C++ 从 HTTP 响应中获取图片

Question

我正在尝试从网站下载图片。问题是即使我从 HTTP 响应正文中获取了所有内容，文件也无法打开。我一直在努力解决这个问题，但我找不到真正的问题。我确实注意到的一件事是，使用 chromium 下载的图片显示的字符与使用命令从我的代码下载的图片不同: $ 猫图片.png |少

#include <netdb.h>
#include <stdlib.h>
#include <string.h>
#include <sys/socket.h>
#include <unistd.h>
#include <fstream>
#include <iostream>

using std::cout;
using std::endl;

//with this function I remove HTTP header info, so I only get content.
char *removeHTTPHeader(char *buffer) {
    char *t = strstr(buffer, "\r\n\r\n");
    t = t + 4;
    return t;
}

void getPicture(const int &socketfd, const int &bSize) {
    std::ofstream file("picture.png",
            std::ofstream::binary | std::ofstream::out);

    char buffer[bSize];
    ssize_t bReceived;

    bReceived = recv(socketfd, buffer, bSize, 0);
    char *t = removeHTTPHeader(buffer);
    file.write(t, strlen(t));
    memset(buffer, 0, bSize);

    while ((bReceived = recv(socketfd, buffer, bSize, 0)) > 0) {
        file.write(buffer, bReceived);
        memset(buffer, 0, bSize);
    }

    file.close();
}

int main() {

    int status;
    addrinfo host_info;
    addrinfo *host_info_list;

    memset(&host_info, 0, sizeof(host_info));

    host_info.ai_family = AF_UNSPEC;
    host_info.ai_socktype = SOCK_STREAM;
    host_info.ai_protocol = 0;

    status = getaddrinfo("www.pngimg.com", "80", &host_info, &host_info_list);
    if (status != 0) {
        cout << "getaddrinfo error" << endl;
    }

    int socketfd;
    socketfd = socket(host_info_list->ai_family, host_info_list->ai_socktype,
            host_info_list->ai_protocol);

    addrinfo *rp;
    for (rp = host_info_list; rp != NULL; rp = rp->ai_next) {

        socketfd = socket(rp->ai_family, rp->ai_socktype, rp->ai_protocol);

        if (socketfd == -1) {
            cout << "socket error" << endl;
        }

        if (connect(socketfd, rp->ai_addr, rp->ai_addrlen) != -1) {
            break;
        }
        close(socketfd);
    }

    if (rp == NULL) {
        cout << "Could not connect!" << endl;
        exit(EXIT_FAILURE);
    }

    freeaddrinfo(host_info_list);

    const char *msg =
            "GET /upload/water_PNG3290.png HTTP/1.1\r\nhost: www.pngimg.com\r\nConnection: close\r\n\r\n";

    status = send(socketfd, msg, strlen(msg), 0);

    if (status == -1) {
        cout << "error sending" << endl;
        exit(EXIT_FAILURE);
    }

    getPicture(socketfd, 1024);
    close(socketfd);

    return 0;
}

这是使用 cat 命令的图片: terminal above is picture from my code, below is picture from chromium "save as"

Answer 1

问题是我不知道在 C 风格的字符串中你不能对二进制数据执行 strlen。这就是为什么我必须在函数 removeHTTPHeader 中添加计数器。以下是我更改的函数 getPicture 和 removeHTTPHeader。

char *removeHTTPHeader(char *buffer, int &bodySize) {
    char *t = strstr(buffer, "\r\n\r\n");
    t = t + 4;

    for (auto it = buffer; it != t; ++it) {
        ++bodySize;
    }

    return t;
}

void getPicture(const int &socketfd, const int &bSize) {
    std::ofstream file("picture.png",
            std::ofstream::binary | std::ofstream::out);

    char buffer[bSize];
    ssize_t bReceived;

    bReceived = recv(socketfd, buffer, bSize, 0);
    int bodySize = 0;

    char *t = removeHTTPHeader(buffer, bodySize);
    bodySize = bReceived - bodySize;

    file.write(t, bodySize);
    memset(buffer, 0, bSize);

    while ((bReceived = recv(socketfd, buffer, bSize, 0)) > 0) {
        file.write(buffer, bReceived);
        memset(buffer, 0, bSize);
    }

    file.close();
}