Massive update

drb-c2-core/app/routers/upload.py

@@ -1,5 +1,5 @@
 from typing import Optional
-from fastapi import APIRouter, UploadFile, File, Form, HTTPException, Security
+from fastapi import APIRouter, BackgroundTasks, UploadFile, File, Form, HTTPException, Security
 from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
 from app.internal.storage import upload_audio
 from app.internal import firestore as fstore
@@ -12,20 +12,32 @@ _bearer = HTTPBearer(auto_error=False)
 
 @router.post("/upload")
 async def upload_call_audio(
+    background_tasks: BackgroundTasks,
     file: UploadFile = File(...),
     call_id: str = Form(...),
     node_id: str = Form(...),
+    talkgroup_id: Optional[int] = Form(None),
+    talkgroup_name: Optional[str] = Form(None),
+    system_id: Optional[str] = Form(None),
     credentials: Optional[HTTPAuthorizationCredentials] = Security(_bearer),
 ):
     """
     Receive an audio recording from an edge node.
-    Upload to GCS, update the call document in Firestore with the audio URL.
+    Upload to GCS, update the call document in Firestore with the audio URL,
+    then kick off the intelligence pipeline as a background task.
     """
     # Verify the per-node API key
     if not credentials:
         raise HTTPException(401, "Missing authorization")
     key_doc = await fstore.doc_get("node_keys", node_id)
-    if not key_doc or key_doc.get("api_key") != credentials.credentials:
+    if not key_doc:
+        logger.warning(f"Upload 401: no key_doc in Firestore for node_id={node_id!r}")
+        raise HTTPException(401, "Invalid node API key")
+    if key_doc.get("api_key") != credentials.credentials:
+        logger.warning(
+            f"Upload 401: key mismatch for node_id={node_id!r} "
+            f"(received prefix: {credentials.credentials[:8]}...)"
+        )
         raise HTTPException(401, "Invalid node API key")
 
     data = await file.read()
@@ -41,4 +53,81 @@ async def upload_call_audio(
         except Exception as e:
             logger.warning(f"Could not update call {call_id} with audio_url: {e}")
 
+        # Convert public GCS URL to gs:// URI for Speech-to-Text
+        gcs_uri = _public_url_to_gcs_uri(audio_url)
+
+        background_tasks.add_task(
+            _run_intelligence_pipeline,
+            call_id=call_id,
+            node_id=node_id,
+            system_id=system_id,
+            talkgroup_id=talkgroup_id,
+            talkgroup_name=talkgroup_name,
+            gcs_uri=gcs_uri,
+        )
+
     return {"url": audio_url}
+
+
+def _public_url_to_gcs_uri(url: str) -> Optional[str]:
+    """
+    Convert a public GCS URL like
+      https://storage.googleapis.com/bucket/calls/file.mp3
+    to a gs:// URI usable by Speech-to-Text.
+    Returns None if the URL doesn't look like a GCS URL.
+    """
+    prefix = "https://storage.googleapis.com/"
+    if url and url.startswith(prefix):
+        return "gs://" + url[len(prefix):]
+    return None
+
+
+async def _run_intelligence_pipeline(
+    call_id: str,
+    node_id: str,
+    system_id: Optional[str],
+    talkgroup_id: Optional[int],
+    talkgroup_name: Optional[str],
+    gcs_uri: Optional[str],
+) -> None:
+    """
+    Post-upload intelligence pipeline (runs as a background task):
+      1. Transcribe audio via Google STT
+      2. Extract tags/incident type from transcript
+      3. Correlate with existing incidents (or create new one)
+      4. Check alert rules and dispatch notifications
+    """
+    from app.internal import transcription, intelligence, incident_correlator, alerter
+
+    transcript: Optional[str] = None
+
+    # Step 1: Transcription
+    if gcs_uri:
+        transcript = await transcription.transcribe_call(call_id, gcs_uri)
+
+    # Step 2: Intelligence extraction
+    tags: list[str] = []
+    incident_type: Optional[str] = None
+    if transcript:
+        tags, incident_type = await intelligence.extract_tags(call_id, transcript)
+
+    # Step 3: Incident correlation
+    if incident_type:
+        await incident_correlator.correlate_call(
+            call_id=call_id,
+            node_id=node_id,
+            system_id=system_id,
+            talkgroup_name=talkgroup_name,
+            tags=tags,
+            incident_type=incident_type,
+        )
+
+    # Step 4: Alert dispatch (always runs — talkgroup ID rules don't need a transcript)
+    await alerter.check_and_dispatch(
+        call_id=call_id,
+        node_id=node_id,
+        talkgroup_id=talkgroup_id,
+        talkgroup_name=talkgroup_name,
+        tags=tags,
+        transcript=transcript,
+    )