WIP: CI: Use iterative/setup-dvc to install dvc

[seisman]

Description of proposed changes

dvc has a complicated dependency and sometimes dvc fails to work due to incompatible changes of other packges (e.g., #1693, #1543, #2338, and recent #2464 (comment)).

In this PR, I tried to use the iterative/setup-dvc action to install dvc instead of manually installing dvc using mamba. It turns out the iterative/setup-dvc action provides a more stable dvc installation. However, iteractive/setup-dvc is extremely slow on Windows (takes about 2.5 minutes).

Reminders

• Run make format and make check to make sure the code follows the style guide.
• Add tests for new features or tests that would have caught the bug that you're fixing.
• Add new public functions/methods/classes to doc/api/index.rst.
• Write detailed docstrings for all functions/methods.
• If wrapping a new module, open a 'Wrap new GMT module' issue and submit reasonably-sized PRs.
• If adding new functionality, add an example to docstrings or tutorials.
• Use underscores (not hyphens) in names of Python files and directories.

Slash Commands

You can write slash commands (/command) in the first line of a comment to perform
specific operations. Supported slash commands are:

• /format: automatically format and lint the code
• /test-gmt-dev: run full tests on the latest GMT development version

[maxrjones]

Regarding the title, are more changes expected on this PR?

[weiji14]

Regarding the title, are more changes expected on this PR?

However, iteractive/setup-dvc is extremely slow on Windows (takes about 2.5 minutes).

Probably need to consider if we want to accept a slower Windows build.

[maxrjones]

Regarding the title, are more changes expected on this PR?

However, iteractive/setup-dvc is extremely slow on Windows (takes about 2.5 minutes).

Probably need to consider if we want to accept a slower Windows build.

My preference is to prioritize functionality (in this case executing the tests completely) over speed. If pinning funcy provides a faster solution that seems like a good alternative.

[weiji14]

Regarding the title, are more changes expected on this PR?

However, iteractive/setup-dvc is extremely slow on Windows (takes about 2.5 minutes).

Probably need to consider if we want to accept a slower Windows build.

My preference is to prioritize functionality (in this case executing the tests completely) over speed. If pinning funcy provides a faster solution that seems like a good alternative.

The funcy error was actually fixed in dvc already at iterative/dvc#9279, but unsure when they'll cut a new release 🙂

[maxrjones]

Regarding the title, are more changes expected on this PR?

However, iteractive/setup-dvc is extremely slow on Windows (takes about 2.5 minutes).

Probably need to consider if we want to accept a slower Windows build.

My preference is to prioritize functionality (in this case executing the tests completely) over speed. If pinning funcy provides a faster solution that seems like a good alternative.

The funcy error was actually fixed in dvc already at iterative/dvc#9279, but unsure when they'll cut a new release 🙂

Nice! Seems like their release frequency is pretty high, so we may be able to wait. Though it would be best to have the tests passing before our release.

[seisman]

I think we should pin funcy temporarily to make the tests pass So we can have the0.9.0 release done.

[weiji14]

Did you want to keep this iterative/setup-dvc workflow or close it?

[seisman]

Closing this PR.