golang中使用原生的http包编写一个web服务

栏目: Go · 发布时间: 5年前

内容简介:在golang中实现一个简单的web服务很简单,代码如下:首先我们是通过ListenAndServe来监听本地端口的,之后ListenAndServe将收到的新建一个Response连同收到的Request作为参数调用ServeMux结构体的ServeHTTP(省略了中间过程).ServeHTTP将Request作为参数调用

在golang中实现一个简单的web服务很简单,代码如下:

package main
import (
    "net/http"
    "fmt"
)
func main()  {
    http.HandleFunc("/", hello)

    http.ListenAndServe(":9090", nil)
}
func hello(w http.ResponseWriter, r *http.Request)  {
    fmt.Fprintf(w, "hello world")
}

首先我们是通过ListenAndServe来监听本地端口的,之后ListenAndServe将收到的新建一个Response连同收到的Request

作为参数调用ServeMux结构体的ServeHTTP(省略了中间过程).ServeHTTP将Request作为参数调用

Handler函数,Handler的返回值为一个Handler类型的接口,ServeHTTP会调用接口实现的ServeHTTP处理Response.

如果Request.URL.Path中有不合法的内容,则调用cleanPath清理,随后将Request.Host以及清理后的

内容传入handler函数,随后返回一个RedirectHandler以及handler所返回的路径。如果Request.URL.Path合法,那么

直接调用handler,返回值与handler返回值相同。

handler中通过判断ServeMux.hosts来决定是否实现pattern = r.Host + r.URL.Path.之后将pattern作为参数调用match,并将

match的返回值返回.

match的判别方式比较”有趣”,它虽然没实现为树形结构(只是用了映射),但是搜索的方法就是树形,因为URL路径就是个树形.它按照树的根节点

与子节点的关系进行判断,譬如路径”/home/select/usercourse”,match在匹配的时候会首先匹配到”/”(假如我们注册了),其次是”/home”,

之后逐层匹配下来,假如我们没注册过”/home/select/usercourse”,但是注册了”/home/select/”,那么match就会匹配到这一层.然后返回

“/home/select/”的Handler以及url(pattern).match函数的匹配规则实现在pathMatch

ServerMux结构体

type ServeMux struct {
        mu    sync.RWMutex
        m     map[string]muxEntry
        hosts bool // whether any patterns contain hostnames
    }

type muxEntry struct {
        explicit bool
        h        Handler
        pattern  string
    }

NewServeMux()

DefaultServeMux变量就是直接调用这个函数,那么这个函数只make了一个变量m,也就是为map分配了空间,在golang的语法中,结构体里未申明的变量也是存在的(即分配了内存)

pathMath()

这个函数是ServeMux用来匹配路劲的主要函数,所以看一下策略还是很重要的,函数中pattern是我们注册的路劲,path是用户请求的路径

func pathMatch(pattern, path string) bool {
    if len(pattern) == 0 {
        // should not happen
        return false
    }
    n := len(pattern)
    if pattern[n-1] != '/' {
        return pattern == path
    }
    return len(path) >= n && path[0:n] == pattern
}

如果我们挂在的路径不是以 / 结尾的,那么就直接判断两个参数是否相同,如果是以 / 结尾的,只要path的路径包含pattern那么久被判定是匹配,也就是说,如若我注册了 /home/select/ ,那么 /home/select/hello 也会被定位到 /home/select/ 上面,挂在的HanderFunc上面,这样做相当于为路径设置了一个index,不符合规则的URL都会被Redirct到这个index上。

* ServeMux.Handler()

注释写的很清楚,这个函数就是处理URL,然后调用*ServeMux.handler().首先调用cleanPath清理请求URL中的不合法内容。如果存在不合法内容,

则将清理过的URL交由*ServeMux.handler()处理并获得匹配到的pattern,然后修改url.Path的内容并调用RedirectHandler.

如果内容合法,则直接调用*ServeMux.handler()并返回结果

  • ServeMux.handler()

    调用ServeMux.match()(封装了pathMatch函数)来获得匹配到的Handler以及对应pattern,如果ServeMux.hosts==true,那么

    传入的参数为host + path,如果找不到的话,调用NotFoundHandler函数,并将其结果返回.

  • ServeMux.Handle()

    Handle函数是用来注册路径与处理过程的.如果该路径已经存在了一个用户注册的Handler则会panic(意思就是说不支持覆盖).判别了合法参数以后就将

    pattern作为key,新建一个muxEntry类型变量作为value加入到map中。

if pattern[0] != ‘/’ {

mux.hosts = true

}

这是这个函数中比较有意思的一个部分,通过这里我们可以看到如果注册路径的时候并不是以’/’开头的,那么ServeMux就会开启hosts,然后会在

请求到达的时候将URL.Host和URL.Path连接在一起放入match中寻找,具体信息请看这里

接下来是关于路径的处理,也就是关于”/home”与”/home/”的区别.我们先来看看作者怎么说

// Helpful behavior:
// If pattern is /tree/, insert an implicit permanent redirect for /tree.
// It can be overridden by an explicit registration.

如果路径的末尾是以’/’结尾并且该路径去掉末尾的’/’以后并没有被注册.那么将会去掉’/’并且为其绑定一个Redirect到现在的路径.

我自己写起来都觉得绕,举个例子就清楚了.

我注册了一个路径”/home/”,但是没有注册”/home”,那么如果用户访问了”/home”会发生什么呢?是的,会被Redirect到”/home/”.

需要注意的是,这里的muxEntry中的explicit没有填,也就是说是false,那么即是可以覆盖的.

  • ServeMux.ServeHTTP()

    ServeHTTP会检测非法的URI(* )

    如果通过检测就会调用自身的Handler()来返回注册的Handler,随后调用Handler的ServeHTTP方法

在http.ListenAndServe中的第二个参数传递处理hander可以自定义,这样就可以自己实现一个路由,在处理函数中必须实现ServeHTTP函数,这个函数在handler中直接执行。例子如下:

package routes

import (
    "encoding/json"
    "encoding/xml"
    "io/ioutil"
    "net/http"
    "net/url"
    "path/filepath"
    "regexp"
    "strconv"
    "strings"
)

const (
    CONNECT = "CONNECT"
    DELETE  = "DELETE"
    GET     = "GET"
    HEAD    = "HEAD"
    OPTIONS = "OPTIONS"
    PATCH   = "PATCH"
    POST    = "POST"
    PUT     = "PUT"
    TRACE   = "TRACE"
)

//commonly used mime-types
const (
    applicationJson = "application/json"
    applicationXml  = "application/xml"
    textXml         = "text/xml"
)

type route struct {
    method  string
    regex   *regexp.Regexp
    params  map[int]string
    handler http.HandlerFunc
}

type RouteMux struct {
    routes  []*route
    filters []http.HandlerFunc
}

func New() *RouteMux {
    return &RouteMux{}
}

// Get adds a new Route for GET requests.
func (m *RouteMux) Get(pattern string, handler http.HandlerFunc) {
    m.AddRoute(GET, pattern, handler)
}

// Put adds a new Route for PUT requests.
func (m *RouteMux) Put(pattern string, handler http.HandlerFunc) {
    m.AddRoute(PUT, pattern, handler)
}

// Del adds a new Route for DELETE requests.
func (m *RouteMux) Del(pattern string, handler http.HandlerFunc) {
    m.AddRoute(DELETE, pattern, handler)
}

// Patch adds a new Route for PATCH requests.
func (m *RouteMux) Patch(pattern string, handler http.HandlerFunc) {
    m.AddRoute(PATCH, pattern, handler)
}

// Post adds a new Route for POST requests.
func (m *RouteMux) Post(pattern string, handler http.HandlerFunc) {
    m.AddRoute(POST, pattern, handler)
}

// Adds a new Route for Static http requests. Serves
// static files from the specified directory
func (m *RouteMux) Static(pattern string, dir string) {
    //append a regex to the param to match everything
    // that comes after the prefix
    pattern = pattern + "(.+)"
    m.AddRoute(GET, pattern, func(w http.ResponseWriter, r *http.Request) {
        path := filepath.Clean(r.URL.Path)
        path = filepath.Join(dir, path)
        http.ServeFile(w, r, path)
    })
}

// Adds a new Route to the Handler
func (m *RouteMux) AddRoute(method string, pattern string, handler http.HandlerFunc) {

    //split the url into sections
    parts := strings.Split(pattern, "/")

    //find params that start with ":"
    //replace with regular expressions
    j := 0
    params := make(map[int]string)
    for i, part := range parts {
        if strings.HasPrefix(part, ":") {
            expr := "([^/]+)"
            //a user may choose to override the defult expression
            // similar to expressjs: ‘/user/:id([0-9]+)’
            if index := strings.Index(part, "("); index != -1 {
                expr = part[index:]
                part = part[:index]
            }
            params[j] = part
            parts[i] = expr
            j++
        }
    }

    //recreate the url pattern, with parameters replaced
    //by regular expressions. then compile the regex
    pattern = strings.Join(parts, "/")
    regex, regexErr := regexp.Compile(pattern)
    if regexErr != nil {
        //TODO add error handling here to avoid panic
        panic(regexErr)
        return
    }

    //now create the Route
    route := &route{}
    route.method = method
    route.regex = regex
    route.handler = handler
    route.params = params

    //and finally append to the list of Routes
    m.routes = append(m.routes, route)
}

// Filter adds the middleware filter.
func (m *RouteMux) Filter(filter http.HandlerFunc) {
    m.filters = append(m.filters, filter)
}

// FilterParam adds the middleware filter iff the REST URL parameter exists.
func (m *RouteMux) FilterParam(param string, filter http.HandlerFunc) {
    if !strings.HasPrefix(param,":") {
        param = ":"+param
    }

    m.Filter(func(w http.ResponseWriter, r *http.Request) {
        p := r.URL.Query().Get(param)
        if len(p) > 0 { filter(w, r) }
    })
}

// Required by http.Handler interface. This method is invoked by the
// http server and will handle all page routing
func (m *RouteMux) ServeHTTP(rw http.ResponseWriter, r *http.Request) {

    requestPath := r.URL.Path

    //wrap the response writer, in our custom interface
    w := &responseWriter{writer: rw}

    //find a matching Route
    for _, route := range m.routes {

        //if the methods don't match, skip this handler
        //i.e if request.Method is 'PUT' Route.Method must be 'PUT'
        if r.Method != route.method {
            continue
        }

        //check if Route pattern matches url
        if !route.regex.MatchString(requestPath) {
            continue
        }

        //get submatches (params)
        matches := route.regex.FindStringSubmatch(requestPath)

        //double check that the Route matches the URL pattern.
        if len(matches[0]) != len(requestPath) {
            continue
        }

        if len(route.params) > 0 {
            //add url parameters to the query param map
            values := r.URL.Query()
            for i, match := range matches[1:] {
                values.Add(route.params[i], match)
            }

            //reassemble query params and add to RawQuery
            r.URL.RawQuery = url.Values(values).Encode() + "&" + r.URL.RawQuery
            //r.URL.RawQuery = url.Values(values).Encode()
        }

        //execute middleware filters
        for _, filter := range m.filters {
            filter(w, r)
            if w.started {
                return
            }
        }

        //Invoke the request handler
        route.handler(w, r)
        break
    }

    //if no matches to url, throw a not found exception
    if w.started == false {
        http.NotFound(w, r)
    }
}

// -----------------------------------------------------------------------------
// Simple wrapper around a ResponseWriter

// responseWriter is a wrapper for the http.ResponseWriter
// to track if response was written to. It also allows us
// to automatically set certain headers, such as Content-Type,
// Access-Control-Allow-Origin, etc.
type responseWriter struct {
    writer  http.ResponseWriter
    started bool
    status  int
}

// Header returns the header map that will be sent by WriteHeader.
func (w *responseWriter) Header() http.Header {
    return w.writer.Header()
}

// Write writes the data to the connection as part of an HTTP reply,
// and sets `started` to true
func (w *responseWriter) Write(p []byte) (int, error) {
    w.started = true
    return w.writer.Write(p)
}

// WriteHeader sends an HTTP response header with status code,
// and sets `started` to true
func (w *responseWriter) WriteHeader(code int) {
    w.status = code
    w.started = true
    w.writer.WriteHeader(code)
}

// -----------------------------------------------------------------------------
// Below are helper functions to replace boilerplate
// code that serializes resources and writes to the
// http response.

// ServeJson replies to the request with a JSON
// representation of resource v.
func ServeJson(w http.ResponseWriter, v interface{}) {
    content, err := json.MarshalIndent(v, "", "  ")
    if err != nil {
        http.Error(w, err.Error(), http.StatusInternalServerError)
        return
    }
    w.Header().Set("Content-Length", strconv.Itoa(len(content)))
    w.Header().Set("Content-Type", applicationJson)
    w.Write(content)
}

// ReadJson will parses the JSON-encoded data in the http
// Request object and stores the result in the value
// pointed to by v.
func ReadJson(r *http.Request, v interface{}) error {
    body, err := ioutil.ReadAll(r.Body)
    r.Body.Close()
    if err != nil {
        return err
    }
    return json.Unmarshal(body, v)
}

// ServeXml replies to the request with an XML
// representation of resource v.
func ServeXml(w http.ResponseWriter, v interface{}) {
    content, err := xml.Marshal(v)
    if err != nil {
        http.Error(w, err.Error(), http.StatusInternalServerError)
        return
    }
    w.Header().Set("Content-Length", strconv.Itoa(len(content)))
    w.Header().Set("Content-Type", "text/xml; charset=utf-8")
    w.Write(content)
}

// ReadXml will parses the XML-encoded data in the http
// Request object and stores the result in the value
// pointed to by v.
func ReadXml(r *http.Request, v interface{}) error {
    body, err := ioutil.ReadAll(r.Body)
    r.Body.Close()
    if err != nil {
        return err
    }
    return xml.Unmarshal(body, v)
}

// ServeFormatted replies to the request with
// a formatted representation of resource v, in the
// format requested by the client specified in the
// Accept header.
func ServeFormatted(w http.ResponseWriter, r *http.Request, v interface{}) {
    accept := r.Header.Get("Accept")
    switch accept {
    case applicationJson:
        ServeJson(w, v)
    case applicationXml, textXml:
        ServeXml(w, v)
    default:
        ServeJson(w, v)
    }

    return
}

以上就是本文的全部内容,希望本文的内容对大家的学习或者工作能带来一定的帮助,也希望大家多多支持 码农网

查看所有标签

猜你喜欢:

本站部分资源来源于网络,本站转载出于传递更多信息之目的,版权归原作者或者来源机构所有,如转载稿涉及版权问题,请联系我们

机器学习及其应用2007

机器学习及其应用2007

周志华 编 / 清华大学 / 2007-10 / 37.00元

机器学习是人工智能的一个核心研究领域,也是近年来计算机科学中最活跃的研究分支之一。目前,机器学习技术不仅在计算机科学的众多领域中大显身手,还成为一些交叉学科的重要支撑技术。本书邀请相关领域的专家撰文,以综述的形式介绍机器学习中一些领域的研究进展。全书共分13章,内容涉及高维数据降维、特征选择、支持向量机、聚类、强化学习、半监督学习、复杂网络、异构数据、商空间、距离度量以及机器学习在自然语言处理中的......一起来看看 《机器学习及其应用2007》 这本书的介绍吧!

JSON 在线解析
JSON 在线解析

在线 JSON 格式化工具

图片转BASE64编码
图片转BASE64编码

在线图片转Base64编码工具

HEX HSV 转换工具
HEX HSV 转换工具

HEX HSV 互换工具