Samples that did not compile or run as shown: - res.user_data.get<T>() inside a generic lambda needs the `template` keyword; use explicit parameter types (tour 09, cookbook s15). - listen() on a Unix domain socket fails with port 0 (tour 09, s22). - "*.dev.local" is not a NO_PROXY pattern (c16). - ssl_backend_error() holds a verify result, not an ERR_get_error() value, after a verification failure; decode each with the matching OpenSSL function (c18). - The content provider's `length` is everything that remains, so the sample read the whole file in one call (s05). Statements corrected: - Client keep-alive is off by default; c14 is rewritten around set_keep_alive(true). - Mounted files are looked up before GET handlers (tour 04, s04). - Params keep insertion order, and to_string(Error::Connection) reads "Could not establish connection" (tour 02). - A chunked provider ends with sink.done(), and post_routing_handler runs before the response is sent (tour 09). - Timeouts surface as Error::Read; Error::Timeout comes from the stream API (c17). The max timeout cuts off the wait for the response only (c13). The progress callback needs Content-Length (c11). - Encoding selection follows q-values, then Brotli, gzip, Zstd (s08), and the client compresses with the first of those it was built with (c15). - stop() cuts a provider-driven response short (s19); a rejected content_reader already gets 400 or 413 (s07); user_data values must be copyable (s12); Client accepts a client certificate too (t04); on_message() is the fallback for every unhandled event and 204/403/404 end reconnection (e04); the pong timeout takes two to three intervals and ends a waiting read() (w02). In the LLM app tutorial, an uncaught exception does not crash the server, so say what it does instead. Drop the server and client timeout settings whose stated purpose, covering inference and download time, they do not serve: those timeouts bound a single socket wait. Update the llama.cpp server layout in chapter 7.
7.0 KiB
title, order
| title | order |
|---|---|
| Basic Client | 2 |
cpp-httplib isn't just for servers -- it also comes with a full HTTP client. Let's use httplib::Client to send GET and POST requests.
Preparing a Test Server
To try out the client, you need a server that accepts requests. Save the following code, then compile and run it the same way you did in the previous chapter. We'll cover the server details in the next chapter.
#include "httplib.h"
#include <iostream>
int main() {
httplib::Server svr;
svr.Get("/hi", [](const auto &, auto &res) {
res.set_content("Hello!", "text/plain");
});
svr.Get("/search", [](const auto &req, auto &res) {
auto q = req.get_param_value("q");
res.set_content("Query: " + q, "text/plain");
});
svr.Post("/post", [](const auto &req, auto &res) {
res.set_content(req.body, "text/plain");
});
svr.Post("/submit", [](const auto &req, auto &res) {
std::string result;
for (auto &[key, val] : req.params) {
result += key + " = " + val + "\n";
}
res.set_content(result, "text/plain");
});
svr.Post("/upload", [](const auto &req, auto &res) {
auto f = req.form.get_file("file");
auto content = f.filename + " (" + std::to_string(f.content.size()) + " bytes)";
res.set_content(content, "text/plain");
});
svr.Get("/users/:id", [](const auto &req, auto &res) {
auto id = req.path_params.at("id");
res.set_content("User ID: " + id, "text/plain");
});
svr.Get(R"(/files/(\d+))", [](const auto &req, auto &res) {
auto id = req.matches[1];
res.set_content("File ID: " + std::string(id), "text/plain");
});
std::cout << "Listening on port 8080..." << std::endl;
svr.listen("0.0.0.0", 8080);
}
GET Request
Once the server is running, open a separate terminal and give it a try. Let's start with the simplest GET request.
#include "httplib.h"
#include <iostream>
int main() {
httplib::Client cli("http://localhost:8080");
auto res = cli.Get("/hi");
if (res) {
std::cout << res->status << std::endl; // 200
std::cout << res->body << std::endl; // Hello!
}
}
Pass the server address to the httplib::Client constructor, then call Get() to send a request. You can retrieve the status code and body from the returned res.
Here's the equivalent curl command.
curl http://localhost:8080/hi
# Hello!
Checking the Response
A response contains header information in addition to the status code and body.
auto res = cli.Get("/hi");
if (res) {
// Status code
std::cout << res->status << std::endl; // 200
// Body
std::cout << res->body << std::endl; // Hello!
// Headers
std::cout << res->get_header_value("Content-Type") << std::endl; // text/plain
}
res->body is a std::string, so if you want to parse a JSON response, you can pass it directly to a JSON library like nlohmann/json.
Query Parameters
To add query parameters to a GET request, you can either write them directly in the URL or use httplib::Params.
auto res = cli.Get("/search", httplib::Params{{"q", "cpp-httplib"}});
if (res) {
std::cout << res->body << std::endl; // Query: cpp-httplib
}
httplib::Params automatically URL-encodes special characters for you.
curl "http://localhost:8080/search?q=cpp-httplib"
# Query: cpp-httplib
Path Parameters
When values are embedded directly in the URL path, no special client API is needed. Just pass the path to Get() as-is.
auto res = cli.Get("/users/42");
if (res) {
std::cout << res->body << std::endl; // User ID: 42
}
curl http://localhost:8080/users/42
# User ID: 42
The test server also has a /files/(\d+) route that uses a regex to accept numeric IDs only.
auto res = cli.Get("/files/42");
if (res) {
std::cout << res->body << std::endl; // File ID: 42
}
curl http://localhost:8080/files/42
# File ID: 42
Pass a non-numeric ID like /files/abc and you'll get a 404. We'll cover how that works in the next chapter.
Request Headers
To add custom HTTP headers, pass an httplib::Headers object. This works with both Get() and Post().
auto res = cli.Get("/hi", httplib::Headers{
{"Authorization", "Bearer my-token"}
});
curl -H "Authorization: Bearer my-token" http://localhost:8080/hi
POST Request
Let's POST some text data. Pass the body as the second argument to Post() and the Content-Type as the third.
auto res = cli.Post("/post", "Hello, Server!", "text/plain");
if (res) {
std::cout << res->status << std::endl; // 200
std::cout << res->body << std::endl; // Hello, Server!
}
The test server's /post endpoint echoes the body back, so you get the same string you sent.
curl -X POST -H "Content-Type: text/plain" -d "Hello, Server!" http://localhost:8080/post
# Hello, Server!
Sending Form Data
You can send key-value pairs just like an HTML form. Use httplib::Params for this.
auto res = cli.Post("/submit", httplib::Params{
{"name", "Alice"},
{"age", "30"}
});
if (res) {
std::cout << res->body << std::endl;
// name = Alice
// age = 30
}
This sends the data in application/x-www-form-urlencoded format.
curl -X POST -d "name=Alice&age=30" http://localhost:8080/submit
POSTing a File
To upload a file, use httplib::UploadFormDataItems to send it as multipart form data.
auto res = cli.Post("/upload", httplib::UploadFormDataItems{
{"file", "Hello, File!", "hello.txt", "text/plain"}
});
if (res) {
std::cout << res->body << std::endl; // hello.txt (12 bytes)
}
Each element in UploadFormDataItems has four fields: {name, content, filename, content_type}.
curl -F "file=Hello, File!;filename=hello.txt;type=text/plain" http://localhost:8080/upload
Error Handling
Network communication can fail -- the server might not be reachable. Always check whether res is valid.
httplib::Client cli("http://localhost:9999"); // Non-existent port
auto res = cli.Get("/hi");
if (!res) {
// Connection error
std::cout << "Error: " << httplib::to_string(res.error()) << std::endl;
// Error: Could not establish connection
return 1;
}
// If we reach here, we have a response
if (res->status != 200) {
std::cout << "HTTP Error: " << res->status << std::endl;
return 1;
}
std::cout << res->body << std::endl;
There are two levels of errors.
- Connection error: The client couldn't reach the server.
resevaluates to false, and you can callres.error()to find out what went wrong. - HTTP error: The server returned an error status (404, 500, etc.).
resevaluates to true, but you need to checkres->status.
Next Steps
Now you know how to send requests from a client. Next, let's take a closer look at the server side. We'll dig into routing, path parameters, and more.
Next: Basic Server