fix #21409 - Closure not created for default arguments of a static function if called from a nested one

[ghost]

Default args sema appears to be already run but sometimes it needs to be recontextualized

[dlang-bot]

Thanks for your pull request and interest in making D better, @baz-bg! We are looking forward to reviewing it, and you should be hearing from a maintainer soon.
Please verify that your PR follows this checklist:

• My PR is fully covered with tests (you can see the coverage diff by visiting the details link of the codecov check)
• My PR is as minimal as possible (smaller, focused PRs are easier to review than big ones)
• I have provided a detailed rationale explaining my changes
• New or modified functions have Ddoc comments (with Params: and Returns:)

Please see CONTRIBUTING.md for more information.

---

If you have addressed all reviews or aren't sure how to proceed, don't hesitate to ping us with a simple comment.

Bugzilla references

Your PR doesn't reference any Bugzilla issue.

If your PR contains non-trivial changes, please reference a Bugzilla issue or create a manual changelog.

⚠️⚠️⚠️ Warnings ⚠️⚠️⚠️

• In preparation for migrating from Bugzilla to GitHub Issues, the issue reference syntax has changed. Please add the word "Bugzilla" to issue references. For example, Fix Bugzilla Issue 12345 or Fix Bugzilla 12345.(Reminder: the edit needs to be done in the Git commit message, not the GitHub pull request.)

Testing this PR locally

If you don't have a local development environment setup, you can use Digger to test this PR:

dub run digger -- build "stable + dmd#21453"

[ibuclaw]

See review.

1. The current proposed fix will run semantic on the variable declaration from a scope different from where it is declared, which isn't sound from a semantic POV.
2. There already exists a function for checking each variable reference in arg, that is lambdaCheckForNestedRef(arg, sc). I suggest to call this function after inlineCopy and resolveLoc. Propagate the return value if an error occurs (return true).
3. Split up tests into one per function, add a couple more different variations of the expression, such as uint c0 = p0 + 1 or uint c0 = fun(p0). I would also suggest an = this example, but it would appear dmd is unable to inline a field if its object has a member function (https://compiler-explorer.com/z/PrxjnzcE3)
4. Please target stable branch.

[ghost]

@ibuclaw ok, expected for the this test. If I change the inline cost of the ThisExp, your test compiles but that might cause unwanted compiler slowdowns.

[ibuclaw]

@ibuclaw ok, expected for the this test. If I change the inline cost of the ThisExp, your test compiles but that might cause unwanted compiler slowdowns.

GDC doesn't use the front-end inliner, so it "just works" for that compiler. The use of inlineCopy here for default arguments is an odd choice, because "inline" doesn't appear anywhere in user code (I forgot I once tried to make DMD exactly same as GDC a few years ago #14309).

If inlineCopy really is unavoidable for DMD, then it should ignore cost when it comes to copying default arguments.

A separate issue can be used to track this though, not here.

[ibuclaw]

@baz-bg thanks for tackling this by the way.

Default arguments is a bit of a "left behind" feature that has not been kept up to date with other development in the language - no safety, purity, nogc checks, etc.