[SEMA] SR-6207: Fix misleading error for unqualified use of static variable from instance method #12644
Conversation
OscarSwanros
changed the title
|
I think this is ready for review. |
| "static member %1 cannot be used on instance of type %0", | ||
| (Type, DeclName)) | ||
| ERROR(could_not_use_type_member_on_instance_without_name,none, | ||
| "static element %0 cannot be referenced as an instance member", |
There was a problem hiding this comment.
Nit: static member is right here.
There was a problem hiding this comment.
While we’re at it, this doesn’t seem much clearer than what was there before
There was a problem hiding this comment.
Just following @jrose-apple's:
The starter bug form of this would be to only include the name if the base expression was a simple DeclRefExpr, or perhaps also a MemberRefExpr
I'm open to suggestions
There was a problem hiding this comment.
IMO, it's clearer because it directly contrasts static with instance members without any extraneous information, but agree that static member is the correct term.
There was a problem hiding this comment.
Sorry, my "only" was ambiguous! I meant "this should be checked in all circumstances, but don't worry about printing the name of the member if it's not easy to do so". I like Huon's suggested wording in the bug report.
test/Constraints/members.swift
Outdated
| @@ -1,3 +1,4 @@ | |||
|
|
|||
There was a problem hiding this comment.
Removed the original whitespace at the top but Github is confused about it.
|
My laptop is going for keyboard repair, so I may be unable to code some more here. Anyone has any further comments or a pointer on how to move forward? |
|
The diagnostics model has moved on since this was written. I'm trying to rebase this locally, but there's a merge conflict on pretty much every commit now. Could you please resubmit this after having rebased it and updated to the new model? If you need some help with that, please let me or @xedin know in SR-6207. Sorry for not reviewing this sooner! |
4 participants
I'm trying to check if the expression is either a MemberRefExpr or a DeclRefExpr, as per @jrose-apple's comment on the ticket. However, I don't think this is the right place/way to do it. Opened the PR just to keep track of things and maybe get some pointers from people more familiar with the matter.Checks for a DeclRefExpr or a MembeRefExpr in 8d58457 now
Resolves #48759.