Indexer: Improve locating stories with specials chars in path

[jankoritak]

Closes #21636

What I did

• Altered StoryIndexGenerator.ts not to treat this.options.workingDir as a glob. This allows the absolute part of the path to contain special glob characters without breaking the match.
• The problem and proposed solution are described inside of the issue.
• Note: As we're altering how we treat the local/absolute portion of the path, I'm unsure whether there is a way to test this.

How to test

1. Insert a directory with a special character (e.g. Special (1)) in your path that contains a repository with a storybook.
2. Run the storybook linked to the next branch.
3. Verify that the stories are not loading (the storybook does not find any)
4. Run the storybook linked to this branch.
5. Verify that the stories are loaded.

Checklist

• Make sure your changes are tested (stories and/or unit, integration, or end-to-end tests)
• Make sure to add/update documentation regarding your changes
• If you are deprecating/removing a feature, make sure to update
  MIGRATION.MD

Maintainers

• If this PR should be tested against many or all sandboxes,
  make sure to add the ci:merged or ci:daily GH label to it.
• Make sure this PR contains one of the labels below.

["cleanup", "BREAKING CHANGE", "feature request", "bug", "documentation", "maintenance", "dependencies", "other"]

[ndelangen]

What does this mean?

[jankoritak]

It means that with cwd, we're now treating only path.join(specifier.directory, specifier.files) as glob. In case the comment brings more confusion than value, I'm happy to remove it.

[ndelangen]

OK, I see now.

Would I need to do the same here:
#22186

https://github.com/storybookjs/storybook/pull/22186/files

[jankoritak]

I'd say yes, in case there are special chars in the path, they won't match without it.

[jankoritak]

Hi @shilman, anything I can do to help push this PR toward a merge?

[shilman]

@jankoritak Sorry for the delay. I'll make some time to review with @ndelangen this coming week.

[supryan]

Thank you all for your hard work on this issue, this is currently impacting me. Hopefully a merge is around the corner!

[ndelangen]

@jankoritak this PR I referenced got merged: #22186

Would you be able to apply the fixes from this PR on that part of the codebase as well?
I'll update the branch for you.

[jankoritak]

@jankoritak this PR I referenced got merged: #22186

Would you be able to apply the fixes from this PR on that part of the codebase as well? I'll update the branch for you.

@ndelangen Sure, I'll look into it. Thanks for the update.

[ndelangen]

@JReinhold or @valentinpalkovic Is either of you able to determine if this is still something we need going forwards?

[valentinpalkovic]

This Pr's fix seems reasonable.

[ndelangen]

@jankoritak could you resolve the merge conflicts?

[jankoritak]

@jankoritak could you resolve the merge conflicts?

Hey @ndelangen, will do. I forgot about this PR, my bad!

[tmeasday]

Do we end up calling glob() twice?

[shilman]

@ndelangen please take another look!

[ndelangen]

@jankoritak I updated the branch, looks like CI found some problems, are you interested in fixing up this PR?

[jankoritak]

@jankoritak I updated the branch, looks like CI found some problems, are you interested in fixing up this PR?

@ndelangen I am. However, I'm swamped at the moment. Would you guys be open to waiting for a couple more weeks till I find time to fine-tune the PR? In case this is a no-go for you, lemme know.

[ndelangen]

That's not a problem, as a heads-up to you:
The 7.6 release (should go out next week) will likely be the last 7.x release, and we'll be focussing on the 8.0 release soon.
So there might be more merge conflict incoming.
I'll help you when it comes to that though, just ping me for assistance.

[JReinhold]

I think some merging has gone bad, as this now does const files = glob(...) twice right after each other.