fix issue 21429 . Allow mixed-in opDispatch overloads

[benjones]

I think I've narrowed down the cause to the area I've edited. There is an overload set containing multiple opDispatch implementations. It think the next step is to resolve the correct on in the ??? block in typesem, but I'm not sure how best to do that. Would appreciate any suggestions there. The Issue also mentions issues with other op overloads, which I haven't looked at yet. I'd like to get something actually working for opDispatch before hunting that down.

[dlang-bot]

Thanks for your pull request and interest in making D better, @benjones! 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 "master + dmd#21444"

[benjones]

It seems like rewriting the expression to e.opDispatch!ident and then running semantic on it to do the resolution would work? But I'm not sure which expression type to use there.

[benjones]

@RazvanN7 or @dkorpel any suggestions here? You two seem to be the most knowledgeable about semantic stuff

[benjones]

OK, I think I got this working for the issue I had. I'll investigate the other overloads like Steven mentioned and am happy for whatever other feedback folks have

[benjones]

OK, opBinary left and right overloads are now included in the fix. I'm assuming similar fixes are needed for opBinaryAssign? Also, I'll look into simplifying the opDispatch logic now that I know that treating an OverloadSet as a TemplateDeclaration seems to work OK here.

[dkorpel]

I'm not familiar with OverloadSet so I'd need to dig in the code a bit before I can say if this approach is right, but if it passes all tests that's a good sign. Operator overload resolution is special because it doesn't do UFCS, otherwise I don't think there are any differences with regular .method calls.

If you're going to fix the issue for all operator overloads, I suggest creating a common function used for all searching of opXXX functions to ensure consistency and making future changes easier.

[benjones]

I'm about to wrap up a fix to opOpAssign. I'll see what similarities there are and what I can extract. I sort of feel like I'm playing whack-a-mole on where else there's an assumption that something is a template, not an overloadSet. Thanks for the feedback!

[benjones]

Skimmed through opover.d and I didn't see any other places where there would be a similar "assume I get a single template from search_function, not an overload set" assumption/problem, so hopefully opDispatch, opBinary, and opOpAssign were the only places.

[benjones]

@thewilsonator any idea who might be good to take a look at this?

[benjones]

Bump

[thewilsonator]

Sorry, I somehow missed this

[thewilsonator]

Could you please rebase or force push to restart the CI