The methods are not expected by CPython and are only explicitly
callable. To get iteration support, PyIterProtocol should be
implemented and to get support for `x in mapping`,
PySequenceProtocol's __contains__ should be implemented.
https://github.com/PyO3/pyo3/issues/611
https://www.python.org/dev/peps/pep-0590/
This was tested on 3.7 using _PyCFunctionFast and 3.8
using PyObject_Vectorcall. Extending pyo3-derive-backend
to generate code using vectorcall is not done here.
This exposes PyObject_Vectorcall with a link_name to the
underscored name on 3.8 because it is expected to be stabilized
on 3.9.
This fixes the "fast" objects being new in 3.7, not 3.6.
This adds a `Python` marker to `GILPool`, to prevent the caller from
misusing it to drain the `ReleasePool` and release Python objects
without the GIL held.
Signed-off-by: Anders Kaseorg <andersk@mit.edu>
Previously, defining `__getattr__` would override all existing fields and methods. This changes it to behave like a `__getattr__` method defined in python, i.e. the custom method is only called if there isn't a field or method of that name
* Add `MODULE` const value to `PyTypeInfo` trait
* Allow setting module name in `pyclass` macro arguments
* Add `$module` argument to `pyobject_native_type` macro in `pyo3::types`
* Set the right module for builtin types in `pyo3::types`
* Add `module` argument to example `word_count.WordCounter` class
* Add `module` argument of `pyclass` macro to guide
* Reformat code using `rustfmt`
* Add tests and entry in CHANGELOG.md for `module` argument in `pyclass`
* Make `$module` parameter in `pyobject_native_type` optional
* Make `pyobject_native_type` declare module as `Some("builtins")` by default
