202605 Binary Installation with Conda Environment

[kcusson83]

More of a question than an issue. The newest binary installation, designed for python 312 includes instructions for installation with a Python virtual environment, but suggests that that is not a prerequisite. Are there instructions for what is needed to install the binary as part of a Conda environment?

[prascle]

Hello,
The latest binary version (May 2026) differs from previous versions (2025 and earlier): the binary package now includes the CloudCompare prerequisites. You no longer need Conda, but you can use it to set up your Python 3.12 environment.

If you want to use a Python 3.12 Conda environment, you will need to install numpy requests psutil scipy quaternion cmake matplotlib:
conda install numpy requests psutil scipy quaternion cmake matplotlib
or
mamba install numpy requests psutil scipy quaternion cmake matplotlib

To create a Conda CloudComPy312 environment from scratch:

conda install -y -n base mamba -c conda-forge
conda update -y -n base -c defaults conda
conda activate 
mamba env create -y -n CloudComPy312 -f CloudComPy312.yml
conda activate CloudComPy312 

with the file CloudComPy312.yml :

name: CloudComPy312
channels:
 - conda-forge
 - nodefaults
dependencies:
 - python=3.12
 - cmake
 - matplotlib
 - numpy
 - psutil
 - quaternion
 - requests
 - scipy

Next, you will need to set the paths for cloudComPy using the envCloudComPy script, as described in the instructions provided with the CloudComPy binary.

I haven't had a chance to test the procedure above yet; please let me know if you run into any problems!

Please note that if you are using Windows, you may encounter issues with a Conda environment that contains certain Qt6 packages (there may be a version conflict between some DLLs included in the cloudComPy package and those provided by your Python environment). If that is the case please don't hesitate to post a detailled issue.

Regards,
Paul

[prascle]

Hello, unfortunately, the procedure described above does not work on Windows. I think matplotlib requires Qt6 via conda, and its DLLs conflict with those of CloudComPy... On Linux or macOS, it should work.
If you don't need matplotlib, remove it from the list above. It works... However, this means you might not be able to create a conda environment with everything you need...

On Windows, you'll run into fewer problems if you use a Python virtual environment (venv) and PyPI packages.

The main goal of this packaging change is to avoid being forced to use a very specific Conda environment that matches the version of CloudComPy. This is quite difficult to manage, as Conda packages evolve rapidly. What works one day may no longer work a few days later.

Please leave this ticket open; I think I'll have to make some improvements to the Windows package one way or another...

Regards,
Paul

[prascle]

Hello,
I've released a new version that includes a fix (compiling cloudComPy with Qt6 from Conda). With this new version, there are fewer DLL conflicts, and it is possible to use a Conda environment with matplotlib, Jupyter, Spyder... The documentation has been updated.

Regards,
Paul

[kcusson83]

Thank you so much for your work on this, I really appreciate it!