cURL C++ Example

curl_example.cpp

#include <iostream>
#include <string>
#include <curl/curl.h>


static size_t WriteCallback(void *contents, size_t size, size_t nmemb, void *userp)
{
    ((std::string*)userp)->append((char*)contents, size * nmemb);
    return size * nmemb;
}

int main(void)
{
  CURL *curl;
  CURLcode res;
  std::string readBuffer;

  curl = curl_easy_init();
  if(curl) {
    curl_easy_setopt(curl, CURLOPT_URL, "http://www.google.com");
    curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, WriteCallback);
    curl_easy_setopt(curl, CURLOPT_WRITEDATA, &readBuffer);
    res = curl_easy_perform(curl);
    curl_easy_cleanup(curl);

    std::cout << readBuffer << std::endl;
  }
  return 0;
}

To run this example, you need to have the curl headers installed

#Install header files and library for cURL
sudo aptitude install libcurl4-openssl-dev
#Download sample cpp file
curl -sLO https://gist.github.com/alghanmi/c5d7b761b2c9ab199157/raw/curl_example.cpp
#Compile and run the code
g++ curl_example.cpp -o curl_example -lcurl
./curl_example

The only caveat is that when dealing with SSL connections, they need to use the OpenSSL libraries.

@alghanmi Works!!!!! your the best!!!!

Hello !
The curl example perfectly works ! What are the role of the "static size_t WriteCallback" function, the "if(curl)" and the "curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, WriteCallback)" ?
Thank you for your help.

@valhockey4

This callback function will be called when receive the http response from the server.
You need to pass a function to handle the response stored in contents. The CURLOPT_WRITEDATA is set the fourth param in the cb function. You can write the response to this buffer and access it at your user context.

Note that you can't pass a std::string as url parameter for curl_easy_setopt:

std::string url = "http://www.google.com";
curl_easy_setopt(curl, CURLOPT_URL, url); // this will fail without triggering any compilation error (in g++, at least)
curl_easy_setopt(curl, CURLOPT_URL, url.c_str()); // this will work

how can i use PUT or POST option in code? thank you!

when i tried url.c_str() I got segmantation fault error. passing readBuffer by reference is also working in my app.

url has to be converted using url.c_str()

Hi,
when i am giving wrong URL then it block with line
res = curl_easy_perform(curl);

is there any idea ,please suggest me.

Thanks in advance

again posting more

int sendMsg(string toSend){

  int status = -1;
  CURL *curl;
 CURLcode res;
 string response;
  string url = "http://129.2.2.2/test.php";
  string data = "name="+toSend;
  curl_global_init(CURL_GLOBAL_ALL);

  curl = curl_easy_init();

  if(curl){
      curl_easy_setopt(curl, CURLOPT_URL, url.c_str());
      curl_easy_setopt(curl, CURLOPT_POST, 1);
      curl_easy_setopt(curl, CURLOPT_POSTFIELDS, data.c_str());
      curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_callback);
     curl_easy_setopt(curl, CURLOPT_WRITEDATA, &response);

    res = curl_easy_perform(curl);  /------------------------  block here-------------------/

      if(res == 0)
	{

	}

}
}

it blocked at line res = curl_easy_perform(curl);

if i am giving correct url then working fine but blocked with wrong url. it should block, it should return error code.
please provide me suggestion.

res = curl_easy_perform(curl);
What is the use of "CURLcode res" variable? you have not used it further in above program? And what is the returning value to res variable?
Please help me..

This code did work right out of the box for me.
With so many projects there are always little details.
This is a great example.
The follow-up message that includes the compiler line cinches it. -lcurl was essential to the compilation.

How can i run this code in code block ?

Thank you so much for writing this as an example!

Hello,

For those folks who are looking for cURL + HTTPS - curl team has nice example on their project page:

• https://curl.haxx.se/libcurl/c/https.html

Also there are tons of useful code snippets for many use cases: https://curl.haxx.se/libcurl/c/example.html

Where would one find the curl.h file in curl's source code? Is it on GitHub by any chance?

Incredible code nevertheless!

Works perfectly

guys i dont know how to integrate my project can you halp me ?

Build started...
1>------ Build started: Project: cmk, Configuration: Release Win32 ------
1>cmk.cpp
1>cmk.obj : error LNK2001: unresolved external symbol _curl_easy_perform
1>cmk.obj : error LNK2001: unresolved external symbol _curl_easy_init
1>cmk.obj : error LNK2001: unresolved external symbol _curl_easy_strerror
1>cmk.obj : error LNK2001: unresolved external symbol _curl_easy_cleanup
1>cmk.obj : error LNK2001: unresolved external symbol _curl_easy_setopt
1>C:\Users\Mustafa_Owner\source\repos\cmk\Release\cmk.exe : fatal error LNK1120: 5 unresolved externals
1>Done building project "cmk.vcxproj" -- FAILED.
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========
========== Elapsed 00:01,275 ==========

main.cpp - > \n
#include
#include <curl/curl.h>

int main()
{
CURL* curl;
CURLcode res;

curl = curl_easy_init();
if (curl) {
    curl_easy_setopt(curl, CURLOPT_URL, "https://example.com");

    /* Forcing HTTP/3 will make the connection fail if the server is not
       accessible over QUIC + HTTP/3 on the given host and port.
       Consider using CURLOPT_ALTSVC instead! */
    curl_easy_setopt(curl, CURLOPT_HTTP_VERSION, (long)CURL_HTTP_VERSION_3);

    /* Perform the request, res will get the return code */
    res = curl_easy_perform(curl);
    /* Check for errors */
    if (res != CURLE_OK)
        fprintf(stderr, "curl_easy_perform() failed: %s\n",
            curl_easy_strerror(res));

    /* always cleanup */
    curl_easy_cleanup(curl);
}

}

@hasirciogli

1. It looks like you did not include the -lcurl library when you compiled, hence the "unresolved external symbol" errors.
2. https://example.com does not support --http3, yet; you could test this using any HTTP/3 verifier.
   Try using any of the HTTP/3 supported websites, some are mentioned here.
3. I compiled and executed your code using https://quic.tech but received a curl_easy_perform() failed: SSL connect error error.
   Most probably because I don't have the OpenSSL libraries installed, and this is expected.
   But the code itself is executing fine.

#include <curl/curl.h>

int main()
{
	CURL* curl;
	CURLcode res;
	curl = curl_easy_init();
	if (curl)
	{
		curl_easy_setopt(curl, CURLOPT_URL, "https://quic.tech");
		curl_easy_setopt(curl, CURLOPT_HTTP_VERSION, (long)CURL_HTTP_VERSION_3);
		res = curl_easy_perform(curl);
		if (res != CURLE_OK)
			fprintf(stderr, "curl_easy_perform() failed: %s\n", curl_easy_strerror(res));
		
		curl_easy_cleanup(curl);
	}
}

Hope that helps :)

I developed a client which is receiving data from server occasionally. But when I receive first data, then program exists. I tried while loop like windows message loop, but it gives me last data.

How can I stay in program so that I can receive more messages from server.

@alghanmi Hi!

First off, thank you for sharing your code example! It was incredibly helpful for getting me started quickly. However, I wanted to point out a couple of discrepancies that I noticed:

1. Missing global initialization:
   The libcurl documentation recommends using curl_global_init() at the beginning of the program to set up the necessary environment for curl operations. See "Global preparation" section:
   - https://curl.se/libcurl/c/libcurl-tutorial.html
2. Callback prototype mismatch:
   The prototype for the callback function seems to be inconsistent with the one documented on the official libcurl site. The prototype for this callback function is described here:
   - https://curl.se/libcurl/c/CURLOPT_WRITEFUNCTION.html

size_t write_callback(char *ptr, size_t size, size_t nmemb, void *userdata);

On a related note, I've created a more detailed setup documentation and added a simple example as well as a slightly more complex one that includes POST requests, which some may find useful:
- https://github.com/voidlib/cpp-curl-example

Thanks again for your contribution—it was a great starting point for me!

Best regards,
@voiduin

this was the best explanation