BUG Installation dependencies with new version of numpy?

[azshe]

Summary of issue

After installing, import serpentTools throws up some dependency errors that I think arise from numpy2.0.0 removing longfloat in favor of longdouble. If the workaround is as simple as just using an older numpy, that's fine, I just don't know if there would be other dependency things cropping up.

Code for reproducing the issue

Installing in a venv with python 3.9 and attempting to import serpentTools. The package list in the venv is here:

`pip list --local
Package Version

---

contourpy 1.3.0
cycler 0.12.1
fonttools 4.56.0
importlib_resources 6.5.2
kiwisolver 1.4.7
matplotlib 3.9.4
numpy 2.0.2
packaging 24.2
pillow 11.1.0
pip 23.0.1
pyparsing 3.2.1
python-dateutil 2.9.0.post0
PyYAML 6.0.2
serpentTools 0.11.0
setuptools 58.1.0
six 1.17.0
zipp 3.21.0`

side note: I'm trying to use python 3.9 to avoid the distutils error from #527 but alas it wasn't going to be this simple 😩

Actual outcome including console output and error traceback if applicable

The full error is here:
python -c "import serpentTools" Traceback (most recent call last): File "<string>", line 1, in <module> File "/home/ashe/Serpent/serpent-tools/serpentTools/__init__.py", line 3, in <module> from serpentTools.parsers import * File "/home/ashe/Serpent/serpent-tools/serpentTools/parsers/__init__.py", line 17, in <module> from serpentTools.parsers.depmatrix import DepmtxReader, readDepmtx File "/home/ashe/Serpent/serpent-tools/serpentTools/parsers/depmatrix.py", line 6, in <module> from numpy import empty, empty_like, longfloat, zeros ImportError: cannot import name 'longfloat' from 'numpy' (/home/ashe/.venvs/my_serpent3.9/lib/python3.9/site-packages/numpy/__init__.py)

Expected outcome

It just works 🍎

Versions

Please provide the following:

• Version from serpentTools.__version__ 0.9.4
• Python version - python --version 3.9.21
• IPython or Jupyter version if applicable
• If using the development version, the output from git describe

[drewejohnson]

Hmmm v 0.11 should have fixed this, but the output of serpentTools.__version__ says 0.9.4

Definitely numpy 2.0 problems though

(sorry for the horrendous delay)

[drewejohnson]

Some quick checks in python 3.9 conda environment, installing with pip and with conda, and creating a py39 venv inside my conda env and installing w/ pip and I can't reproduce this problem

Can you try with a fresh venv and/or with conda?

[azshe]

From an almost fresh WSL2 install of Ubuntu-24.04, here's a copy-pasta straight outta .bash_history:

sudo apt update
sudo apt install -y build-essential libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev wget curl llvm libncursesw5-dev xz-utils tk-dev libxml2-dev libxmlsec1-dev libffi-dev liblzma-dev
cd /tmp
wget https://www.python.org/ftp/python/3.9.18/Python-3.9.18.tgz
tar -xf Python-3.9.18.tgz
cd Python-3.9.18
./configure --enable-optimizations
make -j$(nproc)
sudo make altinstall
python3.9 --version
git clone https://github.com/CORE-GATECH-GROUP/serpent-tools.git
cd ~/serpent-tools/
python3.9 -m venv .venv-py39
source .venv-py39/bin/activate
python --version
python -c "import numpy"
python -m pip install --user --upgrade pip serpentTools
python setup.py install
nano setup.py
pip install serpentTools

From here, importing serpent tools didn't complain, but trying to use it throws a numpy version error. I have not tried this with conda.

[drewejohnson]

Hmmm okay. I'll see what I can cook up.

what is the nano setup.py edit? Everything else is pretty easy to reproduce but if you muck w/ the setup.py file that might be something

[azshe]

Caveat: this was from another user I was working with to try to get serpent tools set up, but I'm fairly certain we just peeked into setup.py and didn't edit anything.