Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Installation not working with python 3.11 #68

Closed
ortega2247 opened this issue May 14, 2023 · 5 comments
Closed

Installation not working with python 3.11 #68

ortega2247 opened this issue May 14, 2023 · 5 comments

Comments

@ortega2247
Copy link

I tried installing netgraph using conda and python 3.11 in a windows machine and got the following error:

ImportError: DLL load failed while importing _imaging: The specified module could not be found.

@paulbrodersen
Copy link
Owner

paulbrodersen commented May 16, 2023

Thanks for raising the issue. Unfortunately, I can't be of help in this case. Netgraph does not contain an _imaging module.

A quick google search suggests that PIL / pillow may be the culprit; to be sure, however, you would have inspect the complete error trace. If that is indeed the case, I would suggest installing pillow separately, or upgrading pillow if you have an existing installation. pillow is a dependency of matplotlib (IIRC) but netgraph does not use any of the related features so any pillow version should work (even none at all) as far as netgraph is concerned.

As I am not author/maintainer of the _imaging module, and can't provide further assistance, I will close the issue for now.

@rkurchin
Copy link

Linking to JOSS review and popping in to note: I think if there are installation hiccups on the latest version of Python, we should make sure any gotchas are documented – @ortega2247, can you confirm whether the stacktrace does implicate pillow and whether installing NetGraph in a clean conda environment (i.e. without preexisting pillow and/or matplotlib installations) still causes this issue?

@paulbrodersen
Copy link
Owner

paulbrodersen commented May 17, 2023

I think if there are installation hiccups on the latest version of Python, we should make sure any gotchas are documented

I am sorry but I can't monitor all upstream packages for installation issues on all platforms. I have neither the time nor resources to do so. The pillow team is very active. If indeed there is an issue with pillow on some windows platforms, they will likely resolve it very soon, but I have no way of knowing if and when that happens. On my windows installation, pillow installs without errors when using conda.

@rkurchin
Copy link

Understood. Another thing you could consider would be setting up CI – there are standard Windows runners (and other OSes as well obviously), which would at least let you catch if the "standard" environment/setup trips up, either due to changes in commits you introduce or to upstream packages.

@paulbrodersen
Copy link
Owner

@rkurchin Absolutely. Setting up CI is already on my TODO list but I haven't got around to it yet. As there are two major tickets that I would like to address first (major release with support for multigraphs, improved integration tests), it won't happen in time for this review.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants