深入理解Tomcat系列之七 详解URL请求

分享一下我老师大神的人工智能教程!零基础,通俗易懂!http://blog.csdn.net/jiangjunshow

也欢迎大家转载本篇文章。分享知识,造福人民,实现我们中华民族伟大复兴!

                       

前言

这里分析一个实际的请求是如何在Tomcat中被处理的,以及最后是怎么样找到要处理的Servlet的?当我们在浏览器中输入http://hostname:port/contextPath/servletPath,前面的hostname与port用于建立tcp连接,由于Http也是基于Tcp协议的,所以这里涉及TCP连接的三次握手。后面的contextPath与servletPath则是与服务器进行请求的信息,contextPath指明了与服务器中哪个Context容器进行交互,服务器会根据这个URL与对应的Context容器建立连接,那么这个过程是如何实现的呢?

在Tomcat7(本文也是基于Tomcat7)中主要通过一个映射来完成的,这个映射的工作交给org.apache.tomcat.util.http.mapper.Mapper类来完成的,这个类保存了Container容器所有子容器的信息,在请求从Connector交给Container容器之前,Mapper会根据hostname和port将host容器与context容器设置到Request的mappingData属性中,这样在Connector的请求进入Container容器之前就知道了交给哪个容器了。
这段代码如下:

代码清单5-4:

// Virtual host mapping        if (mappingData.host == null) {            Host[] hosts = this.hosts;            int pos = findIgnoreCase(hosts, host);            if ((pos != -1) && (host.equalsIgnoreCase(hosts[pos].name))) {                mappingData.host = hosts[pos].object;                contexts = hosts[pos].contextList.contexts;                nesting = hosts[pos].contextList.nesting;            } else {                if (defaultHostName == null) {                    return;                }                pos = find(hosts, defaultHostName);                if ((pos != -1) && (defaultHostName.equals(hosts[pos].name))) {                    mappingData.host = hosts[pos].object;                    contexts = hosts[pos].contextList.contexts;                    nesting = hosts[pos].contextList.nesting;                } else {                    return;                }            }        }        // Context mapping        if (mappingData.context == null) {            int pos = find(contexts, uri);            if (pos == -1) {                return;            }            int lastSlash = -1;            int uriEnd = uri.getEnd();            int length = -1;            boolean found = false;            while (pos >= 0) {                if (uri.startsWith(contexts[pos].name)) {                    length = contexts[pos].name.length();                    if (uri.getLength() == length) {                        found = true;                        break;                    } else if (uri.startsWithIgnoreCase("/", length)) {                        found = true;                        break;                    }                }                if (lastSlash == -1) {                    lastSlash = nthSlash(uri, nesting + 1);                } else {                    lastSlash = lastSlash(uri);                }                uri.setEnd(lastSlash);                pos = find(contexts, uri);            }            uri.setEnd(uriEnd);            if (!found) {                if (contexts[0].name.equals("")) {                    context = contexts[0];                }            } else {                context = contexts[pos];            }            if (context != null) {                mappingData.contextPath.setString(context.name);            }        }        if (context != null) {            ContextVersion[] contextVersions = context.versions;            int versionCount = contextVersions.length;            if (versionCount > 1) {                Object[] contextObjects = new Object[contextVersions.length];                for (int i = 0; i < contextObjects.length; i++) {                    contextObjects[i] = contextVersions[i].object;                }                mappingData.contexts = contextObjects;            }            if (version == null) {                // Return the latest version                contextVersion = contextVersions[versionCount - 1];            } else {                int pos = find(contextVersions, version);                if (pos < 0 || !contextVersions[pos].name.equals(version)) {                    // Return the latest version                    contextVersion = contextVersions[versionCount - 1];                } else {                    contextVersion = contextVersions[pos];                }            }            mappingData.context = contextVersion.object;            mappingData.contextSlashCount = contextVersion.slashCount;        }        // Wrapper mapping        if ((contextVersion != null) && (mappingData.wrapper == null)) {            internalMapWrapper(contextVersion, uri, mappingData);        }
   
   
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92

下面启动服务器,在浏览器中输入http://localhost:8080/examples/jsp/jsp2/el/composite.jsp,断点调试可以mappingData.host属性为localhost,mappingData.contextPath.setString(context.name)中context.name为examples,mappingData.wrapperPath为/jsp/jsp2/el/composite.jsp,这验证了mappingData属性的有效性,那么mappingData属性是如何设置到Request对象的属性中的呢?

通过org.apache.catalina.connector.Request的源码可以知道,其是通过setContextPath方法与setHost方法设置进去的,其源码如下:

代码清单5-5:

    public void setHost(Host host) {        mappingData.host = host;    }    public void setContextPath(String path) {        if (path == null) {            mappingData.contextPath.setString("");        } else {            mappingData.contextPath.setString(path);        }    }
   
   
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11

由于请求是从Connector传过来的,而CoyoteAdapter是Connector中处理请求的最后一个类,那么设置这两个属性的代码肯定在CoyoteAdapter类中,果不其然:

代码清单5-6:

 // This will map the the latest version by default    connector.getMapper().map(serverName, decodedURI, version,                              request.getMappingData());    request.setContext((Context) request.getMappingData().context);    request.setWrapper((Wrapper) request.getMappingData().wrapper);    //Mapper的map方法    public void map(MessageBytes host, MessageBytes uri, String version,                    MappingData mappingData)        throws Exception {        if (host.isNull()) {            host.getCharChunk().append(defaultHostName);        }        host.toChars();        uri.toChars();        internalMap(host.getCharChunk(), uri.getCharChunk(), version,                mappingData);    }
   
   
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18

intenalMap方法执行就是代码清单5-4的内容,这样就把从Connector传入请求,并设置Request对象的mappingData属性的整个流程就打通了。还有一个疑问是为什么Mapper类中可以拥有Container所有子容器的信息呢?答案需要回到Tomcat启动过程图的第21步的startIntenal方法了:

代码清单5-7:

    public void startInternal() throws LifecycleException {        setState(LifecycleState.STARTING);        // Find any components that have already been initialized since the        // MBean listener won't be notified as those components will have        // already registered their MBeans        findDefaultHost();        Engine engine = (Engine) connector.getService().getContainer();        addListeners(engine);        Container[] conHosts = engine.findChildren();        for (Container conHost : conHosts) {            Host host = (Host) conHost;            if (!LifecycleState.NEW.equals(host.getState())) {                // Registering the host will register the context and wrappers                registerHost(host);            }        }    }
   
   
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17

这段代码就是将MapperListener作为一个监听者加到整个Container容器的每一个子容器中,这样任何一个子容器发生变化,MapperListener都将被通知,响应的mappingData属性也会改变。最后可以总结访问请求地址为http://localhost:8080/examples/composite.jsp的处理过程:

  1. 在端口8080启动Server,并通知Service完成启动,Service通知Connector完成初始化和启动的过程
  2. Connector首先收到这个请求,会调用ProtocolHandler完成http协议的解析,然后交给SocketProcessor处理,解析请求头,再交给CoyoteAdapter解析请求行和请求体,并把解析信息封装到Request和Response对象中
  3. 把请求(此时应该是Request对象,这里的Request对象已经封装了Http请求的信息)交给Container容器
  4. Container容器交给其子容器——Engine容器,并等待Engine容器的处理结果
  5. Engine容器匹配其所有的虚拟主机,这里匹配到Host
  6. 请求被移交给hostname为localhost的Host容器,host匹配其所有子容器Context,这里找到contextPath为/examples的Context容器。如果匹配不到就把该请求交给路径名为”“的Context去处理
  7. 请求再次被移交给Context容器,Context继续匹配其子容器Wrapper,由Wrapper容器加载composite.jsp对应的servlet,这里编译的servlet是basic_002dcomparisons_jsp.class文件
  8. Context容器根据后缀匹配原则*.jsp找到composite.jsp编译的java类的class文件
  9. Connector构建一个org.apache.catalina.connector.Request以及org.apache.catalina.connector.Response对象,使用反射调用Servelt的service方法
  10. Context容器把封装了响应消息的Response对象返回给Host容器
  11. Host容器把Response返回给Engine容器
  12. Engine容器返回给Connector
  13. Connetor容器把Response返回给浏览器
  14. 浏览器解析Response报文
  15. 显示资源内容

根据前面的内容,其中的映射关系是由MapperListener类完成的。

           

给我老师的人工智能教程打call!http://blog.csdn.net/jiangjunshow

这里写图片描述

猜你喜欢

转载自blog.csdn.net/qq_43667968/article/details/84104120