Fix commit_status problem when testing

[charles7668]

Close #29661

fix #29656

[lunny]

I want to know why the CI PASS when tests in PR. https://github.com/go-gitea/gitea/actions/runs/8198780164/job/22422906368

[charles7668]

I want to know why the CI PASS when tests in PR.

The problem is related to the behavior of the for range loop with map structures in Go. The order of iteration is not the same every time, so not every test will fail.

[silverwind]

I want to know why the CI PASS when tests in PR.

The problem is related to the behavior of the for range loop with map structures in Go. The order of iteration is not the same every time, so not every test will fail.

The code itself does not depend on order, right? If it does we may need something like https://github.com/iancoleman/orderedmap.

[yp05327]

Maybe the test cases can be merged together, instead of using the index?
e.g.

testCases := []struct{
  commitStatus []*git_model.CommitStatus
  required []string
  expected structs.CommitStatusState
}

[charles7668]

I want to know why the CI PASS when tests in PR.

The problem is related to the behavior of the for range loop with map structures in Go. The order of iteration is not the same every time, so not every test will fail.

The code itself does not depend on order, right? If it does we may need something like https://github.com/iancoleman/orderedmap.

Yes , it doesn't depend .

[charles7668]

Maybe the test cases can be merged together, instead of using the index? e.g.

testCases := []struct{
  commitStatus []*git_model.CommitStatus
  required []string
  expected structs.CommitStatusState
}

This sounds like a great change. However, should I make this change in this PR? It seems like an unrelated change.

[silverwind]

I agree, let's do that in another PR.