Add initial support for chefspec.#66
Conversation
…pec tests for python::default.
There was a problem hiding this comment.
Is there any reason to test this? I would only expect it to work on the ruby in omnibus chef (and when I add test-kitchen support, you want just a single build in the matrix).
There was a problem hiding this comment.
Different people install Chef different ways. There's also backwards compatibility to think about.
There was a problem hiding this comment.
Yeah, this just makes it much more complex to handle test-kitchen later, since we don't want to run those twice. Can probably make it work with some kind of env var magic (create another matrix axis for TEST_TYPE=spec,kitchen, and have the build script run the correct commands in the correct places).
|
I've not really played much with chefspec in favor of test kitchen, any major reason to go this way? |
|
My understanding, as I've not implemented I think of |
|
There are no community cookbook guidelines as Opscode's policies hold no sway here :-) The problem I've had with chefspec (and why I don't use it personally) is that it mocks out a lot of Chef internally, which fails with Poise-style cookbooks and their multi-level resource structure. |
|
You can create more complex build matrixes that would effectively exercise both |
|
test-kitchen is a little different since it doesn't use the local system at all, it just spins up Rackspace instances and talks to them via SSH :-) This looks good for now though, and having a template for a hybrid chefspec/TK travis run is something I want to do anyway. |
Add initial support for chefspec.
|
Let me know if you need/want any help/input. I need to add |
.travis.yml.Gemfile.libraries/matchers.rbfor chefspec.python::default.