如何接受客户端请求并调用处理函数
上篇概括了redis的启动流程,这篇重点介绍redis如何接受客户端请求并调用处理函数来执行命令。 在上一篇里,说到了在initServer()这个函数里边,会调用anetTcpServer和anetUnixServer 这两个函数创建对tcp端口和unix域套接字的监听,那么这里首先重点分析下
上篇概括了redis的启动流程,这篇重点介绍redis如何接受客户端请求并调用处理函数来执行命令。
在上一篇里,美国服务器,说到了在initServer()这个函数里边,会调用anetTcpServer和anetUnixServer 这两个函数创建对tcp端口和unix域套接字的监听,那么这里首先重点分析下这两个函数的具体实现。
int anetTcpServer(char *err, int port, char *bindaddr) { int s; struct sockaddr_in sa; if ((s = anetCreateSocket(err,AF_INET)) == ANET_ERR) return ANET_ERR; memset(&sa,0,sizeof(sa)); sa.sin_family = AF_INET; sa.sin_port = htons(port); sa.sin_addr.s_addr = htonl(INADDR_ANY); if (bindaddr && inet_aton(bindaddr, &sa.sin_addr) == 0) { anetSetError(err, ); close(s); return ANET_ERR; } if (anetListen(err,s,(struct sockaddr*)&sa,sizeof(sa)) == ANET_ERR) return ANET_ERR; return s; }
从代码中我们可以看出,首先调用anetCreateSocket()创建一个套接字并复值给s, 然后对sa 这个sockaddr_in类型的结构体进行初始化, 设置要监听的端口,地址,和地址族, 再调用anetListen() 函数绑定地址并监听端口,这些工作完成后 anetCreateSocket函数返回,网站空间,并将创建的套接字复制给server.ipfd。注意在anetUnixServer()这个函数中完成的工 作类似于anetCreateSocket,只不过是绑定的unix socket。
接下来, 在initServer函数里, 调用了这个函数:aeCreateFileEvent, 这里重点分析第一个,第二个类似。
if (server.ipfd > 0 && aeCreateFileEvent(server.el,server.ipfd,AE_READABLE, acceptTcpHandler,NULL) == AE_ERR) oom(); if (server.sofd > 0 && aeCreateFileEvent(server.el,server.sofd,AE_READABLE, acceptUnixHandler,NULL) == AE_ERR) oom();
首先,从eventLoop的event这个aeFileEvent数组里,取出当前fd对应的acFileEvent,主要是为了在下边给它设置对应事件的处理函数;即根据传入的mask来判断是哪一类事件。
int aeCreateFileEvent(aeEventLoop *eventLoop, int fd, int mask, aeFileProc *proc, void *clientData) { if (fd >= AE_SETSIZE) return AE_ERR; aeFileEvent *fe = &eventLoop->events[fd]; if (aeApiAddEvent(eventLoop, fd, mask) == -1) return AE_ERR; fe->mask |= mask; if (mask & AE_READABLE) fe->rfileProc = proc; if (mask & AE_WRITABLE) fe->wfileProc = proc; fe->clientData = clientData; if (fd > eventLoop->maxfd) eventLoop->maxfd = fd; return AE_OK; }
然后调用 acApiEvent这个事件注册函数:
static int aeApiAddEvent(aeEventLoop *eventLoop, int fd, int mask) { aeApiState *state = eventLoop->apidata; struct epoll_event ee; /* If the fd was already monitored for some event, we need a MOD * operation. Otherwise we need an ADD operation. */ int op = eventLoop->events[fd].mask == AE_NONE ? EPOLL_CTL_ADD : EPOLL_CTL_MOD; ee.events = 0; mask |= eventLoop->events[fd].mask; (mask & AE_READABLE) ee.events |= EPOLLIN; if (mask & AE_WRITABLE) ee.events |= EPOLLOUT; ee.data.u64 = ee.data.fd = fd; if (epoll_ctl(state->epfd,op,fd,&ee) == -1) return -1; return 0; }
可以看到, 这个函数中,主要是调用epoll_ctl(state->epfd,op,fd,&ee)将当前fd设置到及其所关心的事件注册到epoll_create 返回的epoll的句柄里。由于我们这里注册的是AE_READABLE事件,所以当这个fd(即redis监听端口的套接字)有数据可读时(这里我理解的是客户端连接到达),网站空间,就会触发相应的事件处理函数,这里的事件处理函数便是acceptTcpHandler。
下面我们看看acceptTcpHandler这个函数:
这个函数里边首先调用 anetTcpAccept获取客户端与redis连接的socket fd(实际调用 ::accept(s,sa,len)函数返回的fd), 然后调用 acceptCommonHandler()函数,这个函数中调用 createClient()创建 redisClient *c实例, 如果当前redis服务器的连接总数没有超过最大值,则将全局变量server中记录连接数的stat_numconnections加1;如果超过了, 则想客户端输出错误信息,并释放redisClient实例,函数返回。
void acceptTcpHandler(aeEventLoop *el, int fd, void *privdata, int mask) { int cport, cfd; char cip[128]; REDIS_NOTUSED(el); REDIS_NOTUSED(mask); REDIS_NOTUSED(privdata); cfd = anetTcpAccept(server.neterr, fd, cip, &cport); if (cfd == AE_ERR) { redisLog(REDIS_WARNING,, server.neterr); return; } redisLog(REDIS_VERBOSE,, cip, cport); acceptCommonHandler(cfd); }
下面分析createClient这个函数:
redisClient *createClient(int fd) { redisClient *c = zmalloc(sizeof(redisClient)); c->bufpos = 0; anetNonBlock(NULL,fd); anetTcpNoDelay(NULL,fd); if (aeCreateFileEvent(server.el,fd,AE_READABLE, readQueryFromClient, c) == AE_ERR) { close(fd); zfree(c); return NULL; } selectDb(c,0); c->fd = fd; c->querybuf = sdsempty(); c->reqtype = 0; c->argc = 0; c->argv = NULL; c->cmd = c->lastcmd = NULL; c->multibulklen = 0; c->bulklen = -1; c->sentlen = 0; c->flags = 0; c->lastinteraction = time(NULL); c->authenticated = 0; c->replstate = REDIS_REPL_NONE; c->slave_listening_port = 0; c->reply = listCreate(); c->reply_bytes = 0; listSetFreeMethod(c->reply,decrRefCount); listSetDupMethod(c->reply,dupClientReplyValue); c->bpop.keys = NULL; c->bpop.count = 0; c->bpop.timeout = 0; c->bpop.target = NULL; c->io_keys = listCreate(); c->watched_keys = listCreate(); listSetFreeMethod(c->io_keys,decrRefCount); c->pubsub_channels = dictCreate(&setDictType,NULL); c->pubsub_patterns = listCreate(); listSetFreeMethod(c->pubsub_patterns,decrRefCount); listSetMatchMethod(c->pubsub_patterns,listMatchObjects); listAddNodeTail(server.clients,c); initClientMultiState(c); return c; }

Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

Video Face Swap
Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Article

Hot Tools

Notepad++7.3.1
Easy-to-use and free code editor

SublimeText3 Chinese version
Chinese version, very easy to use

Zend Studio 13.0.1
Powerful PHP integrated development environment

Dreamweaver CS6
Visual web development tools

SublimeText3 Mac version
God-level code editing software (SublimeText3)

Hot Topics

Go language provides two dynamic function creation technologies: closure and reflection. closures allow access to variables within the closure scope, and reflection can create new functions using the FuncOf function. These technologies are useful in customizing HTTP routers, implementing highly customizable systems, and building pluggable components.

In C++ function naming, it is crucial to consider parameter order to improve readability, reduce errors, and facilitate refactoring. Common parameter order conventions include: action-object, object-action, semantic meaning, and standard library compliance. The optimal order depends on the purpose of the function, parameter types, potential confusion, and language conventions.

The key to writing efficient and maintainable Java functions is: keep it simple. Use meaningful naming. Handle special situations. Use appropriate visibility.

1. The SUM function is used to sum the numbers in a column or a group of cells, for example: =SUM(A1:J10). 2. The AVERAGE function is used to calculate the average of the numbers in a column or a group of cells, for example: =AVERAGE(A1:A10). 3. COUNT function, used to count the number of numbers or text in a column or a group of cells, for example: =COUNT(A1:A10) 4. IF function, used to make logical judgments based on specified conditions and return the corresponding result.

The advantages of default parameters in C++ functions include simplifying calls, enhancing readability, and avoiding errors. The disadvantages are limited flexibility and naming restrictions. Advantages of variadic parameters include unlimited flexibility and dynamic binding. Disadvantages include greater complexity, implicit type conversions, and difficulty in debugging.

The benefits of functions returning reference types in C++ include: Performance improvements: Passing by reference avoids object copying, thus saving memory and time. Direct modification: The caller can directly modify the returned reference object without reassigning it. Code simplicity: Passing by reference simplifies the code and requires no additional assignment operations.

The difference between custom PHP functions and predefined functions is: Scope: Custom functions are limited to the scope of their definition, while predefined functions are accessible throughout the script. How to define: Custom functions are defined using the function keyword, while predefined functions are defined by the PHP kernel. Parameter passing: Custom functions receive parameters, while predefined functions may not require parameters. Extensibility: Custom functions can be created as needed, while predefined functions are built-in and cannot be modified.

Exception handling in C++ can be enhanced through custom exception classes that provide specific error messages, contextual information, and perform custom actions based on the error type. Define an exception class inherited from std::exception to provide specific error information. Use the throw keyword to throw a custom exception. Use dynamic_cast in a try-catch block to convert the caught exception to a custom exception type. In the actual case, the open_file function throws a FileNotFoundException exception. Catching and handling the exception can provide a more specific error message.
