练习来自:https://tour.golang.org/concurrency/10

描述:

In this exercise you'll use Go's concurrency features to parallelize a web crawler.

Modify the Crawl function to fetch URLs in parallel without fetching the same URL twice.

Hint: you can keep a cache of the URLs that have been fetched on a map, but maps alone are not safe for concurrent use!

这是我的答案:

package main

import (
    "fmt"
    "sync"
)

type Fetcher interface {
    // Fetch returns the body of URL and
    // a slice of URLs found on that page.
    Fetch(url string) (body string, urls []string, err error)
}

var crawledURLs = make(map[string]bool)
var mux sync.Mutex

func CrawlURL(url string, depth int, fetcher Fetcher, quit chan bool) {
    defer func() { quit <- true }()
    if depth <= 0 {
        return
    }

    mux.Lock()
    _, isCrawled := crawledURLs[url]
    if isCrawled {
        return
    }
    crawledURLs[url] = true
    mux.Unlock()

    body, urls, err := fetcher.Fetch(url)
    if err != nil {
        fmt.Println(err)
        return
    }
    fmt.Printf("found: %s %q\n", url, body)
    quitThis := make(chan bool)
    for _, u := range urls {
        go CrawlURL(u, depth-1, fetcher, quitThis)
    }
    for range urls {
        <-quitThis
    }
    return
}

// Crawl uses fetcher to recursively crawl
// pages starting with url, to a maximum of depth.
func Crawl(url string, depth int, fetcher Fetcher) {
    CrawlURL(url, depth, fetcher, make(chan bool))
    return
}

func main() {
    Crawl("https://golang.org/", 4, fetcher)
}

// fakeFetcher is Fetcher that returns canned results.
type fakeFetcher map[string]*fakeResult

type fakeResult struct {
    body string
    urls []string
}

func (f fakeFetcher) Fetch(url string) (string, []string, error) {
    if res, ok := f[url]; ok {
        return res.body, res.urls, nil
    }
    return "", nil, fmt.Errorf("not found: %s", url)
}

// fetcher is a populated fakeFetcher.
var fetcher = fakeFetcher{
    "https://golang.org/": &fakeResult{
        "The Go Programming Language",
        []string{
            "https://golang.org/pkg/",
            "https://golang.org/cmd/",
        },
    },
    "https://golang.org/pkg/": &fakeResult{
        "Packages",
        []string{
            "https://golang.org/",
            "https://golang.org/cmd/",
            "https://golang.org/pkg/fmt/",
            "https://golang.org/pkg/os/",
        },
    },
    "https://golang.org/pkg/fmt/": &fakeResult{
        "Package fmt",
        []string{
            "https://golang.org/",
            "https://golang.org/pkg/",
        },
    },
    "https://golang.org/pkg/os/": &fakeResult{
        "Package os",
        []string{
            "https://golang.org/",
            "https://golang.org/pkg/",
        },
    },
}

输出:

found: https://golang.org/ "The Go Programming Language"
not found: https://golang.org/cmd/
found: https://golang.org/pkg/ "Packages"
found: https://golang.org/pkg/os/ "Package os"
fatal error: all goroutines are asleep - deadlock!

我想知道为什么会发生死锁？是不是我用错了 channel ？

注意到我忘记释放 if isCrawled {} 分支中的互斥量， 所以我像这样编辑了我的代码:

...
    if isCrawled {
        mux.Unlock() // added this line
        return
    }
...

但是死锁依然存在，输出结果不同:

found: https://golang.org/ "The Go Programming Language"
not found: https://golang.org/cmd/
found: https://golang.org/pkg/ "Packages"
found: https://golang.org/pkg/os/ "Package os"
found: https://golang.org/pkg/fmt/ "Package fmt"
fatal error: all goroutines are asleep - deadlock!

最佳答案

主要问题是您在 if isCrawled {} 中返回之前忘记释放互斥体。分支。

此外，如果您确实需要同步 goroutine，我建议使用同步 API。 channel 更适合用于通信和共享数据。

这是 sync.WaitGroup 的解决方案: https://play.golang.org/p/slrnmr3sPrs

这里是只有 channel 的解决方案:https://play.golang.org/p/FbPXxPSXvFL

问题是你第一次调用 CrawlURL()你不是从你作为论点传递的 channel 阅读。因此，一旦该函数尝试通过 defer func() { quit <- true }() 向其中发送内容。 ，它永远阻塞并且永远不会返回。

package main

import (
    "fmt"
    "sync"
)

type Fetcher interface {
    // Fetch returns the body of URL and
    // a slice of URLs found on that page.
    Fetch(url string) (body string, urls []string, err error)
}

var crawledURLs = make(map[string]bool)
var mux sync.Mutex

func CrawlURL(url string, depth int, fetcher Fetcher, quit chan bool) {
    //For very first function instance, this would block forever if 
    //nobody is receiving from the other end of this channel
    defer func() { quit <- true }()

    if depth <= 0 {
        return
    }

    mux.Lock()
    _, isCrawled := crawledURLs[url]
    if isCrawled {
        mux.Unlock()
        return
    }
    crawledURLs[url] = true
    mux.Unlock()

    body, urls, err := fetcher.Fetch(url)
    if err != nil {
        fmt.Println(err)
        return
    }
    fmt.Printf("found: %s %q\n", url, body)
    quitThis := make(chan bool)
    for _, u := range urls {
        go CrawlURL(u, depth-1, fetcher, quitThis)
    }
    for range urls {
        <-quitThis
    }
    return
}

// Crawl uses fetcher to recursively crawl
// pages starting with url, to a maximum of depth.
func Crawl(url string, depth int, fetcher Fetcher) {
    lastQuit := make(chan bool)
    go CrawlURL(url, depth, fetcher, lastQuit)
    //You need to receive from this channel in order to
    //unblock the called function
    <-lastQuit
    return
}

func main() {
    Crawl("https://golang.org/", 10, fetcher)
}

// fakeFetcher is Fetcher that returns canned results.
type fakeFetcher map[string]*fakeResult

type fakeResult struct {
    body string
    urls []string
}

func (f fakeFetcher) Fetch(url string) (string, []string, error) {
    if res, ok := f[url]; ok {
        return res.body, res.urls, nil
    }
    return "", nil, fmt.Errorf("not found: %s", url)
}

// fetcher is a populated fakeFetcher.
var fetcher = fakeFetcher{
    "https://golang.org/": &fakeResult{
        "The Go Programming Language",
        []string{
            "https://golang.org/pkg/",
            "https://golang.org/cmd/",
        },
    },
    "https://golang.org/pkg/": &fakeResult{
        "Packages",
        []string{
            "https://golang.org/",
            "https://golang.org/cmd/",
            "https://golang.org/pkg/fmt/",
            "https://golang.org/pkg/os/",
        },
    },
    "https://golang.org/pkg/fmt/": &fakeResult{
        "Package fmt",
        []string{
            "https://golang.org/",
            "https://golang.org/pkg/",
        },
    },
    "https://golang.org/pkg/os/": &fakeResult{
        "Package os",
        []string{
            "https://golang.org/",
            "https://golang.org/pkg/",
        },
    },
}

关于去旅行练习 : Web Crawler - all goroutines are asleep - deadlock，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/56882761/