Update intelligence

drb-c2-core/app/internal/summarizer.py

@@ -0,0 +1,114 @@
+"""
+Background incident summary loop.
+
+Runs every SUMMARY_INTERVAL_MINUTES. Finds all active incidents with
+summary_stale=True, fetches all their call transcripts, and calls Gemini
+once per incident to produce a concise factual summary.
+
+By batching this way: Gemini is never called per-call — only periodically
+and only for incidents that have actually changed since the last run.
+"""
+import asyncio
+from datetime import datetime, timezone
+from typing import Optional
+from app.internal.logger import logger
+from app.internal import firestore as fstore
+from app.config import settings
+
+
+async def summarizer_loop() -> None:
+    interval = settings.summary_interval_minutes * 60
+    logger.info(f"Summarizer started — interval: {settings.summary_interval_minutes}m")
+    while True:
+        await asyncio.sleep(interval)
+        try:
+            await _run_summary_pass()
+        except Exception as e:
+            logger.error(f"Summarizer pass failed: {e}")
+
+
+async def _run_summary_pass() -> None:
+    stale = await fstore.collection_list("incidents", status="active", summary_stale=True)
+    if not stale:
+        return
+
+    logger.info(f"Summarizer: processing {len(stale)} stale incident(s)")
+    for inc in stale:
+        await _summarize_incident(inc)
+
+
+async def _summarize_incident(inc: dict) -> None:
+    incident_id = inc.get("incident_id")
+    if not incident_id:
+        return
+
+    call_ids: list[str] = inc.get("call_ids", [])
+    if not call_ids:
+        return
+
+    # Fetch transcripts for all calls in this incident
+    transcripts: list[str] = []
+    for cid in call_ids:
+        doc = await fstore.doc_get("calls", cid)
+        if doc and doc.get("transcript"):
+            transcripts.append(doc["transcript"])
+
+    if not transcripts:
+        # No transcripts yet — clear stale flag and wait for next pass
+        await fstore.doc_set("incidents", incident_id, {"summary_stale": False})
+        return
+
+    summary = await asyncio.to_thread(_sync_summarize, inc, transcripts)
+
+    now = datetime.now(timezone.utc).isoformat()
+    updates: dict = {
+        "summary_stale": False,
+        "summary_last_run": now,
+    }
+    if summary:
+        updates["summary"] = summary
+        logger.info(f"Summarizer: updated summary for incident {incident_id}")
+    else:
+        logger.warning(f"Summarizer: Gemini returned nothing for incident {incident_id}")
+
+    await fstore.doc_set("incidents", incident_id, updates)
+
+
+def _sync_summarize(inc: dict, transcripts: list[str]) -> Optional[str]:
+    from app.config import settings
+    import google.generativeai as genai
+
+    if not settings.gemini_api_key:
+        return None
+
+    genai.configure(api_key=settings.gemini_api_key)
+    model = genai.GenerativeModel("gemini-1.5-flash")
+
+    inc_type = inc.get("type", "unknown")
+    location = inc.get("location") or "unknown location"
+    tg_ids = ", ".join(inc.get("talkgroup_ids", [])) or "unknown"
+    numbered = "\n".join(f"{i+1}. {t}" for i, t in enumerate(transcripts))
+
+    prompt = f"""You are analyzing P25 public safety radio communications for a single active incident.
+
+Incident type: {inc_type}
+Location: {location}
+Talkgroup(s): {tg_ids}
+
+Transcripts ({len(transcripts)} calls, chronological):
+{numbered}
+
+Write a concise factual summary of this incident in 2-4 sentences. Include:
+- What happened
+- Location (most specific mentioned)
+- Units or resources involved if mentioned
+- Current status if determinable
+
+Be factual. Do not speculate beyond what the transcripts say. Do not use bullet points."""
+
+    try:
+        response = model.generate_content(prompt)
+        return response.text.strip() or None
+    except Exception as e:
+        logger.warning(f"Gemini summary failed: {e}")
+        return None