mirror of https://github.com/pulumi/pulumi.git
11eff25a20
Long term I want to introduce a new snake_case based naming system to schemas. There's an internal design doc about this at https://docs.google.com/document/d/1ns07htpLjw0dJPn5p7TBPVULvwwYc1Pl7l1c1jdGXoY/edit. As part of that we get the nice feature that _nearly_ all current names are either valid and the same in both system (e.g. "foo"), or they look different in the current system (e.g. "fooBar"). But there are a handful of odd cases where we have providers using names that look like they should be new style snake case names (e.g. kubernetes has a load of "x_kubernetes_" properties). To provide a way to ensure k8s can stay on the current name generation (even though it gives pretty odd results) this change adds a test with a schema demonstrating a load of snake_case names. When we do add the new naming system this test _should not diff_, excepting possibly setting an option if the new naming system is opt-out rather than opt-in. |
||
|---|---|---|
| .. | ||
| pulumi_legacy_names | ||
| codegen-manifest.json | ||
| setup.py | ||