]: [u8;3]) {}
```
Co-authored-by: Adam Reichold <adam.reichold@t-online.de>
Co-authored-by: mejrs <59372212+mejrs@users.noreply.github.com>
3055: Fix compile error when trying to use static slot methods, fixes: #3039 r=davidhewitt a=willstott101
Whilst having static slot methods is of dubious value in the first place, there's little reason not to support it by bringing the different function call macros more in-line with each other.
I had no idea which test file to use so I just made a new one... `test_methods.rs` has no slots in and the test files that do use slots seem to be specific to their protocol. Let me know if you'd like the test somewhere else.
Thanks!
Co-authored-by: Will Stott <willstott101@gmail.com>
3044: Add `PyTuple::to_list` r=davidhewitt a=davidhewitt
Companion to #3043. I've included benchmarks which suggests that this method is indeed about 25% faster on my machine.
Co-authored-by: David Hewitt <1939362+davidhewitt@users.noreply.github.com>
3056: ci: pin syn to 1.0 for MSRV r=davidhewitt a=davidhewitt
Looks like syn 2.0 requires Rust 1.56 - it'll be interesting for us to update to this one day, but not quite yet!
Co-authored-by: David Hewitt <1939362+davidhewitt@users.noreply.github.com>
3049: Add Blog post in Articles and other media section r=messense a=AdilZouitine
Hi there 👋,
This pull request adds a link to a blog post in the PyO3 README under the "Articles and other media" section. The blog post, titled "How we extended the River stats module with Rust using PyO3," details our experience in utilizing PyO3 to improve the performance of the River library's stats module.
Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com>
3043: `PyList::as_tuple()` -> `PyList::to_tuple()` r=mejrs a=davidhewitt
As discussed as a follow-up to #3042
Co-authored-by: David Hewitt <1939362+davidhewitt@users.noreply.github.com>
3042: add `as_tuple()` method to `PyList` r=davidhewitt a=samuelcolvin
From [benchmarks](25fcf98e12/benches/main.rs (L165-L192)), this is significantly faster than `PyTuple::new(py, the_list)`:
```
test list_as_tuple_direct ... bench: 299 ns/iter (+/- 51)
test list_as_tuple_iterate ... bench: 521 ns/iter (+/- 41)
```
Co-authored-by: Samuel Colvin <s@muelcolvin.com>
3036: fix const-ness of some FFI name & doc members r=adamreichold a=davidhewitt
I just noticed that these are [`*const` since Python 3.7](007d7ff73f (diff-830b405713d1c40982ffa918864e39c40c1b318d79b8dd2b523646dc3bd18b52)).
This might break existing compiles, so will not include this in 0.18.2.
Co-authored-by: David Hewitt <1939362+davidhewitt@users.noreply.github.com>
3031: mangle exported functions for PyPy r=davidhewitt a=mattip
I run HEAD of PyO3 with HEAD of some PyPy branches, and noticed that the py3.10 branch (which implements python3.10) [fails to properly build](https://github.com/pypy/binary-testing/actions/runs/4334873617/jobs/7569059852#step:6:179). The failure was a missing export:
```
/home/runner/work/binary-testing/binary-testing/pyo3/src/exceptions.rs:715: \
undefined reference to `PyExc_EncodingWarning'
collect2: error: ld returned 1 exit status
```
So I grepped around in the code for `Py_3_10` to see what new functions were added and found a few that needed PyPy-specific exports.
Co-authored-by: Matti Picus <matti.picus@gmail.com>
3014: feat: add #[pyo3(get, set)] for Cell r=davidhewitt a=AntoineRR
This fixes#2659.
The types for which `#[pyo3(get, set)]` should now work are `Cell`, `Arc` and `Box`.
There is one issue regarding `Box`, the implementation of `FromPyObject` conflicts with another one. I could not find what the issue was, especially since the other implementations for `Arc` and `Cell` work as expected. The related code and test has been commented out for now. Maybe someone could help me fix this issue if I don't figure it out myself? There is also the possibility to remove the implementation for `Box` of course.
Co-authored-by: Antoine Romero-Romero <ant.romero2@orange.fr>
3032: fix clippy and ui tests for Rust 1.68 r=adamreichold a=davidhewitt
The usual 😄
Co-authored-by: David Hewitt <1939362+davidhewitt@users.noreply.github.com>
3019: Use stable toolchain for Valgrind job as version 3.20 should handle the DWARF version. r=adamreichold a=adamreichold
Co-authored-by: Adam Reichold <adam.reichold@t-online.de>
3009: Remove stale references to tox.ini from template substitution scripts. r=messense a=adamreichold
Closes#3008
Co-authored-by: Adam Reichold <adam.reichold@t-online.de>
2993: fix `non_snake_case` lint for `#[pyfunction]` generated code r=davidhewitt a=davidhewitt
As promised in #2990
Co-authored-by: David Hewitt <1939362+davidhewitt@users.noreply.github.com>
2990: Fix `clippy::redundant_closure` lint firing for pyfunction defaults r=adamreichold a=davidhewitt
Fixes#2988
I'll push a follow-up for the FIXME as a separate PR.
2991: unpin 3.11 ci jobs r=adamreichold a=davidhewitt
With Python 3.11.2 released for a couple weeks now, I think #2817 is probably not an issue in CI any more.
Co-authored-by: David Hewitt <1939362+davidhewitt@users.noreply.github.com>
