mirror of
https://github.com/rancher/plugins.git
synced 2025-09-06 23:05:23 +00:00
Update ginkgo to v2 in go.mod, go.sum, vendor
This commit updates ginkgo to v2. Note that because ginkgo/v2 requires go1.18, it was updated as well. Signed-off-by: liornoy <lnoy@redhat.com> Co-authored-by: Sascha Grunert <sgrunert@redhat.com>
This commit is contained in:
90
vendor/github.com/onsi/ginkgo/v2/internal/spec_context.go
generated
vendored
Normal file
90
vendor/github.com/onsi/ginkgo/v2/internal/spec_context.go
generated
vendored
Normal file
@@ -0,0 +1,90 @@
|
||||
package internal
|
||||
|
||||
import (
|
||||
"context"
|
||||
"sort"
|
||||
"sync"
|
||||
|
||||
"github.com/onsi/ginkgo/v2/types"
|
||||
)
|
||||
|
||||
type SpecContext interface {
|
||||
context.Context
|
||||
|
||||
SpecReport() types.SpecReport
|
||||
AttachProgressReporter(func() string) func()
|
||||
}
|
||||
|
||||
type specContext struct {
|
||||
context.Context
|
||||
|
||||
cancel context.CancelFunc
|
||||
lock *sync.Mutex
|
||||
progressReporters map[int]func() string
|
||||
prCounter int
|
||||
|
||||
suite *Suite
|
||||
}
|
||||
|
||||
/*
|
||||
SpecContext includes a reference to `suite` and embeds itself in itself as a "GINKGO_SPEC_CONTEXT" value. This allows users to create child Contexts without having down-stream consumers (e.g. Gomega) lose access to the SpecContext and its methods. This allows us to build extensions on top of Ginkgo that simply take an all-encompassing context.
|
||||
|
||||
Note that while SpecContext is used to enforce deadlines by Ginkgo it is not configured as a context.WithDeadline. Instead, Ginkgo owns responsibility for cancelling the context when the deadline elapses.
|
||||
|
||||
This is because Ginkgo needs finer control over when the context is canceled. Specifically, Ginkgo needs to generate a ProgressReport before it cancels the context to ensure progress is captured where the spec is currently running. The only way to avoid a race here is to manually control the cancellation.
|
||||
*/
|
||||
func NewSpecContext(suite *Suite) *specContext {
|
||||
ctx, cancel := context.WithCancel(context.Background())
|
||||
sc := &specContext{
|
||||
cancel: cancel,
|
||||
suite: suite,
|
||||
lock: &sync.Mutex{},
|
||||
prCounter: 0,
|
||||
progressReporters: map[int]func() string{},
|
||||
}
|
||||
ctx = context.WithValue(ctx, "GINKGO_SPEC_CONTEXT", sc) //yes, yes, the go docs say don't use a string for a key... but we'd rather avoid a circular dependency between Gomega and Ginkgo
|
||||
sc.Context = ctx //thank goodness for garbage collectors that can handle circular dependencies
|
||||
|
||||
return sc
|
||||
}
|
||||
|
||||
func (sc *specContext) SpecReport() types.SpecReport {
|
||||
return sc.suite.CurrentSpecReport()
|
||||
}
|
||||
|
||||
func (sc *specContext) AttachProgressReporter(reporter func() string) func() {
|
||||
sc.lock.Lock()
|
||||
defer sc.lock.Unlock()
|
||||
sc.prCounter += 1
|
||||
prCounter := sc.prCounter
|
||||
sc.progressReporters[prCounter] = reporter
|
||||
|
||||
return func() {
|
||||
sc.lock.Lock()
|
||||
defer sc.lock.Unlock()
|
||||
delete(sc.progressReporters, prCounter)
|
||||
}
|
||||
}
|
||||
|
||||
func (sc *specContext) QueryProgressReporters() []string {
|
||||
sc.lock.Lock()
|
||||
keys := []int{}
|
||||
for key := range sc.progressReporters {
|
||||
keys = append(keys, key)
|
||||
}
|
||||
sort.Ints(keys)
|
||||
reporters := []func() string{}
|
||||
for _, key := range keys {
|
||||
reporters = append(reporters, sc.progressReporters[key])
|
||||
}
|
||||
sc.lock.Unlock()
|
||||
|
||||
if len(reporters) == 0 {
|
||||
return nil
|
||||
}
|
||||
out := []string{}
|
||||
for _, reporter := range reporters {
|
||||
out = append(out, reporter())
|
||||
}
|
||||
return out
|
||||
}
|
Reference in New Issue
Block a user