C/C++编程：url解析

深碍√TFBOYSˉ_ 2022-09-02 14:58 350阅读 0赞

# url解析 #

## 源码 ##

//
    // Created by oceanstar on 2021/8/6.
    //
    
    #ifndef OCEANSTAR_HTTP_HTTP_UTILS_H
    #define OCEANSTAR_HTTP_HTTP_UTILS_H
    
    #include <cstdio>
    #include "global.h"
    namespace oceanstar {
        
        class  http_utils{
        
        public:
            http_utils() {
        }
            ~http_utils() {
        }
            /**
            * 从完整的 url 中获得 WEB 服务器 IP 地址及端口号
            * @param url {const char*} HTTP url，非空
            * @param domain {char*} 存储域名
            * @param size {size_t} domain 内存大小
            * @param port {unsigned short*} 存储端口号大小
            * @return {bool} 是否成功获得
            */
            static bool get_addr(const char* url, char* domain, size_t size,
                                 unsigned short* port);
            /**
             * 从完整的 url 中获得 WEB 服务器地址，格式：domain:port
             * @param url {const char*} HTTP url，非空
             * @param addr {char*} 存储结果，存储格式：domain:port
             * @param size {size_t} out 缓冲区大小
             * @return {bool} 是否成功获得
             */
            static bool get_addr(const char* url, char* addr, size_t size);
    
        };
    };
    
    
    namespace oceanstar{
        
        using namespace std;
        class  http_url {
        
                public:
                http_url(void);
                ~http_url(void) {
        }
    
                bool parse(const char* url);
    
                public:
                /**
                 * 返回 URL 中的协议类型：http 或 https
                 * @return {const char*}
                 */
                const char* get_proto(void) const {
        
                    return proto_;
                }
    
                /**
                 * 返回 URL 中的域名字段
                 * @return {const char*} 返回空串则表示没有该字段
                 */
                const char* get_domain(void) const {
        
                    return domain_.c_str();
                }
    
                /**
                 * 返回根据 URL 提取的 HTTP 协议服务端端口号，内部缺省值为 80
                 * @return {unsigned short}
                 */
                unsigned short get_port(void) const {
        
                    return port_;
                }
    
                /**
                 * 返回根据 URL 提取的相对路径部分（不含 ? 后面的参数）
                 * @return {const char*}
                 */
                const char* get_url_path(void) const {
        
                    return url_path_.c_str();
                }
    
                /**
                 * 返回从 URL 中提取的参数字段
                 * @return {const char*}
                 */
                const char* get_url_params(void) const {
        
                    return url_params_.c_str();
                }
    
                /**
                 * 清理解析过程中的中间状态，以便重复使用该类对象解析下一个 URL
                 */
                void reset(void);
    
                private:
                char proto_[16];
                string domain_;
                unsigned short port_;
                string url_path_;
                string url_params_;
    
                bool parse_url_part(const char* url);
                const char* parse_domain(const char* url);
        };
    };
    
    #endif //OCEANSTAR_HTTP_HTTP_UTILS_H

//
    // Created by oceanstar on 2021/8/6.
    //
    
    #include <iostream>
    #include <string.h>
    #include "http_utils.h"
    
    
    namespace oceanstar{
        
    #define	HTTP_PREFIX	"http://"
    #define	HTTPS_PREFIX	"https://"
    
        bool http_utils::get_addr(const char* url, char* domain, size_t size,
                                  unsigned short* pport)
        {
        
            const char* ptr;
            unsigned short default_port;
    
            if (!strncasecmp(url, HTTP_PREFIX, sizeof(HTTP_PREFIX) - 1)) {
        
                ptr = url + sizeof(HTTP_PREFIX) - 1;
                default_port = 80;
            } else if (!strncasecmp(url, HTTPS_PREFIX, sizeof(HTTPS_PREFIX) - 1)) {
        
                ptr = url + sizeof(HTTPS_PREFIX) - 1;
                default_port = 443;
            } else {
        
                logger_error("invalid url: %s", url);
                return false;
            }
    
            if (*ptr == 0) {
        
                logger_error("invalid url: %s", url);
                return false;
            }
    
            char buf[256];
            strncpy(buf, ptr, sizeof(buf));
    
            char* slash = strchr(buf, '/');
            if (slash) {
        
                *slash = 0;
            }
    
            unsigned short port;
    
            char* col = strchr(buf, ':');
            if (col == NULL) {
        
                port = default_port;
            } else {
        
                *col++ = 0;
                port = (unsigned short) atoi(col);
                if (port == 0 || port == 65535) {
        
                    port = default_port;
                }
            }
    
            if (pport) {
        
                *pport = port;
            }
            strncpy(domain, buf, size);
            return true;
        }
    
    
        bool http_utils::get_addr(const char* url, char* addr, size_t size)
        {
        
            char  buf[256];
            unsigned short port;
    
            if (!get_addr(url, buf, sizeof(buf), &port)) {
        
                return false;
            }
            snprintf(addr, size, "%s:%d", buf, port);
            return true;
        }
    }
    
    namespace oceanstar{
        
        http_url::http_url(void) {
        
            strncpy(proto_, "http", sizeof(proto_));
            port_ = 80;
        }
        void http_url::reset(void) {
        
            strncpy(proto_, "http", sizeof(proto_));
            port_ = 80;
            domain_.clear();
            url_path_.clear();
            url_params_.clear();
        }
    
        const char* http_url::parse_domain(const char *url) {
        
            if (*url == '/') {
        
                logger_error("invalid url: %s", url);
                return NULL;
            }
            const char* ptr = strchr(url, '/');
            if (ptr == NULL) {
        
                domain_ = url;
                return NULL;
            }
    
            char buf[256];
            size_t size = ptr - url + 1;
            if (size > sizeof(buf)) {
         // xxx: sanity check
                size = sizeof(buf);
            }
            strncpy(buf, url, size);
    
            // fixme: Is it error if buf contains IPV6 Addr ---zsx
            char* col = strchr(buf, ':');
            if (col != NULL) {
        
                *col++ = 0;
                port_ = (unsigned short) atoi(col);
                if (port_ == 0 || port_ == 65535) {
        
                    port_ = 80;
                }
            }
            domain_ = buf;
            return ptr;
        }
    
    
        bool http_url::parse(const char *url) {
        
            const char* ptr;
    
            if (!strncasecmp(url, HTTP_PREFIX, sizeof(HTTP_PREFIX) - 1)) {
        
                ptr = url + sizeof(HTTP_PREFIX) - 1;
            } else if (!strncasecmp(url, HTTPS_PREFIX, sizeof(HTTPS_PREFIX) - 1)) {
        
                ptr = url + sizeof(HTTPS_PREFIX) - 1;
                port_ = 443;
                strncpy(proto_, "https", sizeof(proto_));
            } else if (*url == '/'){
        
                ptr = url;
            } else {
        
                logger_error("invalid url: %s", url);
                return false;
            }
    
            if (*ptr == 0) {
        
                logger_error("invalid url: %s", url);
                return false;
            }
    
            if (ptr == url) {
        
                // 说明是仅含相对路径的 url
                return parse_url_part(url);
            } else {
        
                // 说明包含有完整路径的 url，下面先提取域名字段，再提取相对 url
                ptr = parse_domain(ptr);
                if (ptr == NULL) {
        
                    url_path_ = "/";
                    return true;
                }
                return parse_url_part(ptr);
            }
        }
    
    #define SKIP_WHILE(cond, ptr) {
           while(*ptr && (cond)) ptr++; }
    
        bool http_url::parse_url_part(const char *url) {
        
            if (*url != '/') {
        
                logger_error("invalid url: %s", url);
                return false;
            }
    
            const char* ptr = url;
            SKIP_WHILE(*ptr == '/', ptr);
            if (*ptr == 0) {
        
                url_path_ = "/";
                return true;
            }
    
            const char* qm = strchr(ptr, '?');
            if (qm == NULL) {
        
                url_path_ = url;
                return true;
            }
            char buf[strlen(url)];
            strncpy(buf, url, qm - url);
            url_path_ = buf;
            ++qm;
            if (*qm != 0) {
        
                url_params_ = qm;
            }
            return true;
        }
    
    }

## 使用 ##

### 例子 ###

#include <cstring>
    #include "http_utils.h"
    
    
    using namespace oceanstar;
    int main(int argc, char *argv[])
    {
        
        unsigned short port ;
        char host[256];
        memset(host,  0, sizeof(host));
        std::string url = "http://www.gmail.com:443/path/test.cgi?name=value&name2=value2 ";
        if(http_utils::get_addr(url.c_str(), host, sizeof(host), &port)){
        
            logger_info("host: %s, port: %d", host, port);
        }else{
        
            logger_error("s(%d): url invalid ", __FUNCTION__ , __LINE__);
        }
    
    
        char addr[256];
        memset(addr,  0, sizeof(addr));
    
        if(http_utils::get_addr(url.c_str(), addr, sizeof(addr))){
        
            logger_info("addr: %s", addr);
        }else{
        
            logger_error("s(%d): url invalid ", __FUNCTION__ , __LINE__);
        }
    
        return (0);
    }

![在这里插入图片描述][a89da76ec44b4bb791c4df4811c5eb61.png]

### 例子 ###

#include <cstring>
    #include "http_utils.h"
    
    
    using namespace oceanstar;
    int main(int argc, char *argv[])
    {
        
        unsigned short port ;
        char host[256];
        memset(host,  0, sizeof(host));
        std::string url = "http://www.gmail.com:443/path/test.cgi?name=value&name2=value2 ";
    
    
        http_url httpUrl;
    
        if(!httpUrl.parse(url.c_str())){
        
            exit(0);
        }
    
        logger_info("proto: %s", httpUrl.get_proto());
        logger_info("domain: %s", httpUrl.get_domain());
        logger_info("port: %d", httpUrl.get_port());
        logger_info("url_path: %s", httpUrl.get_url_path());
        logger_info("url_params: %s", httpUrl.get_url_params());
    
        printf("\n\n");
    
        httpUrl.reset();
    
    
        url = "http://127.0.0.1:443/path/test.cgi";
        if(!httpUrl.parse(url.c_str())){
        
            exit(0);
        }
    
        logger_info("proto: %s", httpUrl.get_proto());
        logger_info("domain: %s", httpUrl.get_domain());
        logger_info("port: %d", httpUrl.get_port());
        logger_info("url_path: %s", httpUrl.get_url_path());
        logger_info("url_params: %s", httpUrl.get_url_params());
    
        return (0);
    }

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3poaXpoZW5nZ3Vhbg_size_16_color_FFFFFF_t_70]

# 通过127.0.0.1:80解析中获取host和port #

#include <cstring>
    #include "http_utils.h"
    
    
    using namespace oceanstar;
    
    // s必须是堆内存
    static char  *acl_split_at_right(char *s, int delimiter)
    {
        
        char   *cp;
    
        if ((cp = strrchr(s, delimiter)) != 0){
        
            *cp = 0;
            cp++;
        }
    
        return cp;
    }
    
    static const char *acl_host_port(char *buf, char **host, char *def_host,char **port)
    {
        
        char *cp = buf;
    
        /* host:port, host:, host, :port */
        if ((cp = acl_split_at_right(buf, ':')) != 0) {
        
            *host = *buf ? buf : def_host;
            *port = *cp ? cp : NULL;
        }
    
        if (*host == 0)
            return "missing host information";
    
    
        return NULL;
    }
    
    
    static int host_port(char *buf, char **host, char **port)
    {
        
        char *def_host = "";
        const char *ptr = acl_host_port(buf, host, def_host, port);
    
        if (ptr != NULL) {
        
            logger_error("%s(%d), %s: invalid addr %s, %s",
                         __FILE__, __LINE__, __FUNCTION__, buf, ptr);
            return -1;
        }
    
        if (*port != NULL && atoi(*port) < 0) {
        
            logger_error("%s(%d), %s: invalid port: %s, addr: %s",
                         __FILE__, __LINE__, __FUNCTION__,
                         *port ? *port : "null", buf);
            return -1;
        }
    
        if (*host && **host == 0)
            *host = 0;
        if (*host == NULL)
            *host = "0";
    
        return 0;
    }
    
    int main(int argc, char *argv[])
    {
        
    
        char *addr = strdup("127.0.0.0:80");
        char *host = NULL, *port = NULL;
        if (host_port(addr, &host, &port) < 0) {
        
            printf("formet error");
            free(addr);
            return NULL;
        }
    
        printf("host: %s, port: %s", host, port);
        free(addr);
    
      
        return (0);
    }

# 通过url获取host【版本1】 #

#include <iostream>
    /* data: "http://www.gmail.com:443/path/test.cgi?name=value&name2=value2 */
    static void __get_host_from_url(char *buf, size_t size, const char *url)
    {
        
        const char *ptr1, *ptr2;
        size_t n;
    
        buf[0] = 0;
        if (strncasecmp(url, "http://", sizeof("http://") - 1) == 0)
            ptr1 = url + sizeof("http://") - 1;
        else if (strncasecmp(url, "https://", sizeof("https://") - 1) == 0)
            ptr1 = url + sizeof("https://") - 1;
        else
            ptr1 = url;
    
        if (ptr1 == NULL || *ptr1 == 0 || *ptr1 == '/')
            return;
    
        ptr2 = strchr(ptr1, '/');
        if (ptr2)
            n = ptr2 - ptr1;
        else
            n = strlen(ptr1);
    
     //   n++;
    
        if (n > size)
            n = size;
    
        strncpy(buf, ptr1, (int) n);
    }
    
    int main() {
        
        char host[256];
        memset(host,  0, sizeof(host));
        __get_host_from_url(host, sizeof(host), "http://www.gmail.com:443/path/test.cgi?name=value&name2=value2 ");
        std::cout << host << std::endl;
    }

![在这里插入图片描述][5f37f2d92b734bd39d1acaab40c54518.png]

[a89da76ec44b4bb791c4df4811c5eb61.png]: /images/20220829/7e6f42750dc84ae8a8f2325ba9d2035a.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3poaXpoZW5nZ3Vhbg_size_16_color_FFFFFF_t_70]: /images/20220829/c9e3e9f866064f2aa73a4dbf9b626337.png
[5f37f2d92b734bd39d1acaab40c54518.png]: /images/20220829/460dd5d78cfe477c9991d3bbcea2838e.png