Conversation
| # [lock] brings pyodide-lock and uv for jupyter_lite_config.py; pyodide-lock | ||
| # needs 3.12, which every docs build has | ||
| "jupyterlite-pyodide-kernel[lock] >= 0.8; python_version >= '3.12'", | ||
| "jupyterlite-pyodide-kernel >= 0.8[lock]; python_version >= '3.12'", |
There was a problem hiding this comment.
Not sure why this is getting moved, but it's causing CI failures, so will need to look.
There was a problem hiding this comment.
But this is still using the local spec-zero script. I think we just never had a case of a dep with a group having a version specifier
There was a problem hiding this comment.
Ahh okay. Then +1 for fixing it in mne-tools and getting rid of local spec-zero here 😄
| "mne-connectivity", | ||
| "mne-gui-addons", | ||
| "neo", | ||
| "neo", # released 2023-06-30, will become 0.12 on 2028-07-01 |
There was a problem hiding this comment.
Also @thht a bit weird that something we don't pin at all got a comment
There was a problem hiding this comment.
@larsoner I honestly have no idea what my involvement is here...
There was a problem hiding this comment.
Ahh sorry @thht I got my Thomas GitHub names confused because of our recent emails 🤦 😆
There was a problem hiding this comment.
Yes, us Thomasses are everywhere 🤣
|
... so I'm going to close this, we shouldn't merge it. Once a PR switches us to mne-tools and the bug is fixed, we workflow-dispatch it to do a proper update |
|
Sounds good. I'll have a look |
4 participants
Created by the
spec_zerojob of this GitHub Action run.Adjustments may need to be made to shims in
mne/fixes.pyand elsewhere in this or another PR.make -C tools/dev depis a good starting point for finding potential updates.