Remove support for base access with explicit base type. #35456
Conversation
| ERR_NotDeclaredInBase = 8710, | ||
| //ERR_NotBaseOrImplementedInterface = 8708, | ||
| //ERR_NotImplementedInBase = 8709, | ||
| //ERR_NotDeclaredInBase = 8710, | ||
|
|
There was a problem hiding this comment.
I think it may be better to just delete these, since it will likely be inappropriate to include these diagnostics in the C# 8.0 range of codes if it is introduced in a later language version.
There was a problem hiding this comment.
One possible reason to keep these error codes is to make it easier to revert this PR later on (to re-introduce the feature). We can reserve those error codes.
In reply to: 280623450 [](ancestors = 280623450)
|
@dotnet/roslyn-compiler Please review, need a second sign-off |
| boundType = new BoundTypeExpression(node.TypeClause.BaseType, alias, baseType, hasErrors); | ||
| } | ||
|
|
||
| TypeSymbol baseType = this.ContainingType is null ? null : this.ContainingType.BaseTypeNoUseSiteDiagnostics; | ||
| bool inStaticContext; |
There was a problem hiding this comment.
inStaticContext [](start = 17, length = 15)
nit: consider inlining declaration
| @@ -4391,144 +4391,5 @@ public void BaseExpression_01() | |||
| } | |||
| EOF(); | |||
| } | |||
|
|
|||
| [Fact] | |||
There was a problem hiding this comment.
Fact [](start = 9, length = 4)
General question about this PR:
I don't think the removal of base(Something) is a breaking change, since the feature has not RTM'ed. But if anyone has been using this feature in preview, their code will break. Does that rise to the level where we want to add a note to breaking changes?
There was a problem hiding this comment.
The feature became available only a couple of weeks ago in Preview 3 of VS.
In reply to: 280927546 [](ancestors = 280927546)
Related to #32054.