onetool 0.0.1-alpha.7

use tracing_subscriber::EnvFilter;

const MODEL: &str = "gpt-4o-mini"; // or "gemini-2.0-flash" or other model supporting tool calls

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Initialize tracing subscriber (controlled via RUST_LOG env var)
    tracing_subscriber::fmt()
        .with_env_filter(EnvFilter::from_default_env())
        .init();

    // Create the Lua REPL (sandboxed environment)
    let repl = onetool::Repl::new().expect("Failed to create REPL");

    let genai_client = genai::Client::default();

    // Create the tool orchestrator for easier tool handling
    let lua_repl = onetool::genai::LuaRepl::new(&repl);

    let chat_req = genai::chat::ChatRequest::new(vec![genai::chat::ChatMessage::user(
        "What's the sum of the 10 first prime numbers?",
    )])
    .with_tools(vec![lua_repl.definition()]);

    println!("--- Getting function call from model");
    let chat_res = genai_client
        .exec_chat(MODEL, chat_req.clone(), None)
        .await
        .unwrap();

    let tool_calls = chat_res.into_tool_calls();

    if tool_calls.is_empty() {
        return Err("Expected tool calls in the response".into());
    }

    println!("--- Tool calls received:");
    for tool_call in &tool_calls {
        println!("Function: {}", tool_call.fn_name);
        println!("Arguments: {}", tool_call.fn_arguments);
    }

    // Use the orchestrator to handle the tool call
    let tool_response = lua_repl.call(&tool_calls[0]);

    let chat_req = chat_req
        .append_message(tool_calls)
        .append_message(tool_response);

    println!("\n--- Getting final response with function results");
    let chat_res = genai_client.exec_chat_stream(MODEL, chat_req, None).await?;

    println!("\n--- Final response:");
    genai::chat::printer::print_chat_stream(chat_res, None).await?;

    Ok(())
}