Pyfhel icon indicating copy to clipboard operation
Pyfhel copied to clipboard

`ImportError` when running `from Pyfhel import Pyfhel`

Open lynkos opened this issue 11 months ago • 3 comments

Description Trying to run code from steps 1-3 in Pyfhel's MultDepth and Relinearization tutorial results in an ImportError when running from Pyfhel import Pyfhel.

Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/Users/kiran/Documents/workspace/Projects/Q-SECURE/project/test/test.py", line 5, in <module>
    from Pyfhel import Pyfhel
  File "/Users/kiran/miniconda3/envs/q-secure_env/lib/python3.11/site-packages/Pyfhel/__init__.py", line 1, in <module>
    from .Pyfhel import Pyfhel
ImportError: dlopen(/Users/kiran/miniconda3/envs/q-secure_env/lib/python3.11/site-packages/Pyfhel/Pyfhel.cpython-311-darwin.so, 0x0002): symbol not found in flat namespace '__ZN4seal13parms_id_zeroE'

Code To Reproduce Error Activate conda virtual environment and install Pyfhel via pip. I also set/added the environment variables from Pyfhel's Installing a C/C++ Compiler instructions to my .bash_profile. Up until now, everything's fine. Running the following command is what produces the aforementioned errors:

python -m project.test.test

Expected behavior Successfully run code from steps 1-3 in Pyfhel's MultDepth and Relinearization tutorial, with the described outputs printed to my terminal.

Setup:

  • OS: Sonoma 14.4
  • Python: 3.11.8
  • C compiler version: [Clang 14.0.6] on darwin
  • Pyfhel Version: 3.4.2

lynkos avatar Mar 15 '24 20:03 lynkos

@lynkos The error indicates that SEAL library was not installed/found alongside Pyfhel.

How did you install Pyfhel? If possible, you could install it directly with pip instead of via conda, I recall some issues with the latter method.

ibarrond avatar Mar 25 '24 12:03 ibarrond

Hi @ibarrond, thanks for reaching out!

How did you install Pyfhel? If possible, you could install it directly with pip instead of via conda, I recall some issues with the latter method.

I installed Pyfhel (in my conda environment) as described in README.md:

pip install Pyfhel

When that didn't work, I uninstalled it and reinstalled it by cloning the repo:

git clone --recursive https://github.com/ibarrond/Pyfhel.git
pip install .

Unfortunately, I still got the same error.

lynkos avatar Mar 30 '24 01:03 lynkos

Hi @lynkos were you able to resolve the issue? I am getting the same error.

tanish-malekar avatar Aug 16 '24 08:08 tanish-malekar