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

try mamba, remove docs job in azure #312

Merged
merged 5 commits into from
Aug 11, 2020

Conversation

bollwyvl
Copy link
Collaborator

References

Code changes

  • updates and uses mamba instead of conda
  • removes docs jobs (because RTD)

User-facing changes

  • n/a

Backwards-incompatible changes

  • n/a

Chores

  • linted
  • tested
  • documented
    • not sure if we'll want to also recommend this, since we use it?
  • changelog entry

@bollwyvl
Copy link
Collaborator Author

before after
Screenshot_2020-08-10 Pipelines - Run 20200811 2 logs Screenshot_2020-08-10 Pipelines - Run 20200811 3 logs

@bollwyvl
Copy link
Collaborator Author

Hah, windows still failing, will try pinning conda (looks line conda-forge maybe jumped the gun with 4.8.4)

@bollwyvl
Copy link
Collaborator Author

before after
Screenshot_2020-08-10 Pipelines - Run 20200811 2 logs(2) Screenshot_2020-08-10 Pipelines - Run 20200811 3 logs(1)

yeah, so looks like a minute across the board (still-running job flaked). Not bad! I've tried pinning conda, and telling it not to update incidental packages, will see if that offsets the additional expense of mamba and friends (or makes it worse).

@bollwyvl
Copy link
Collaborator Author

Still fail on windows, but got further. interesting that our huge solve actually went slower, but there's so many factors...

before after
Screenshot_2020-08-11 Pipelines - Run 20200811 2 logs Screenshot_2020-08-11 Pipelines - Run 20200811 4 logs

@bollwyvl
Copy link
Collaborator Author

The tea laves (win/36) suggest we're saving a couple minutes on windows, which makes sense. So we can either:

  • merge this now, just rely on those gains for future stuff
  • see how some other approaches (like conda-lock) work, in isolation (as it would become less reliant on runtime solves)
before after
Screenshot_2020-08-11 Pipelines - Run 20200811 2 logs(1) Screenshot_2020-08-11 Pipelines - Run 20200811 5 logs

@krassowski
Copy link
Member

Yes, this seems like a step in the right direction. Feel free to merge or explore other options! I do have high hopes for GitHub Actions.

@bollwyvl bollwyvl merged commit d7ac678 into jupyter-lsp:master Aug 11, 2020
@bollwyvl
Copy link
Collaborator Author

Welp, it's easy to back out, and making subsequent experimentation faster is just, well, faster!

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

Successfully merging this pull request may close these issues.

2 participants