Fix Git submodule command on Windows

[shivjm]

Projectile produces an error when finding project files in a Git repository with submodules on Windows (#1600). This is because the command to find the submodules is hard-coded, including the quotes. This PR adds a projectile-get-git-sub-projects-command function to generate the command with the correct quoting at runtime. It also adds projectile-git-submodule-command-function, which is set to projectile-get-git-sub-projects-command by default, obsoleting projectile-git-submodule-command.

I’m not sure I can add tests for this since it’s specific to the combination of Git, submodules, and Windows. I did run checkdoc but it showed a whole lot of unrelated errors.

---

Before submitting a PR make sure the following things have been done (and denote this
by checking the relevant checkboxes):

• The commits are consistent with our contribution guidelines
• You've added tests (if possible) to cover your change(s)
• All tests are passing (eldev test)
• The new code is not generating bytecode or M-x checkdoc warnings
• You've updated the changelog (if adding/changing user-visible functionality)
• You've updated the readme (if adding/changing user-visible functionality)

Thanks!

[bbatsov]

I'm not particularly concerned about the tests here, but I'm wondering if there's a simpler way to fix this. Also we'll need to be extra clear in the function docstrings as to why this is not a simple string as most other commands. The public docs have to be updated as well to reflect the change.

[bbatsov]

Probably this should be marked as private and also explain why the command has to be generated at runtime.

[shivjm]

Does this make sense?

(defun projectile--get-git-sub-projects-command ()
  "Get the sub-projects command for Git.
Defined as a function in order to generate the properly-quoted command at
runtime."

[bbatsov]

One thing that's not clear to me is whether this needs to be generated every time or can be generated just the first time (which would mean you can simply move the format in the original defcustom.

[shivjm]

One thing that's not clear to me is whether this needs to be generated every time or can be generated just the first time (which would mean you can simply move the format in the original defcustom.

I also thought of this, and the only concern I had is that (if I’ve understood correctly) it would cause problems if the same configuration were reused across platforms. Otherwise I definitely think it would be better to just generate it once.

[bbatsov]

I also thought of this, and the only concern I had is that (if I’ve understood correctly) it would cause problems if the same configuration were reused across platforms.

I don't quite understand what you mean by this. Can you elaborate a bit?

[shivjm]

My understanding is that the custom variable will be processed once. Say you’re running Emacs on Linux. The custom variable will be set accordingly. If you then reuse that config, including the custom settings, for a Windows machine, the custom variable won’t be updated—it’ll remain in Linux form.

Assuming this is so, that might simply be an edge case we can ignore. I don’t know how important it is.

[github-actions]

This pull request has been automatically marked as stale because it has not had any recent activity. It will be closed soon if no further activity occurs. Thank you for your contribution and understanding!

[github-actions]

This issue been automatically closed due to lack of activity. Feel free to re-open it if you ever come back to it.