title: "What's Next"
Great job finishing the Tour! You now have a solid grasp of the cpp-httplib basics. But there's a lot more to explore. Here's a quick overview of features we didn't cover in the Tour, organized by category.
When you're working with LLM streaming responses or downloading large files, you don't want to load the entire response into memory. Use stream::Get() to process data chunk by chunk.
httplib::Client cli("http://localhost:11434");
auto result = httplib::stream::Get(cli, "/api/generate");
if (result) {
while (result.next()) {
std::cout.write(result.data(), result.size());
}
}
You can also pass a content_receiver callback to Get(). This approach works with Keep-Alive.
httplib::Client cli("http://localhost:8080");
cli.Get("/stream", [](const char *data, size_t len) {
std::cout.write(data, len);
return true;
});
On the server side, you have set_content_provider() and set_chunked_content_provider(). Use the former when you know the size, and the latter when you don't.
// With known size (sets Content-Length)
svr.Get("/file", [](const auto &, auto &res) {
auto size = get_file_size("large.bin");
res.set_content_provider(size, "application/octet-stream",
[](size_t offset, size_t length, httplib::DataSink &sink) {
// Send 'length' bytes starting from 'offset'
return true;
});
});
// Unknown size (Chunked Transfer Encoding)
svr.Get("/stream", [](const auto &, auto &res) {
res.set_chunked_content_provider("text/plain",
[](size_t offset, httplib::DataSink &sink) {
sink.write("chunk\n", 6);
return true; // Return false to finish
});
});
For uploading large files, make_file_provider() comes in handy. It streams the file instead of loading it all into memory.
httplib::Client cli("http://localhost:8080");
auto res = cli.Post("/upload", {}, {}, {
httplib::make_file_provider("file", "/path/to/large-file.zip")
});
We provide an SSE client as well. It supports automatic reconnection and resuming via Last-Event-ID.
httplib::Client cli("http://localhost:8080");
httplib::sse::SSEClient sse(cli, "/events");
sse.on_message([](const httplib::sse::SSEMessage &msg) {
std::cout << msg.event << ": " << msg.data << std::endl;
});
sse.start(); // Blocking, with auto-reconnection
You can also set separate handlers for each event type.
sse.on_event("update", [](const httplib::sse::SSEMessage &msg) {
// Only handles "update" events
});
The client has helpers for Basic auth, Bearer Token auth, and Digest auth.
httplib::Client cli("https://api.example.com");
cli.set_basic_auth("user", "password");
cli.set_bearer_token_auth("my-token");
We support compression and decompression with gzip, Brotli, and Zstandard. Define the corresponding macro when you compile.
| Method | Macro |
| -- | -- |
| gzip | CPPHTTPLIB_ZLIB_SUPPORT |
| Brotli | CPPHTTPLIB_BROTLI_SUPPORT |
| Zstandard | CPPHTTPLIB_ZSTD_SUPPORT |
httplib::Client cli("https://example.com");
cli.set_compress(true); // Compress request body
cli.set_decompress(true); // Decompress response body
You can connect through an HTTP proxy.
httplib::Client cli("https://example.com");
cli.set_proxy("proxy.example.com", 8080);
cli.set_proxy_basic_auth("user", "password");
You can set connection, read, and write timeouts individually.
httplib::Client cli("https://example.com");
cli.set_connection_timeout(5, 0); // 5 seconds
cli.set_read_timeout(10, 0); // 10 seconds
cli.set_write_timeout(10, 0); // 10 seconds
If you're making multiple requests to the same server, enable Keep-Alive. It reuses the TCP connection, which is much more efficient.
httplib::Client cli("https://example.com");
cli.set_keep_alive(true);
You can hook into request processing before and after handlers run.
svr.set_pre_routing_handler([](const auto &req, auto &res) {
// Runs before every request
return httplib::Server::HandlerResponse::Unhandled; // Continue to normal routing
});
svr.set_post_routing_handler([](const auto &req, auto &res) {
// Runs after the response is sent
res.set_header("X-Server", "cpp-httplib");
});
Use res.user_data to pass data from middleware to handlers. This is useful for sharing things like decoded auth tokens.
svr.set_pre_routing_handler([](const auto &req, auto &res) {
res.user_data["auth_user"] = std::string("alice");
return httplib::Server::HandlerResponse::Unhandled;
});
svr.Get("/me", [](const auto &req, auto &res) {
auto user = std::any_cast<std::string>(res.user_data.at("auth_user"));
res.set_content("Hello, " + user, "text/plain");
});
You can also customize error and exception handlers.
svr.set_error_handler([](const auto &req, auto &res) {
res.set_content("Custom Error Page", "text/html");
});
svr.set_exception_handler([](const auto &req, auto &res, std::exception_ptr ep) {
res.status = 500;
res.set_content("Internal Server Error", "text/plain");
});
You can set a logger on both the server and the client.
svr.set_logger([](const auto &req, const auto &res) {
std::cout << req.method << " " << req.path << " " << res.status << std::endl;
});
In addition to TCP, we support Unix Domain Sockets. You can use them for inter-process communication on the same machine.
// Server
httplib::Server svr;
svr.set_address_family(AF_UNIX);
svr.listen("/tmp/httplib.sock", 0);
// Client
httplib::Client cli("http://localhost");
cli.set_address_family(AF_UNIX);
cli.set_hostname_addr_map({{"localhost", "/tmp/httplib.sock"}});
auto res = cli.Get("/");
Want to dig deeper? Check out these resources.