Fixes #44877

This issue is already closed (it was for records) - probably there is another issue for record structs?

This issue is already closed (it was for records) - probably there is another issue for record structs?

Thanks, bad copy-paste :)

We should test that adding fields to record struct reports RudeEditKind.InsertIntoStruct rude edit.

Are there any differences in synthesized members between records and record structs?

Are there any differences in synthesized members between records and record structs?

PrintMembers has different visibility, and there is no Clone$ method or copy constructor, but none of that matters for EnC. If fields were able to be added to structs then we'd at least want to add a test to ensure the semantic edits are right due to the missing copy constructor, but based on how the code is written (ie it enumerates constructors rather than assuming they exist) I don't think it would be an issue.

@tmat is there anything else for this? any other tests you think are important?

One more thing we might want to test is changing record to record struct and vice versa. I'm guessing we already correctly report rude edit.

We do, but I'll add a test anyway.

Thanks!