path: root/packages/meshbay-node/src/meshbay_node/transport/quic_server.py

"""
MeshBay Node — QUIC chunk server (MNP v2).

Replaces the TCP+TLS ChunkServer with QUIC transport.
Advantages over TCP:
  - UDP-based → works with hole punching (Spike 4 confirmed Cone NAT on SFR)
  - Multiplexed streams — each request is an independent QUIC stream
  - 0-RTT reconnection (connection resumption)
  - Built-in TLS 1.3

Wire protocol:
  - Each bidirectional QUIC stream carries one request/response exchange
  - Messages: length-prefixed msgpack (4-byte big-endian, same as TCP+TLS)
  - MNP handshake on stream 0 (control stream); subsequent streams = requests

Application protocol (MNP) is identical to TCP+TLS version.
The transport is the only change — all crypto, auth, and message types stay the same.
"""

import asyncio
import base64
import logging
import struct
import subprocess
from pathlib import Path
from typing import Any, Callable

import blake3
import jwt
import msgpack
from aioquic.asyncio import QuicConnectionProtocol, serve
from aioquic.quic.configuration import QuicConfiguration
from aioquic.quic.events import QuicEvent, StreamDataReceived, StreamReset
from cryptography.hazmat.primitives.asymmetric.ed25519 import Ed25519PrivateKey

from meshbay_common import MNP_VERSION
from meshbay_common.crypto import (
    chunk_key as derive_chunk_key,
    encrypt_chunk,
    sign_chunk,
    pk_to_b64,
)
from meshbay_common.protocol import MNP
from meshbay_node.indexer import GroupIndex
from meshbay_node.transport.tls_cert import server_ssl_context

log = logging.getLogger(__name__)

CHUNK_SIZE = 1024 * 1024
MAX_MSG    = 64 * 1024 * 1024
ALPN       = ["meshbay-mnp"]


class Denylist:
    """Shared denylist for revoked users and invalidated JWTs."""

    def __init__(self):
        self.user_ids: set[str] = set()
        self.jtis: set[str] = set()

    def is_denied(self, user_id: str, jti: str) -> bool:
        return user_id in self.user_ids or jti in self.jtis

    def deny_user(self, user_id: str) -> None:
        self.user_ids.add(user_id)
        log.info("Denied user: %s", user_id[:8])

    def deny_jti(self, jti: str) -> None:
        self.jtis.add(jti)
        log.info("Denied jti: %s", jti[:8])


# ── Wire helpers ──────────────────────────────────────────────────────────────

def _pack(obj: dict) -> bytes:
    data = msgpack.packb(obj, use_bin_type=True)
    return struct.pack(">I", len(data)) + data


class _StreamBuffer:
    """Accumulate incoming QUIC stream data and extract length-prefixed messages."""

    def __init__(self):
        self._buf = bytearray()

    def feed(self, data: bytes):
        self._buf.extend(data)

    def messages(self):
        while len(self._buf) >= 4:
            length = struct.unpack(">I", self._buf[:4])[0]
            if length > MAX_MSG:
                raise ValueError(f"Message too large: {length}")
            if len(self._buf) < 4 + length:
                break
            msg_bytes = bytes(self._buf[4:4 + length])
            del self._buf[:4 + length]
            yield msgpack.unpackb(msg_bytes, raw=False)


# ── Per-connection server protocol ────────────────────────────────────────────

class _MNPServerProtocol(QuicConnectionProtocol):
    """
    One instance per QUIC connection.
    Handles the MNP handshake and all subsequent streams.
    """

    def __init__(self, *args, node_ctx: dict, **kwargs):
        super().__init__(*args, **kwargs)
        self._ctx        = node_ctx    # shared server context (keys, index, etc.)
        self._user_id: str | None = None
        self._group_id: str | None = None
        self._buffers: dict[int, _StreamBuffer] = {}

    def quic_event_received(self, event: QuicEvent) -> None:
        if isinstance(event, StreamDataReceived):
            sid = event.stream_id
            if sid not in self._buffers:
                self._buffers[sid] = _StreamBuffer()
            self._buffers[sid].feed(event.data)
            for msg in self._buffers[sid].messages():
                self._handle_message_sync(sid, msg)

        elif isinstance(event, StreamReset):
            self._buffers.pop(event.stream_id, None)

    def _handle_message_sync(self, stream_id: int, msg: dict) -> None:
        """Handle an MNP message synchronously (called from quic_event_received)."""
        mtype = msg.get("type")
        try:
            if mtype == MNP.HANDSHAKE:
                self._do_handshake_sync(stream_id, msg)
            elif self._user_id is None:
                self._send(stream_id, {"type": "error", "detail": "Handshake required"})
            elif mtype == MNP.INDEX_SYNC:
                self._do_index_sync_sync(stream_id)
            elif mtype == MNP.FILE_REQUEST:
                self._do_file_request_sync(stream_id, msg)
            elif mtype == MNP.STREAM_SEGMENT:
                self._do_stream_segment_sync(stream_id, msg)
            elif mtype == MNP.CHAT_MESSAGE:
                self._do_chat_message_sync(stream_id, msg)
            else:
                log.warning("Unknown MNP message type: %s", mtype)
        except Exception as e:
            log.error("Error handling %s: %s", mtype, e)
            self._send(stream_id, {"type": "error", "detail": str(e)})

    def _do_handshake_sync(self, stream_id: int, msg: dict) -> None:
        token = msg.get("token", "")
        group_id = msg.get("group_id", "")
        try:
            decoded = jwt.decode(token, self._ctx["hub_pk_pem"], algorithms=["EdDSA"])
        except Exception as e:
            self._send(stream_id, {"type": "error", "detail": f"Invalid JWT: {e}"})
            self._quic.close()
            return

        denylist = self._ctx.get("denylist")
        if denylist and denylist.is_denied(decoded.get("sub", ""), decoded.get("jti", "")):
            self._send(stream_id, {"type": "error", "detail": "Token revoked"})
            self._quic.close()
            return

        if group_id and group_id not in decoded.get("groups", []):
            self._send(stream_id, {"type": "error", "detail": "Not a member of this group"})
            self._quic.close()
            return

        if group_id and "groups" in self._ctx and group_id not in self._ctx["groups"]:
            self._send(stream_id, {"type": "error", "detail": "Group not hosted on this node"})
            self._quic.close()
            return

        self._user_id = decoded["sub"]
        self._group_id = group_id

        peers = self._ctx.get("_peers")
        if peers is not None:
            peers[self._user_id] = self

        log.info("QUIC handshake OK — user=%s group=%s", self._user_id[:8], group_id[:8] if group_id else "none")
        self._send(stream_id, {
            "type":    MNP.HANDSHAKE_ACK,
            "v":       MNP_VERSION,
            "node_pk": pk_to_b64(self._ctx["sk_node"].public_key()),
        })

    def _group_ctx(self) -> dict:
        """Resolve the active group context (multi-group or legacy single-group)."""
        if "groups" in self._ctx and self._group_id:
            return self._ctx["groups"][self._group_id]
        return self._ctx

    def _do_index_sync_sync(self, stream_id: int) -> None:
        ctx = self._group_ctx()
        wire = ctx["index"].serialize()
        self._send(stream_id, {
            "type":      MNP.INDEX_SYNC,
            "v":         MNP_VERSION,
            "index_b64": base64.b64encode(wire).decode(),
        })

    def _do_file_request_sync(self, stream_id: int, msg: dict) -> None:
        """Serve file chunk synchronously (blocking I/O — acceptable for test sizes)."""
        ctx = self._group_ctx()
        file_id     = msg["file_id"]
        chunk_index = msg["chunk_index"]
        entry = ctx["index"].get_entry(file_id)
        if not entry:
            self._send(stream_id, {"type": "error", "detail": "File not found"})
            return

        file_path = ctx["shared_root"] / entry.path / entry.name
        if not file_path.exists():
            self._send(stream_id, {"type": "error", "detail": "File not on disk"})
            return

        chunk_data = _read_and_encrypt(
            self._ctx["sk_node"],
            ctx["gek"],
            file_path,
            chunk_index,
        )
        self._send(stream_id, chunk_data)

    def _do_stream_segment_sync(self, stream_id: int, msg: dict) -> None:
        """Extract and serve one HLS segment via ffmpeg."""
        ctx = self._group_ctx()
        file_id = msg["file_id"]
        segment_index = msg["segment_index"]
        segment_duration = msg.get("segment_duration", 4)

        entry = ctx["index"].get_entry(file_id)
        if not entry:
            self._send(stream_id, {"type": "error", "detail": "File not found"})
            return

        file_path = ctx["shared_root"] / entry.path / entry.name
        if not file_path.exists():
            self._send(stream_id, {"type": "error", "detail": "File not on disk"})
            return

        start_time = segment_index * segment_duration
        segment_data = _extract_segment(file_path, start_time, segment_duration)
        if segment_data is None:
            self._send(stream_id, {"type": "error", "detail": "Segment extraction failed"})
            return

        self._send(stream_id, {
            "type": MNP.STREAM_SEGMENT,
            "v": MNP_VERSION,
            "file_id": file_id,
            "segment_index": segment_index,
            "data_b64": base64.b64encode(segment_data).decode(),
            "size": len(segment_data),
        })

    def _do_chat_message_sync(self, stream_id: int, msg: dict) -> None:
        """Receive a chat message, store it, and broadcast to other connected peers."""
        chat_store = self._ctx.get("chat_store")
        if chat_store:
            import asyncio
            asyncio.ensure_future(chat_store.save_message(
                sender_id=msg.get("sender_id", self._user_id),
                iteration=msg.get("iteration", 0),
                payload=msg.get("payload", b"").encode() if isinstance(msg.get("payload"), str) else msg.get("payload", b""),
                thread_id=msg.get("thread_id"),
            ))

        peers = self._ctx.get("_peers", {})
        broadcast = {
            "type": MNP.CHAT_MESSAGE,
            "v": MNP_VERSION,
            "sender_id": msg.get("sender_id", self._user_id),
            "iteration": msg.get("iteration", 0),
            "payload": msg.get("payload", ""),
            "thread_id": msg.get("thread_id"),
            "group_id": self._group_id or "",
        }
        for uid, proto in peers.items():
            if uid != self._user_id and proto is not self:
                try:
                    proto._send(0, broadcast)
                except Exception:
                    pass

        self._send(stream_id, {"type": "ack", "v": MNP_VERSION})

    def connection_lost(self, exc) -> None:
        peers = self._ctx.get("_peers")
        if peers and self._user_id:
            peers.pop(self._user_id, None)
        super().connection_lost(exc)

    def _send(self, stream_id: int, obj: dict) -> None:
        self._quic.send_stream_data(stream_id, _pack(obj))
        self.transmit()


def _read_and_encrypt(
    sk_node: Ed25519PrivateKey,
    gek: bytes,
    file_path: Path,
    chunk_index: int,
) -> dict:
    """Read and encrypt one chunk (blocking — runs in executor)."""
    with open(file_path, "rb") as f:
        f.seek(chunk_index * CHUNK_SIZE)
        plaintext = f.read(CHUNK_SIZE)

    file_hash = blake3.blake3(file_path.read_bytes()).digest()
    pt_hash   = blake3.blake3(plaintext).digest()
    ckey      = derive_chunk_key(gek, file_hash, chunk_index)
    nonce, ct = encrypt_chunk(ckey, plaintext)
    ct_hash   = blake3.blake3(ct).digest()
    sig       = sign_chunk(sk_node, chunk_index, nonce, ct_hash)

    return {
        "type":           MNP.FILE_CHUNK,
        "v":              MNP_VERSION,
        "chunk_index":    chunk_index,
        "plaintext_size": len(plaintext),
        "nonce_b64":      base64.b64encode(nonce).decode(),
        "ct_b64":         base64.b64encode(ct).decode(),
        "ct_hash_b64":    base64.b64encode(ct_hash).decode(),
        "pt_hash_b64":    base64.b64encode(pt_hash).decode(),
        "sig_b64":        base64.b64encode(sig).decode(),
        "pk_node_b64":    pk_to_b64(sk_node.public_key()),
        "file_hash_b64":  base64.b64encode(file_hash).decode(),
    }


def _extract_segment(file_path: Path, start_time: float, duration: float) -> bytes | None:
    """Extract one HLS segment via ffmpeg. Returns MPEG-TS bytes or None on failure."""
    try:
        result = subprocess.run(
            ["ffmpeg", "-hide_banner", "-loglevel", "error",
             "-ss", str(start_time),
             "-i", str(file_path),
             "-t", str(duration),
             "-c:v", "copy", "-c:a", "copy",
             "-f", "mpegts", "pipe:1"],
            capture_output=True, timeout=30,
        )
        if result.returncode == 0 and result.stdout:
            return result.stdout
        return None
    except Exception:
        return None


# ── QuicChunkServer ────────────────────────────────────────────────────────────

class QuicChunkServer:
    """
    QUIC-based MNP chunk server (MNP v2).
    Drop-in replacement for ChunkServer with UDP transport.
    """

    def __init__(
        self,
        sk_node: Ed25519PrivateKey,
        hub_pk_pem: bytes,
        gek: bytes,
        shared_root: Path,
        index: GroupIndex,
        host: str = "::",   # listen IPv4 + IPv6 (dual-stack Linux)
        port: int = 19000,
        cert_path: Path | None = None,
        key_path: Path | None = None,
        groups: dict[str, dict] | None = None,
        denylist: Denylist | None = None,
    ):
        self._ctx = {
            "sk_node":     sk_node,
            "hub_pk_pem":  hub_pk_pem,
            "gek":         gek,
            "shared_root": shared_root,
            "index":       index,
        }
        if groups:
            self._ctx["groups"] = groups
        self._denylist = denylist or Denylist()
        self._ctx["denylist"] = self._denylist
        self._ctx["_peers"] = {}
        self._host      = host
        self._port      = port
        self._cert_path = cert_path or Path.home() / ".config/meshbay/node_tls.crt"
        self._key_path  = key_path  or Path.home() / ".config/meshbay/node_tls.key"
        self._server    = None
        self._task      = None
        self._session_tickets: dict[bytes, Any] = {}

    @property
    def port(self) -> int:
        return self._port

    @property
    def denylist(self) -> Denylist:
        return self._denylist

    def _make_config(self) -> QuicConfiguration:
        from meshbay_node.transport.tls_cert import generate_self_signed_cert
        if not self._cert_path.exists():
            generate_self_signed_cert(self._cert_path, self._key_path)
        config = QuicConfiguration(is_client=False, alpn_protocols=ALPN)
        config.load_cert_chain(str(self._cert_path), str(self._key_path))
        return config

    def _store_ticket(self, ticket: Any) -> None:
        self._session_tickets[ticket.ticket] = ticket

    def _fetch_ticket(self, label: bytes) -> Any | None:
        return self._session_tickets.pop(label, None)

    async def start(self) -> None:
        config = self._make_config()
        ctx = self._ctx

        def protocol_factory(*args, **kwargs):
            return _MNPServerProtocol(*args, node_ctx=ctx, **kwargs)

        self._server = await serve(
            self._host, self._port,
            configuration=config,
            create_protocol=protocol_factory,
            session_ticket_handler=self._store_ticket,
            session_ticket_fetcher=self._fetch_ticket,
        )
        log.info("QuicChunkServer listening on %s:%d (QUIC/UDP)", self._host, self._port)

    def punch_nat(self, peer_ip: str, peer_port: int) -> None:
        """
        Send a probe UDP packet FROM the QUIC server's own socket.

        Critical for Port-Restricted Cone NAT (SFR résidentiel) :
        the NAT only allows inbound from (peer_ip, peer_port) if the server
        previously sent a packet TO (peer_ip, peer_port) from this same socket.

        The QUIC client must connect FROM peer_port for the NAT entry to match.
        """
        if self._server and hasattr(self._server, '_transport') and self._server._transport:
            self._server._transport.sendto(b'MESHBAY:NAT:PUNCH', (peer_ip, peer_port))
            log.info("NAT probe sent → %s:%d", peer_ip, peer_port)
        else:
            log.warning("punch_nat: server transport not available")

    async def stop(self) -> None:
        if self._server:
            self._server.close()
            self._server = None
            log.info("QuicChunkServer stopped")