Robovoice/Robovoice.Stt.Dhcp/DhcpSttSource.cs

using System.Net;
using System.Net.NetworkInformation;
using System.Net.Sockets;
using System.Text;
using Robovoice.Core;

namespace Robovoice.Stt.Dhcp;

public sealed class DhcpSttSource : ISttSource
{
    private const string Magic = "HKMSTR";
    private const int DhcpClientPort = 68;
    private const int DhcpServerPort = 67;
    private static readonly TimeSpan NopInterval = TimeSpan.FromMilliseconds(50);

    private Socket? _sock;
    private CancellationTokenSource? _cts;
    private Task? _receiveTask;
    private Task? _nopTask;
    private EndPoint _broadcastEp = new IPEndPoint(IPAddress.Broadcast, DhcpServerPort);
    private uint _nonce;
    private uint _session;
    private bool _disposed;

    public string InterfaceIp { get; set; } = string.Empty;

    public Action<string>? Log { get; set; }

    public event TranscriptEventHandler? TranscriptReceived;

    public Task StartAsync(CancellationToken ct = default)
    {
        ObjectDisposedException.ThrowIf(_disposed, this);
        if (_cts is not null)
            return Task.CompletedTask;

        string ip = string.IsNullOrEmpty(InterfaceIp) ? AutoDetectInterfaceIp() ?? "0.0.0.0" : InterfaceIp;

        _sock = new Socket(AddressFamily.InterNetwork, SocketType.Dgram, ProtocolType.Udp);
        _sock.SetSocketOption(SocketOptionLevel.Socket, SocketOptionName.ReuseAddress, true);
        _sock.EnableBroadcast = true;
        _sock.Bind(new IPEndPoint(IPAddress.Parse(ip), DhcpClientPort));

        Log?.Invoke($"STT: bound to {ip}:{DhcpClientPort}, broadcasting to :{DhcpServerPort}");

        _cts = CancellationTokenSource.CreateLinkedTokenSource(ct);
        _receiveTask = ReceiveLoopAsync(_cts.Token);
        return Task.CompletedTask;
    }

    public async Task StopAsync(CancellationToken ct = default)
    {
        StopNop();

        if (_cts is not null)
            _cts.Cancel();

        _sock?.Dispose();
        _sock = null;

        if (_receiveTask is not null)
        {
            try { await _receiveTask.WaitAsync(ct); }
            catch { }
            _receiveTask = null;
        }

        _cts?.Dispose();
        _cts = null;
    }

    public void SendOn()
    {
        if (_cts is null)
            return;

        _session++;
        _nonce = 0;
        Log?.Invoke($"STT: session {_session} started");
        SendNop();
        _nopTask = NopLoopAsync(_cts.Token);
    }

    public void SendOff()
    {
        StopNop();
        SendControl($"HKMSTR:OFF {_session} {_nonce}");
    }

    private void StopNop()
    {
        if (_nopTask is not null)
        {
            try { _nopTask.Wait(2000); } catch { }
            _nopTask = null;
        }
    }

    private async Task NopLoopAsync(CancellationToken ct)
    {
        while (!ct.IsCancellationRequested)
        {
            try { await Task.Delay(NopInterval, ct); }
            catch (OperationCanceledException) { break; }

            SendNop();
        }
    }

    private void SendNop()
    {
        _nonce++;
        SendControl($"HKMSTR {_session} {_nonce}");
    }

    private void SendControl(string message)
    {
        if (_sock is null)
            return;

        try
        {
            byte[] payload = Encoding.UTF8.GetBytes(message + "\n");
            _sock.SendTo(payload, _broadcastEp);
        }
        catch (Exception ex)
        {
            Log?.Invoke($"STT: send failed: {ex.Message}");
        }
    }

    private async Task ReceiveLoopAsync(CancellationToken ct)
    {
        byte[] buffer = new byte[4096];
        EndPoint fromEp = new IPEndPoint(IPAddress.Any, 0);

        while (!ct.IsCancellationRequested)
        {
            int received;
            try
            {
                if (!_sock!.Poll(500_000, SelectMode.SelectRead))
                    continue;
                received = _sock.ReceiveFrom(buffer, ref fromEp);
            }
            catch (OperationCanceledException)
            {
                break;
            }
            catch (ObjectDisposedException)
            {
                break;
            }
            catch (Exception ex)
            {
                Log?.Invoke($"STT: receive error: {ex.Message}");
                continue;
            }

            string text = Encoding.UTF8.GetString(buffer, 0, received).TrimEnd('\n', '\r');
            if (!text.StartsWith(Magic))
                continue;

            TranscriptMessage? message = ParseReply(text);
            if (message is null)
                continue;

            TranscriptReceived?.Invoke(this, new TranscriptEventArgs
            {
                Message = message,
            });
        }
    }

    private TranscriptMessage? ParseReply(string text)
    {
        // Format: HKMSTR:P <session> <text>  or  HKMSTR:F <session> <text>
        // <text> may be empty.
        string prefix;
        TranscriptType type;

        if (text.StartsWith("HKMSTR:P "))
        {
            prefix = "HKMSTR:P ";
            type = TranscriptType.Partial;
        }
        else if (text.StartsWith("HKMSTR:F "))
        {
            prefix = "HKMSTR:F ";
            type = TranscriptType.Final;
        }
        else
        {
            return null;
        }

        string rest = text[prefix.Length..];

        int spaceIndex = rest.IndexOf(' ');
        if (spaceIndex < 0)
        {
            if (uint.TryParse(rest, out uint sessionOnly))
            {
                if (sessionOnly != _session)
                    return null;
                return new TranscriptMessage(type, string.Empty);
            }
            return null;
        }

        string sessionStr = rest[..spaceIndex];
        if (!uint.TryParse(sessionStr, out uint session))
            return null;

        if (session != _session)
        {
            Log?.Invoke($"STT: dropping stale reply (session {session} != current {_session})");
            return null;
        }

        string transcript = rest[(spaceIndex + 1)..];
        return new TranscriptMessage(type, transcript);
    }

    public static List<(string Ip, string Name)> GetAvailableInterfaces()
    {
        var result = new List<(string, string)>();

        foreach (var nic in NetworkInterface.GetAllNetworkInterfaces())
        {
            if (nic.OperationalStatus != OperationalStatus.Up)
                continue;
            if (nic.NetworkInterfaceType == NetworkInterfaceType.Loopback)
                continue;

            string desc = nic.Description;
            if (desc.Contains("WireGuard", StringComparison.OrdinalIgnoreCase))
                continue;

            foreach (var addr in nic.GetIPProperties().UnicastAddresses)
            {
                if (addr.Address.AddressFamily != AddressFamily.InterNetwork)
                    continue;

                string ip = addr.Address.ToString();
                if (ip.StartsWith("192.168.") || ip.StartsWith("10.") ||
                    ip.StartsWith("172.16.") || ip.StartsWith("172.17.") ||
                    ip.StartsWith("172.18.") || ip.StartsWith("172.19.") ||
                    ip.StartsWith("172.20.") || ip.StartsWith("172.21.") ||
                    ip.StartsWith("172.22.") || ip.StartsWith("172.23.") ||
                    ip.StartsWith("172.24.") || ip.StartsWith("172.25.") ||
                    ip.StartsWith("172.26.") || ip.StartsWith("172.27.") ||
                    ip.StartsWith("172.28.") || ip.StartsWith("172.29.") ||
                    ip.StartsWith("172.30.") || ip.StartsWith("172.31."))
                {
                    result.Add((ip, $"{nic.Name} ({ip})"));
                }
            }
        }

        return result;
    }

    private static string? AutoDetectInterfaceIp()
    {
        foreach (var (ip, _) in GetAvailableInterfaces())
            return ip;
        return null;
    }

    public async ValueTask DisposeAsync()
    {
        if (_disposed) return;
        await StopAsync();
        _disposed = true;
    }
}