Refactoring test runner #180
Conversation
💚 Build Succeeded
Expand to view the summary
Build stats
Test stats 🧪
|
| return r.run() | ||
| } | ||
|
|
||
| func (r runner) TearDown() { |
There was a problem hiding this comment.
nit: it may be hard to debug issues with TearDown if it starts to depend on state and it's not r *runner) anymore.
There was a problem hiding this comment.
Changed to pointer receivers in 7de930c.
internal/testrunner/testrunner.go
Outdated
|
|
||
| // TearDown cleans up any test runner resources. It must be called | ||
| // after the test runner has finished executing. | ||
| TearDown() |
There was a problem hiding this comment.
is it fair for TearDown method to return an error?
There was a problem hiding this comment.
Yeah, I had it returning an error at one point but looking at the current implementations of TearDown, none of them actually return an error so I took it out.
But, since you're also thinking about it, I'm once again leaning towards allowing TearDown to return an error. I will add it back, thanks!
There was a problem hiding this comment.
Added in 18bb93e.
This PR defines a
TestRunnerinterface. All test runners must implement this interface. This will give consumers of test runners access to their lifecycle methods (RunandTearDown) as well as to informational methods (TypeandString).Related: #175 (comment)
Related: #168