Cross-validation of dependent traits

Suppose I have an object with traits 'a', 'b', and 'c', with the constraints that a>b>c and default values a=1, b=10, c=100. I'd like to write a validator that checks this constraint, such that I can assign all three values at once to get a new consistent state (either on creation, like MyObject(a=5, b=6, c=7). How would I write such a validator?

If it's not possible, would it be possible to include in the proposal argument to a validator the new values of any other traits in the same assignment, so I can compare against what the object state will be instead of what it currently is?

The issue names no files, tests, or entry points. Start by tracing trait validation and multi-trait assignment behavior, then determine whether dependent values can be checked against the proposed consistent state; done would require a settled validator design and corresponding documented behavior.

由索引模型根据 Issue 内容生成。