使用context实现多个goroutine的依赖管理

flybywind · · 2087 次点击 · · 开始浏览    
这是一个创建于 的文章,其中的信息可能已经有所发展或是发生改变。

解决的问题

在很多实际情况,比如处理网络请求时,我们需要启动多个goroutine来处理不同的逻辑,比如一个主要的goroutine用来响应请求,生成网页,同时它还启动一个子线程用来获取数据库信息,还有一个则写日志等等。正常情况都没有问题,但是一旦出现异常,如何优雅的退出这些子线程,同时释放掉可能占用的资源呢?

context

在golang中,人们发明了context接口处理这种情况。早在14年,这个库就出现了,并且提出了基于context的并发编程范式(英文好的同学可以直接撸这篇文章)。
今年8月go1.7发布后,它正式成为了标准库的一员。

如何使用

在golang的context库中,首先定义了context的接口,然后给出了context接口的4种实现:

  • WithCancel(parent Context) (Context, CancelFunc)
    初始化一个可以被cancel的context,同时把新context对象作为child放入parent的children数组中。当parent终止时,child也会接受到信号。这个过程叫propagateCancel

  • WithDeadline(parent Context, deadline time.Time) (Context, CancelFunc)
    同样初始化一个context,除了实现跟WithCancel同样的功能外,还增加了一个时间变量,一旦当前时间超过这个deadline,那么这个context以及它的所有子孙都被被cancel。

  • WithTimeout(parent Context, timeout time.Duration) (Context, CancelFunc)
    WithDeadline类似,如果说WithDeadline是一个绝对时间上的限制,那么WithTimeout就是一个相对时间的限制

  • WithValue(parent Context, key, val interface{}) Context
    单纯给parent增加value,不需要propagateCancel。value可以用来跨进程、跨api的传递数据,最好是和某个请求相关的参数,不要传递太多大量数据。

所以关键就在于propagateCancel,实际工程中,所有context共同组成了一个依赖树,他们都继承自一个祖先。一旦parent被cancel,就会通过propagateCancel递归的传播给下面的所有子孙。可以看出,context就好比信使,或者说通讯协议,通过遵循context接口构建的这个框架,能够保证子线程及时获得与他相关的父线程的状态,从而由子线程根据情况作出反应。至于怎么反应,就取决于各位码农的能力和搬砖当时的心情了。。。

另外,golang有一套静态分析工具可以分析context的传播过程,所以为了方便这个工具的使用,实际使用中有几个规定:

  • 不要把context作为struct内部变量使用,而是把它和其他变量一块作为参数传入下一个函数。

  • context变量需要作为函数的第一个参数传入,命名一般为ctx

具体例子

这个例子来源于基于context的并发编程范式,但是为了符合国情我做了些修改:
包括3部分:

  • server.go
    主线程,会创建一个server服务器,可以通过localhost:9090/search访问。接到请求后,它会创建父context,同时生成一个新goroutine,去fakesrv(本来应该去google上的)上请求数据。

  • google.go
    替换原来的google网址,改成由fakesrv提供的一个网址。主要就是演示一下context的运行过程,请求fakesrv的工作在一个新goroutine中进行,同时它还有一个访问数据库的操作。如果父context因为timeout超时了,那么对fakesrv和数据库的访问也会终止。在代码中,演示了如何监听context信息的过程。

  • query.go
    解析url中的query参数

  • fakesrv.go
    提供http://localhost:9000/context...供google.go访问。

mycontext/serve.go

// The server program issues Google search requests and demonstrates the use of
// the go.net Context API. It serves on port 8080.
//
// The /search endpoint accepts these query params:
//   q=the Google search query
//   timeout=a timeout for the request, in time.Duration format
//
// For example, http://localhost:8080/search?q=golang&timeout=1s serves the
// first few Google search results for "golang" or a "deadline exceeded" error
// if the timeout expires.
package main

import (
    "html/template"
    "log"
    "net/http"
    "time"

    "context"
    "mycontext/google"
    "mycontext/query"
)

func main() {
    http.HandleFunc("/search", handleSearch)
    log.Fatal(http.ListenAndServe(":9090", nil))
}

// handleSearch handles URLs like /search?q=golang&timeout=1s by forwarding the
// query to google.Search. If the query param includes timeout, the search is
// canceled after that duration elapses.
func handleSearch(w http.ResponseWriter, req *http.Request) {
    // ctx is the Context for this handler. Calling cancel closes the
    // ctx.Done channel, which is the cancellation signal for requests
    // started by this handler.
    var (
        ctx    context.Context
        qctx   *query.QueryCtx
        cancel context.CancelFunc
    )
    timeout, err := time.ParseDuration(req.FormValue("timeout"))
    if err == nil {
        // The request has a timeout, so create a context that is
        // canceled automatically when the timeout expires.
        ctx, cancel = context.WithTimeout(context.Background(), timeout)
    } else {
        ctx, cancel = context.WithCancel(context.Background())
    }
    defer cancel() // Cancel ctx as soon as handleSearch returns.
    qctx, err = query.NewQueryCtx(ctx, req)
    if err != nil {
        http.Error(w, "no query", http.StatusBadRequest)
        return
    }

    // Run the Google search and print the results.
    start := time.Now()
    results, err := google.Search(qctx)
    elapsed := time.Since(start)
    if err != nil {
        http.Error(w, err.Error(), http.StatusInternalServerError)
        return
    }
    if err := resultsTemplate.Execute(w, struct {
        Results          google.Results
        Timeout, Elapsed time.Duration
    }{
        Results: results,
        Timeout: timeout,
        Elapsed: elapsed,
    }); err != nil {
        log.Print(err)
        return
    }
}

var resultsTemplate = template.Must(template.New("results").Parse(`
<html>
<head/>
<body>
  <ol>
  {{range .Results}}
    <li>{{.Title}} - <span>{{.SubTitle}}</span></li>
  {{end}}
  </ol>
  <p>{{len .Results}} results in {{.Elapsed}}; timeout {{.Timeout}}</p>
</body>
</html>
`))

mycontext/google/google.go

// Package google provides a function to do Google searches using the Google Web
// Search API. See https://developers.google.com/web-search/docs/
//
// This package is an example to accompany https://blog.golang.org/context.
// It is not intended for use by others.
//
// Google has since disabled its search API,
// and so this package is no longer useful.
package google

import (
    "context"
    "encoding/json"
    "log"
    "mycontext/query"
    "net/http"
    "time"
)

// Results is an ordered list of search results.
type Results []Result

// A Result contains the title and URL of a search result.
type Result struct {
    Title, SubTitle string
}

// Search sends query to Google search and returns the results.
func Search(ctx *query.QueryCtx) (Results, error) {
    // Prepare the Google Search API request.
    req, err := http.NewRequest("GET", "http://localhost:9000/context_demo", nil)
    if err != nil {
        return nil, err
    }

    ctx.SetReq(req)
    // Issue the HTTP request and handle the response. The httpDo function
    // cancels the request if ctx.Done is closed.
    var results Results
    err = httpDo(ctx, req, func(resp *http.Response, err error) error {
        if err != nil {
            return err
        }
        defer resp.Body.Close()

        // Parse the JSON search result.
        // https://developers.google.com/web-search/docs/#fonje
        var data struct {
            ResponseData struct {
                Results []struct {
                    Title, SubTitle string
                }
            }
        }
        if err := json.NewDecoder(resp.Body).Decode(&data); err != nil {
            return err
        }
        for _, res := range data.ResponseData.Results {
            results = append(results, Result{Title: res.Title, SubTitle: res.SubTitle})
        }
        return nil
    })
    // httpDo waits for the closure we provided to return, so it's safe to
    // read results here.
    return results, err
}

// httpDo issues the HTTP request and calls f with the response. If ctx.Done is
// closed while the request or f is running, httpDo cancels the request, waits
// for f to exit, and returns ctx.Err. Otherwise, httpDo returns f's error.
func httpDo(ctx *query.QueryCtx, req *http.Request, f func(*http.Response, error) error) error {
    // Run the HTTP request in a goroutine and pass the response to f.
    tr := &http.Transport{}
    client := &http.Client{Transport: tr}
    // WithCancel会在ctx的children中增加cancelDb,这样当
    // ctx 结束的时候,cancelDb也会受到消息
    cancelDb, cancel := context.WithCancel(ctx.Context)
    defer cancel()
    c := make(chan error, 1)
    go func() { c <- f(client.Do(req)) }()
    go func(ctx context.Context) {
        t := time.NewTimer(2 * time.Second)

        select {
        case <-t.C:
            log.Println("db access finished!")
        case <-ctx.Done():
            log.Println("canceld by parent, release resource")
        }
    }(cancelDb)
    select {
    case <-ctx.Done():
        tr.CancelRequest(req)
        <-c // Wait for f to return.
        return ctx.Err()
    case err := <-c:
        return err
    }
}

mycontext/query/query.go

package query

import (
    "context"
    "fmt"
    "net/http"
)

func NewQueryCtx(ctx context.Context, req *http.Request) (*QueryCtx, error) {
    q := req.FormValue("q")
    if q == "" {
        return nil, fmt.Errorf("no query supplied!")
    }
    return &QueryCtx{ctx, q}, nil
}

type QueryCtx struct {
    context.Context
    val string
}

func (ctx *QueryCtx) SetReq(req *http.Request) {
    q := req.URL.Query()
    q.Set("q", ctx.val)

    req.URL.RawQuery = q.Encode()
}

mycontext/fakesrv/main.go

package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "log"
    "math/rand"
    "net/http"
    "strconv"
    "strings"
    "time"
)

func init() {
    log.SetFlags(log.Lshortfile)
}

type Results struct {
    ResponseData struct {
        Results []Content
    }
}

// A Result contains the title and URL of a search result.
type Content struct {
    Title, SubTitle string
}

func main() {

    http.HandleFunc("/context_demo", handleContext)
    http.ListenAndServe(":9000", nil)
}

func handleContext(resp http.ResponseWriter, req *http.Request) {
    defer func() {
        if e := recover(); e != nil {
            if msg, ok := e.(string); ok {
                resp.Write([]byte(msg))
            } else {
                panic(e)
            }
        }
    }()
    check_error := func(err error, msg string) {
        if err != nil {
            if msg != "" {
                panic(err.Error() + ":" + msg)
            } else {
                panic(err.Error())
            }
        }
    }
    if req.Method == "GET" {
        q := req.FormValue("q")
        seg := strings.Split(q, ":")
        if len(seg) < 2 {
            log.Println("query format wrong")
            resp.Write([]byte("query format wrong"))
            return
        }
        title := seg[0]
        num, err := strconv.Atoi(seg[1])
        check_error(err, "")
        rs := Results{}
        for i := 0; i < num; i++ {
            rs.ResponseData.Results = append(rs.ResponseData.Results,
                Content{fmt.Sprintf("%s %d", title, i), RandomString(20)})
        }
        buff := bytes.NewBuffer(nil)
        err = json.NewEncoder(buff).Encode(rs)
        check_error(err, "")
        time.Sleep(time.Second * 2)
        resp.Write(buff.Bytes())
    } else {
        resp.Write([]byte("请使用get方法!"))
    }
}
func RandomString(strlen int) string {
    rand.Seed(time.Now().UTC().UnixNano())
    const chars = "abcdefghijklmnopqrstuvwxyz0123456789"
    result := make([]byte, strlen)
    for i := 0; i < strlen; i++ {
        result[i] = chars[rand.Intn(len(chars))]
    }
    return string(result)
}

Makefile

run:
    go build 
    ./mycontext &
    cd fakesrv && go build && ./fakesrv &

test:
    @echo "======= test without timeout ======="
    curl localhost:9090/search?q=title:6
    @echo "======= test with timeout 1s ======="
    curl localhost:9090/search?q=title:6\&timeout=1s
    @echo "======= test with timeout 4s ======="
    curl localhost:9090/search?q=title:6\&timeout=4s

测试

在命令行运行如下命令,即可看到具体结果
make run
make test


有疑问加站长微信联系(非本文作者)

本文来自:Segmentfault

感谢作者:flybywind

查看原文:使用context实现多个goroutine的依赖管理

入群交流(和以上内容无关):加入Go大咖交流群,或添加微信:liuxiaoyan-s 备注:入群;或加QQ群:692541889

2087 次点击  
加入收藏 微博
暂无回复
添加一条新回复 (您需要 登录 后才能回复 没有账号 ?)
  • 请尽量让自己的回复能够对别人有帮助
  • 支持 Markdown 格式, **粗体**、~~删除线~~、`单行代码`
  • 支持 @ 本站用户;支持表情(输入 : 提示),见 Emoji cheat sheet
  • 图片支持拖拽、截图粘贴等方式上传