I don't understand why some checks fail.
How do I know, if they are relevant?

Do you think, this is sufficient as a solution?

How do I know, if they are relevant?

In this case those two builds can be ignored, the failure is unrelated and @drammock plans to look into it soon.

Do you think, this is sufficient as a solution?

Maybe we should wait a couple of days to hear from the upstream issue fieldtrip/fieldtrip#1676 , and then merge this if we can't get an updated file, WDYT?

Yes sure, that would interest me too (maybe there are several versions of Neuromag122?). Thank you for looking into it

... as mentioned in the issue, let's get rid of the .mat file and the associated lines that are currently commented out. That plus a doc/changes/latest.inc and we should be good to go here!

Ok:)

@marsipu I realized that we are missing some sort of test. This is a little bit tricky because we don't have a Neuromag 122 test file. Do you have or could you create some tiny raw (you could crop it to 1 sec and save it in MNE) file that we could add to the testing dataset? I would probably put it in misc. Then we could for example add a tiny test that connectivity can be computed and the result is a sparse array of shape (122, 122).

Ok sure, I'll do it as soon as I got time

Oh, I rebased the branch but I didn't think about the commits then being part of the pull request. Is this a problem, should I revert the new commits from main?

Oh, I rebased the branch but I didn't think about the commits then being part of the pull request. Is this a problem, should I revert the new commits from main?

Hmmm, if you rebase properly then the commits from main should not even show up in the PR. Moreover, they should come before your commits instead of after them. For example one commit that shows up in this PR now:

cdd332b

Is actually a different commit actually in main:

0de802a

How did you do the rebase?

One option is that, if your latest commit is actually meant to be b2ef4a0, you can do git reset --hard b2ef4a0. Then something like this should work:

git fetch upstream
git rebase upstream/main
git push origin -f neuromag122_adjacency  # WARNING: Will force push

Thank you for your help, I started the rebase in a terminal with

git fetch upstream
git rebase upstream/main

but then pushed the changes with PyCharm (because I am still a bit confused about how the GitHub authentication works and in PyCharm my credentials are already stored). Maybe that combination was the problem.

I added the test-file to mne-testing and updated the version in mne, just the hash-update is missing.

Then we could for example add a tiny test that connectivity can be computed and the result is a sparse array of shape (122, 122).

I added a first attempt of a test, not sure if it is at the right spot (I haven't got a lot of experience with testing yet)

Ok, I don't know why it was not changed in the conversation, but I already changed the testing-version in 2e2f76c before. so I marked the conversation as resolved, right?

I rebased the branch onto upstream/main to resolve the conflict in latest.inc.
Is there something else I can do?

Thanks @marsipu !