Refactor redactIfNecessary function
#1746
Conversation
It's useful to call `redactIfNecessary` on arbitrary shape types, not only member shapes. It is now an extension function on `Shape`, with a specialization on `MemberShape`, since note that the `sensitive` trait cannot be applied to member shapes [0, 1], so we traverse to its target directly. Ultimately this is just to make the diff of #1342 lighter. [0]: https://awslabs.github.io/smithy/1.0/spec/core/documentation-traits.html#sensitive-trait [1]: https://awslabs.github.io/smithy/1.0/spec/core/documentation-traits.html#sensitive-trait
|
A new generated diff is ready to view.
A new doc preview is ready to view. |
| fun MemberShape.redactIfNecessary(model: Model, safeToPrint: String) = | ||
| model.expectShape(this.target).redactIfNecessary(safeToPrint) | ||
|
|
||
| fun Shape.redactIfNecessary(safeToPrint: String) = |
There was a problem hiding this comment.
I think this could be error prone. For example, if someone calls Shape.redactIfNecessary on a MemberShape, which is possible since a MemberShape is a shape, then that might result in the wrong code generated behavior, right? Perhaps we should just have one function that always takes a model, and do any necessary switching logic based on the type in that.
There was a problem hiding this comment.
Yeah, it crossed my mind that this could be error prone.
if someone calls Shape.redactIfNecessary on a MemberShape, which is possible since a MemberShape is a shape, then that might result in the wrong code generated behavior, right?
Yes. I think there's an argument to be made that it's a mistake on behalf of the caller, since sensitive cannot be applied to member shapes. I only introduced MemberShape.redactIfNecessary as syntax sugar for the caller to not have to traverse to the member's target, but I don't know if I prefer to lay that responsibility on the caller, or to have a single function that always takes in model like you suggest. All 3 ways are reasonable.
…dact-if-necessary-function
david-perez
force-pushed
the
davidpz/refactor-redact-if-necessary-function
branch
from
939fcb6
to
869afa4
Compare
|
A new generated diff is ready to view.
A new doc preview is ready to view. |
david-perez
deleted the
davidpz/refactor-redact-if-necessary-function
branch
It's useful to call
redactIfNecessaryon arbitrary shape types, notonly member shapes.
It is now an extension function on
Shape, with a specialization onMemberShape, since note that thesensitivetrait cannot be appliedto member shapes [0, 1], so we traverse to its target directly.
Ultimately this is just to make the diff of #1342 lighter.
Testing
No diff in generated artifacts.
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.