Brian Wilkerson
Brian Wilkerson
> I would prefer it if we had a generic way of marking mutually exclusive arguments ... The first option that comes to mind (though maybe not the best) would...
If we define an annotation then we'll implement, warning rather than lints, and we can move this issue to the 'sdk' repository. But for now this is a reasonable place...
It would be nice to be a little more explicit by describing when they can be avoided, or at least the conditions under which this lint will fire. Is the...
Either way, do we want it to catch more complex cases, such as ``` void f(int? x, int? y) { if (x != null && y != null) { x!...
If we have multiple lints that would benefit from knowing what the context type is during inference, then it seems likely that other lint authors might want that information as...
Unless I'm misunderstanding what you're asking for, I don't believe that we could associate the context type with the element model. The context type is associated with a particular location...
I suspect the bug is in `LintContext.canBeConst`. It probably doesn't correctly detect that `if` and `for` elements can't be in a `const` literal.
@goderbauer
We could potentially add one or more annotations that would let us cover this case. @pq was looking at a similar situation awhile back. We had a client that wanted...
Or the lint should be changed to use a case-insensitive comparison. @pq