A preliminary exploration of Goroutine and channel in Go language

This article will give you a preliminary understanding of Goroutine and channel in the Go language. I hope it will be helpful to you!

The implementation of the CSP concurrency model of the Go language contains two main components: one is Goroutine and the other ischannel. This article will introduce their basic usage and precautions.

Goroutine

Goroutine is the basic execution unit of the Go application. It is a lightweight user-level thread , the bottom layer is concurrency achieved through coroutine (coroutine). As we all know, a coroutine is a user thread running in user mode, so Goroutine is also scheduled when the Go program is running.

Basic usage

Syntax: go function/method

You can create a ## through go keyword function/method #Goroutine.

Code example:

import (
   "fmt"
   "time"
)

func printGo() {
   fmt.Println("具名函数")
}

type G struct {
}

func (g G) g() {
   fmt.Println("方法")
}

func main() {
   // 基于具名函数创建 goroutine
   go printGo()
   // 基于方法创建 goroutine
   g := G{}
   go g.g()
   // 基于匿名函数创建 goroutine
   go func() {
      fmt.Println("匿名函数")
   }()
   // 基于闭包创建 goroutine
   i := 0
   go func() {
      i++
      fmt.Println("闭包")
   }()
   time.Sleep(time.Second) // 避免 main goroutine 结束后，其创建的 goroutine 来不及运行，因此在此休眠 1 秒
}

Execution result:

闭包
具名函数
方法
匿名函数

When multiple

Goroutine exist, their execution order is not fixed. Therefore, the results will be different every time you print.

As can be seen from the code, through the

go keyword, we can create goroutine based on the named function / method, also goroutine can be created based on anonymous functions/closures.

So how does

Goroutine exit? Under normal circumstances, as long as the execution of the Goroutine function ends or the execution returns, it means the exit of Goroutine. If Goroutine's function or method has a return value, it will be ignored when Goroutine exits.

channel

channel plays an important role in the Go concurrency model. It can be used to implement communication between Goroutine, and can also be used to implement synchronization between Goroutine.

Basic operations of channel

#channel is a composite data type. When declaring, you need to specify the elements in channel type.

Declaration syntax: var ch chan string

Declare a

channel whose element type is string through the above code. Only elements of type string can be stored. channel is a reference type and must be initialized to write data. It is initialized by make.

import (
   "fmt"
)

func main() {
   var ch chan string
   ch = make(chan string, 1)
   // 打印 chan 的地址
   fmt.Println(ch)
   // 向 ch 发送 "Go" 数据
   ch <- "Go"
   // 从 ch 中接收数据
   s := <-ch
   fmt.Println(s) // Go
}

Through

ch <- xxx, you can send data to the channel variable ch, via x := <- ch Data can be received from the channel variable ch.

Buffered channel and unbuffered channel

If the capacity is not specified when initializing the

channel, an unbuffered # will be created. ##channel: <pre class="brush:js;toolbar:false;">ch := make(chan string)</pre>The sending and receiving operations of the unbuffered

channel

are synchronous. After the send operation is performed, the corresponding Goroutine will block. , until there is another Goroutine to perform the receive operation, and vice versa. What will happen if the send operation and execution operation are placed under the same Goroutine? Take a look at the following code: <pre class="brush:js;toolbar:false;">import ( &quot;fmt&quot; ) func main() { ch := make(chan int) // 发送数据 ch &lt;- 1 // fatal error: all goroutines are asleep - deadlock! // 接收数据 n := &lt;-ch fmt.Println(n) }</pre> After the program is run, you will get

fatal error

at ch <-, prompting all Goroutine In a dormant state, it is deadlocked. To avoid this situation, we need to execute the sending and receiving operations of channel in different Goroutine. <pre class="brush:js;toolbar:false;">import ( &quot;fmt&quot; ) func main() { ch := make(chan int) go func() { // 发送数据 ch &lt;- 1 }() // 接收数据 n := &lt;-ch fmt.Println(n) // 1 }</pre> It can be concluded from the above example: the sending and receiving operations of unbuffered

channel

must be carried out in two different Goroutine, otherwise it will Occurrencedeadlock image.

If the capacity is specified, a buffered

---

channel

is created: <pre class="brush:js;toolbar:false;">ch := make(chan string, 5)</pre>Buffered

channel

and unbuffered chennel is different. When performing a send operation, as long as the buffer of channel is not full, Goroutine will not hang until the buffer is full. channel Performing a send operation will cause Goroutine to hang. Code example: <pre class="brush:js;toolbar:false;">func main() { ch := make(chan int, 1) // 发送数据 ch &lt;- 1 ch &lt;- 2 // fatal error: all goroutines are asleep - deadlock! }</pre>

Declare the send-only type and receive-only type of channel

• channel# that can both send and receive

  ##

  ch := make(chan int, 1)

  The channel
  variable is obtained through the above code, and we can perform sending and receiving operations on it.
  Only receiving
channel

• ch := make(<-chan int, 1)

  The channel
  variable is obtained through the above code, we can only receive it .
  Only sent
channel

• ch := make(chan<- int, 1)

  The channel
  variable is obtained through the above code, we can only send it .

通常只发送 channel 类型和只接收 channel 类型，会被用作函数的参数类型或返回值：

func send(ch chan<- int) {
   ch <- 1
}

func recv(ch <-chan int) {
   <-ch
}

channel 的关闭

通过内置函  close(c chan<- Type)，可以对 channel 进行关闭。

• 在发送端关闭 channel

  在 channel 关闭之后，将不能对 channel 执行发送操作，否则会发生 panic，提示 channel 已关闭。

  func main() {
     ch := make(chan int, 5)
     ch <- 1
     close(ch)
     ch <- 2 // panic: send on closed channel
  }

• 管道 channel 之后，依旧可以对 channel 执行接收操作，如果存在缓冲区的情况下，将会读取缓冲区的数据，如果缓冲区为空，则获取到的值为 channel 对应类型的零值。

  import "fmt"
  
  func main() {
     ch := make(chan int, 5)
     ch <- 1
     close(ch)
     fmt.Println(<-ch) // 1
     n, ok := <-ch
     fmt.Println(n)  // 0
     fmt.Println(ok) // false
  }

• 如果通过 for-range 遍历 channel 时，中途关闭 channel 则会导致 for-range 循环结束。

小结

本文首先介绍了 Goroutine的创建方式以及其退出的时机是什么。

其次介绍了如何创建 channel 类型变量的有缓冲与无缓冲的创建方式。需要注意的是，无缓冲的 channel 发送与接收操作，需要在两个不同的 Goroutine 中执行，否则会发送 error。

接下来介绍如何定义只发送和只接收的 channel 类型。通常只发送 channel 类型和只接收 channel 类型，会被用作函数的参数类型或返回值。

最后介绍了如何关闭 channel，以及关闭之后的一些注意事项。

【相关推荐：Go视频教程、编程教学】

The above is the detailed content of A preliminary exploration of Goroutine and channel in Go language. For more information, please follow other related articles on the PHP Chinese website!