feat(matrix-bridge-dagi): add room mapping, ingress loop, synapse setup (PR-M1.2 + PR-M1.3)

PR-M1.2 — room-to-agent mapping:
- adds room_mapping.py: parse BRIDGE_ROOM_MAP (format: agent:!room_id:server)
- RoomMappingConfig with O(1) room→agent lookup, agent allowlist check
- /bridge/mappings endpoint (read-only ops summary, no secrets)
- health endpoint now includes mappings_count
- 21 tests for parsing, validation, allowlist, summary

PR-M1.3 — Matrix ingress loop:
- adds ingress.py: MatrixIngressLoop asyncio task
- sync_poll → extract → dedupe → _invoke_gateway (POST /v1/invoke)
- gateway payload: agent_id, node_id, message, metadata (transport, room_id, event_id, sender)
- exponential backoff on errors (2s..60s)
- joins all mapped rooms at startup
- metric callbacks: on_message_received, on_gateway_error
- graceful shutdown via asyncio.Event
- 5 ingress tests (invoke, dedupe, callbacks, empty-map idle)

Synapse setup (docker-compose.synapse-node1.yml):
- fixed volume: bind mount ./synapse-data instead of named volume
- added port mapping 127.0.0.1:8008:8008

Synapse running on NODA1 (localhost:8008), bot @dagi_bridge:daarion.space created,
room !QwHczWXgefDHBEVkTH:daarion.space created, all 4 values in .env on NODA1.

Made-with: Cursor
This commit is contained in:
Apple
2026-03-03 07:51:13 -08:00
parent d8506da179
commit dbfab78f02
6 changed files with 1010 additions and 5 deletions

View File

@@ -4,6 +4,7 @@ Bridges Matrix/Element rooms to DAGI agents via Gateway.
M1 scope: 1 room ↔ 1 agent (Sofiia), audit via sofiia-console internal endpoint.
"""
import asyncio
import logging
import os
import time
@@ -31,6 +32,8 @@ except ImportError: # pragma: no cover
_PROM_OK = False
from .config import BridgeConfig, load_config
from .ingress import MatrixIngressLoop
from .room_mapping import RoomMappingConfig, parse_room_map
logging.basicConfig(
level=logging.INFO,
@@ -71,6 +74,9 @@ _cfg: Optional[BridgeConfig] = None
_config_error: Optional[str] = None
_matrix_reachable: Optional[bool] = None # probed at startup
_gateway_reachable: Optional[bool] = None # probed at startup
_room_map: Optional[RoomMappingConfig] = None
_ingress_task: Optional[asyncio.Task] = None
_ingress_stop: Optional[asyncio.Event] = None
async def _probe_url(url: str, timeout: float = 5.0) -> bool:
@@ -87,14 +93,24 @@ async def _probe_url(url: str, timeout: float = 5.0) -> bool:
# ── Lifespan ──────────────────────────────────────────────────────────────────
@asynccontextmanager
async def lifespan(app_: Any):
global _cfg, _config_error, _matrix_reachable, _gateway_reachable
global _cfg, _config_error, _matrix_reachable, _gateway_reachable, _room_map
try:
_cfg = load_config()
# Parse room mapping
_room_map = parse_room_map(
os.getenv("BRIDGE_ROOM_MAP", ""),
_cfg.bridge_allowed_agents,
)
logger.info(
"✅ matrix-bridge-dagi started | node=%s build=%s homeserver=%s room=%s agents=%s",
"✅ matrix-bridge-dagi started | node=%s build=%s homeserver=%s "
"room=%s agents=%s mappings=%d",
_cfg.node_id, _cfg.build_sha, _cfg.matrix_homeserver_url,
_cfg.sofiia_room_id, list(_cfg.bridge_allowed_agents),
_room_map.total_mappings,
)
# Connectivity smoke probes (non-blocking failures)
_matrix_reachable = await _probe_url(
f"{_cfg.matrix_homeserver_url}/_matrix/client/versions"
@@ -112,12 +128,52 @@ async def lifespan(app_: Any):
logger.warning("⚠️ DAGI Gateway NOT reachable: %s", _cfg.dagi_gateway_url)
if _PROM_OK:
_bridge_up.set(1)
except RuntimeError as exc:
# Start ingress loop (fire-and-forget asyncio task)
if _room_map and _room_map.total_mappings > 0:
_ingress_stop = asyncio.Event()
def _on_msg(room_id: str, agent_id: str) -> None:
if _PROM_OK:
_messages_received.labels(room_id=room_id, agent_id=agent_id).inc()
def _on_gw_error(error_type: str) -> None:
if _PROM_OK:
_gateway_errors.labels(error_type=error_type).inc()
ingress = MatrixIngressLoop(
matrix_homeserver_url=_cfg.matrix_homeserver_url,
matrix_access_token=_cfg.matrix_access_token,
matrix_user_id=_cfg.matrix_user_id,
gateway_url=_cfg.dagi_gateway_url,
node_id=_cfg.node_id,
room_map=_room_map,
on_message_received=_on_msg,
on_gateway_error=_on_gw_error,
)
_ingress_task = asyncio.create_task(
ingress.run(_ingress_stop),
name="matrix_ingress_loop",
)
logger.info("✅ Ingress loop task started")
else:
logger.warning("⚠️ No room mappings — ingress loop NOT started")
except (RuntimeError, ValueError) as exc:
_config_error = str(exc)
logger.error("❌ Config error: %s", _config_error)
if _PROM_OK:
_bridge_up.set(0)
yield
# Shutdown: cancel ingress loop
if _ingress_stop:
_ingress_stop.set()
if _ingress_task and not _ingress_task.done():
_ingress_task.cancel()
try:
await asyncio.wait_for(_ingress_task, timeout=5.0)
except (asyncio.CancelledError, asyncio.TimeoutError):
pass
logger.info("matrix-bridge-dagi shutting down")
# ── App ───────────────────────────────────────────────────────────────────────
@@ -166,9 +222,32 @@ async def health() -> Dict[str, Any]:
"allowed_agents": list(_cfg.bridge_allowed_agents),
"gateway": _cfg.dagi_gateway_url,
"gateway_reachable": _gateway_reachable,
"mappings_count": _room_map.total_mappings if _room_map else 0,
"config_ok": True,
}
# ── Bridge Mappings (read-only ops endpoint) ───────────────────────────────────
@app.get("/bridge/mappings")
async def bridge_mappings() -> Dict[str, Any]:
"""
Returns room-to-agent mapping summary.
Safe for ops visibility — no secrets included.
"""
if _cfg is None or _room_map is None:
return {
"ok": False,
"error": _config_error or "service not initialised",
"mappings": [],
}
return {
"ok": True,
"total": _room_map.total_mappings,
"allowed_agents": list(_cfg.bridge_allowed_agents),
"mappings": _room_map.as_summary(),
}
# ── Metrics ───────────────────────────────────────────────────────────────────
@app.get("/metrics")
async def metrics():