-
Notifications
You must be signed in to change notification settings - Fork 6
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 failure on Python 3.9.5 (RHEL 8.8, fresh environment) #11
Comments
Sorry to push but I was wondering if there'd been any progress on this? I know we're waiting for the JOSS review to finish up but if we can resolve this now it'll speed things up once the reviewers are ready to accept. |
Thanks for pushing, I wasn't able to look into it at this time, but I do want to address it promptly. I do expect being able to this Wednesday, but I'll try looking into it before too. I am not familiar with RHEL 8.8, but I'll see if I can reproduce the error. |
I was able to reproduce the error in a RHEL 8.10 docker container. After setting up a minimalist python version installation (no conda, just python 3.9 with a pip installer), I ended up with the same error. After doing a few tests, I saw that the pip I got installed was version 20.2.4, so I tried upgrading it to its latest version (24.2 through pip). After completing the upgrade, I was able to pip install I have yet to test any version of pip in between, but could you let me know what version of pip are you running for now? |
I was using pip version 21.1.1, I think inherited from our OS installation (possibly including that I loaded Python 3.9.5 via Environment Modules). I tried creating a fresh virtual environment (still using the Python 3.9.5 module), upgrading pip to 24.2 and installing
for both. I'm guessing there should be a list of source files in there that for some reason is coming up empty. More error output
|
I'm not finding a way to reproduce that error. I first thought it was due to I guess you installed using the recommended |
I originally install using |
For some reason it tried to build SciPy from source, so once I installed SciPy's system-level dependencies, I progressed to the message I now remember seeing, which is that |
The Python requirement under 3.12 is intentional at the moment, but I hope to revise it soon. To clarify, that last comment was from your installation attempt in your local machine running Fedora 40? |
Sorry, yes, the local install failed purely because Fedora 40 has Python 3.12. I also hope to circle back to debugging the Python 3.9 build on my University cluster. (I'll comment on the JOSS review in that thread.) |
I spent some time dabbling with What I did find is that if I simply commented out the I'll try to return to this tomorrow to see if I can safely comment out |
Thanks for looking into it. I'll try to test that out on my end too. The |
Though I'll leave this open, it's peculiar enough to my particular machines that I'm going to press on with the JOSS post-review process but I'll try to chip away at this and ask some colleagues and collaborators if they've ever run into anything similar. I noticed I had |
To test the installation process, I just created a fresh Python 3.9.5 environment and ran
which eventually errored out with
My guess is that this is coming up because the function
get_versions
inpy-EnBiD-ananke
(though I think the same function is used inpy-ananke
andpy-Galaxia-ananke
) is returning one of the dictionaries withdate = None
.This is RHEL 8.8.
pip freeze
returns nothing, presumably because I'm in a fresh virtual environment.Related to openjournals/joss-reviews#6234.
The text was updated successfully, but these errors were encountered: