CI: limit circleci redirector check

[sappelhoff]

following:

• MAINT: Limit redirector check mne-tools/mne-python#11156

@Remi-Gau I think you recently noted that there are so many of the checks for this CI step. Perhaps this is a solution.

[codecov]

Codecov Report

Base: 88.33% // Head: 88.33% // No change to project coverage 👍

Coverage data is based on head (4fb182b) compared to base (296ec42).
Patch has no changes to coverable lines.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1327   +/-   ##
=======================================
  Coverage   88.33%   88.33%           
=======================================
  Files          11       11           
  Lines        1080     1080           
=======================================
  Hits          954      954           
  Misses        126      126           

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[effigies]

Worth trying...

[effigies]

LOL no. Bad syntax. But it worked before...

[sappelhoff]

so weird. it works nicely on https://github.com/mne-tools/mne-python/actions

[effigies]

mne-tools/mne-python@b5abb5a

[effigies]

And I'm not sure that it's working nicely. It's generating a ton of cancellation messages, rather than a bunch of green checks.

[sappelhoff]

@larsoner do you have advice for the redirect action? why is it performing so many checks?

[sappelhoff]

okay I guess the problem is that on: [status] is just very unspecific and gets triggered on every tiny event there is. A proper solution would probably tackle the on: setting.

[larsoner]

scientific-python/circleci-artifacts-redirector-action#27
scipy/scipy#17002

Haven't found a good solution yet but there is a lot of related discussion above

[larsoner]

@sappelhoff in particular can you see if you agree with my summary about why repository_dispatch is intractable here:

scientific-python/circleci-artifacts-redirector-action#27 (comment)

I am not as experienced with repository dispatch, but IIRC you've used it for some doc building gymnastics before (apologies if I'm mis-remembering)

[sappelhoff]

Thanks for the links! Seems like it's as good as it gets for now.

I don't think I have used repository_dispatch in the past (or at least not while knowing what it's called), but your conclusions sound sensible in that post.