在那里我和GO有一些乐趣,我对我想要达到的目标非常好奇。我这里有一个包,只是从Reddit得到一个特殊的饲料。当我收到父JSON文件时,我想要检索子数据。如果您看到下面的代码,我会启动一系列goroutines然后阻止,等待它们使用sync
包完成。我想要的是,一旦第一系列goroutines使用之前的结果完成第二系列goroutines。有一些我正在考虑如loop和switch语句。但是,最好和最有效的方法是什么
func (m redditMatcher) retrieve(dataPoint *collect.DataPoint) (*redditCommentsDocument, error) {
if dataPoint.URI == "" {
return nil, errors.New("No datapoint uri provided")
}
// Get options data -> returns empty struct
// if no options are present
options := m.options(dataPoint.Options)
if len(options.subreddit) <= 0 {
return nil, fmt.Errorf("Matcher fail: Reddit - Subreddit option manditory\n")
}
// Create an buffered channel to receive match results to display.
results := make(chan *redditCommentsDocument, len(options.subreddit))
// Generte requests for each subreddit produced using
// goroutines concurency model
for _, s := range options.subreddit {
// Set the number of goroutines we need to wait for while
// they process the individual subreddit.
waitGroup.Add(1)
go retrieveComment(s.(string), dataPoint.URI, results)
}
// Launch a goroutine to monitor when all the work is done.
waitGroup.Wait()
// HERE I WOULD TO CALL ANOTHER SERIES OFF GOROUTINES
for commentFeed := range results {
// HERE I WOULD LIKE TO CALL GO ROUTINES USING THE RESULTS
// PROVIDED FROM THE PREVIOUS FUNCTIONS
waitGroup.Add(1)
log.Printf("%s\n\n", commentFeed.Kind)
}
waitGroup.Wait()
close(results)
return nil, nil
}
答案 0 :(得分:1)
如果你想等待所有第一个系列完成,那么你可以直接传入一个指向你的waitgroup的指针,在调用所有第一个系列函数之后等待(在waitgroup上调用Done()
) ,然后开始第二个系列。这是一个可运行的带注释的代码示例:
package main
import(
"fmt"
"sync"
"time"
)
func first(wg *sync.WaitGroup) {
defer wg.Done()
fmt.Println("Starting a first")
// do some stuff... here's a sleep to make some time pass
time.Sleep(250 * time.Millisecond)
fmt.Println("Done with a first")
}
func second(wg *sync.WaitGroup) {
defer wg.Done()
fmt.Println("Starting a second")
// do some followup stuff
time.Sleep(50 * time.Millisecond)
fmt.Println("Done with a second")
}
func main() {
wg := new(sync.WaitGroup) // you'll need a pointer to avoid a copy when passing as parameter to goroutine function
// let's start 5 firsts and then wait for them to finish
wg.Add(5)
go first(wg)
go first(wg)
go first(wg)
go first(wg)
go first(wg)
wg.Wait()
// now that we're done with all the firsts, let's do the seconds
// how about two of these
wg.Add(2)
go second(wg)
go second(wg)
wg.Wait()
fmt.Println("All done")
}
输出:
Starting a first
Starting a first
Starting a first
Starting a first
Starting a first
Done with a first
Done with a first
Done with a first
Done with a first
Done with a first
Starting a second
Starting a second
Done with a second
Done with a second
All done
但是如果你想在“第一次”完成后立即启动“秒”,那么在第一次运行时,只需要在通道上执行阻止接收运算符的秒数:
package main
import(
"fmt"
"math/rand"
"sync"
"time"
)
func first(res chan int, wg *sync.WaitGroup) {
defer wg.Done()
fmt.Println("Starting a first")
// do some stuff... here's a sleep to make some time pass
time.Sleep(250 * time.Millisecond)
fmt.Println("Done with a first")
res <- rand.Int() // this will block until a second is ready
}
func second(res chan int, wg *sync.WaitGroup) {
defer wg.Done()
fmt.Println("Wait for a value from first")
val := <-res // this will block until a first is ready
fmt.Printf("Starting a second with val %d\n", val)
// do some followup stuff
time.Sleep(50 * time.Millisecond)
fmt.Println("Done with a second")
}
func main() {
wg := new(sync.WaitGroup) // you'll need a pointer to avoid a copy when passing as parameter to goroutine function
ch := make(chan int)
// lets run first twice, and second once for each first result, for a total of four workers:
wg.Add(4)
go first(ch, wg)
go first(ch, wg)
// don't wait before starting the seconds
go second(ch, wg)
go second(ch, wg)
wg.Wait()
fmt.Println("All done")
}
哪个输出:
Wait for a value from first
Starting a first
Starting a first
Wait for a value from first
Done with a first
Starting a second with val 5577006791947779410
Done with a first
Starting a second with val 8674665223082153551
Done with a second
Done with a second
All done