Feature/numpy2

[Mateasek]

Description of work

Adds compatibility with numpy>2 while keeping compatibility with numpy<2

Fixes

Implements #87 #86: It seems it was enough to replace np.fromstring with np.frombuffer, tested for numpy < 2 and numpy > 2.

Pytests were also updated (np.bool -> np.bool_) to be compatible with numpy < 2 and numpy > 2

To test

I could not pass server tests, but I don't think that is relevant to this PR, dataclass tests pass. I also tested by remotely getting data from JET ppfs. It worked and the data made sense.

Reviewer checklist

• The author suggested tests are successful.
• Relevant documentation (user and developer) has been added, and is clear and complete.
• Suitable unit tests have been added. If unit tests are not included in this MR, please comment as to why and create an issue to address this.

[BradleyMak]

Thanks for working on this @Mateasek! The changes you've made look good, but the issue described actually upgrading the version of numpy used to >=2.0.0, so the poetry configuration (in pyproject.toml and poetry.lock) should also be updated in this MR to reflect that, ensuring that any related dependencies are also upgraded if required. Once this has been added, I'll do some more testing.

[Mateasek]

Hi @BradleyMak ,

The changes I applied should make it compatible with numpy < 3, should I use it in the pyproject.toml? I'm happy to set it to whatever value you think is suitable. Packaging is not my strong suite.

[BradleyMak]

Hi @Mateasek,

Running poetry add numpy@^2.0 (and resolving any dependency errors that poetry raises) followed by poetry lock should be sufficient. If you encounter any issues with this feel free to let me know.

[Mateasek]

Thanks for the guidance. As requested, I set the numpy=''^2.0' which also required python="^3.9"