Add calledOnceWith and calledOnceWithExactly#117
Add calledOnceWith and calledOnceWithExactly#117domenic merged 1 commit intochaijs:masterfrom hexpunk:feature/116
Conversation
lib/sinon-chai.js
Outdated
| sinonMethod("calledWith", "been called with arguments %*", "%D"); | ||
| sinonMethod("calledOnceWith", "been called exactly once with arguments %*", "%C"); | ||
| sinonMethod("calledWithExactly", "been called with exact arguments %*", "%D"); | ||
| sinonMethod("calledOnceWithExactly", "been called exactly once with exact arguments %*", "%C"); |
There was a problem hiding this comment.
So, here's the thing about these, specifically their nonNegatedSuffixes. There's really no good way to address these new ones.
Using the ", but it was called %c%C" suffix doesn't make sense when it's only been called once. And using "%D" causes some odd formatting issues in the case of multiple calls.
With "%D", it ends up looking like this:
AssertionError: expected spy to have been called with exact arguments A, BCall 1:
C A
A B
Call 2:
C A
B
I'm about 99.999% sure this is a sinon formatter issue and not an issue with sinon-chai.
There was a problem hiding this comment.
This should be resolved by this ticket: sinonjs/sinon#1717
I'm still open to suggestions on how these nonNegatedSuffixes should be.
There was a problem hiding this comment.
I think just using "%D" should work pretty well, after that Sinon bug is fixed at least. Do you agree?
package.json
Outdated
| "chai": "^4.1.0", | ||
| "eslint": "^3.19.0", | ||
| "chai": "^4.1.2", | ||
| "eslint": "^4.18.2", |
There was a problem hiding this comment.
Let's leave dev dependency updates for a separate PR, and only update Sinon in this PR.
domenic
left a comment
domenic
left a comment
There was a problem hiding this comment.
Can you add a few messages tests too?
|
LGTM, but can you regenerate package-lock.json? I don't think you rolled back the changes there. |
|
Good call. I did it all via Github. |
|
What do you mean by "messages tests"? As in testing the formatting of the failure messages? |
|
Yeah, see messages.js. Historically we've had a lot of regressions there (e.g. via Sinon upgrades that make our current messages nonsensical) that those tests have caught. |
|
I'll take a crack at that in the next couple of days when I get time. Thanks for reviewing this. |
|
@domenic Just added some messages tests. Can you please take another look? |
|
Sorry for the massive delay here :(. This will go out in a release today! |
2 participants
This should resolve #116.