Back in Go 1.14, the testing package introduced a t.Cleanup method that allows registering multiple callback functions in test code, which are executed in reverse order of registration after the test completes. Looking at its implementation, can you register another Cleanup callback nested inside a Cleanup callback? As of Go 1.15, you cannot.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
|
package testing
// T is a type passed to Test functions to manage test state.
type T struct {
mu sync.RWMutex
cleanup func() // optional function to be called at the end of the test
...
}
// Cleanup registers a function to be called when the test and all its
// subtests complete. Cleanup functions will be called in last added,
// first called order.
func (t *T) Cleanup(f func()) {
t.mu.Lock()
defer t.mu.Unlock()
oldCleanup := t.cleanup
t.cleanup = func() {
if oldCleanup != nil {
defer func() {
...
oldCleanup()
}()
}
...
f()
}
...
}
// runCleanup is called at the end of the test
func (t *T) runCleanup(ph panicHandling) (panicVal interface{}) {
t.mu.Lock()
cleanup := t.cleanup
t.cleanup = nil
t.mu.Unlock()
if cleanup == nil {
return nil
}
...
cleanup()
return nil
}
|
早在 Go 1.14 中，testing 包就引入过一个 t.Cleanup 的方法，允许在测试代码中注册多个回调函数，并以注册顺序的逆序在测试结束后被执行。从其实现来看，你能在一个 Cleanup 里注册的回调中，嵌套注册另一个 Cleanup 吗？现在（1.15）还不能。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
|
package testing
// T is a type passed to Test functions to manage test state.
type T struct {
mu sync.RWMutex
cleanup func() // optional function to be called at the end of the test
...
}
// Cleanup registers a function to be called when the test and all its
// subtests complete. Cleanup functions will be called in last added,
// first called order.
func (t *T) Cleanup(f func()) {
t.mu.Lock()
defer t.mu.Unlock()
oldCleanup := t.cleanup
t.cleanup = func() {
if oldCleanup != nil {
defer func() {
...
oldCleanup()
}()
}
...
f()
}
...
}
// runCleanup is called at the end of the test
func (t *T) runCleanup(ph panicHandling) (panicVal interface{}) {
t.mu.Lock()
cleanup := t.cleanup
t.cleanup = nil
t.mu.Unlock()
if cleanup == nil {
return nil
}
...
cleanup()
return nil
}
|