Files
server-26/drb-c2-core/app/routers/incidents.py
T
Logan Cusano d18e4f0743
Build & Deploy / Build & push images (push) Successful in 4m2s
Build & Deploy / Deploy to VM (push) Successful in 1m53s
Build & Deploy / Report a failed deploy (push) Skipped
Make "AI is off" true, and stop the transcript PATCH from destroying calls
config/ai_features was not the switch it was documented to be. Three paths
spent money with it off, and one path read it wrong, so per-system opt-outs
did not opt anything out.

- Correlation in the ingest pipeline tested the raw global flag instead of the
  per-system resolution. With a system opted out, extraction was skipped but
  the no-scenes fallback still correlated the call with empty tags, taking the
  thin/recency path and attaching it to whatever incident was most recent on
  that system. The opt-out did not disable correlation, it disabled good
  correlation and left the worst kind running. (#75)

- Transcript correction ran on every transcribed call gated only by an env var,
  spending Gemini tokens and a Places lookup per proposed location. An
  "STT-only" window was never STT-only and its cost could not be attributed.
  Now behind transcript_correction_enabled. (#76)

- _run_extraction_pipeline and the vocabulary learner, both reachable from
  PATCH /calls/{id}/transcript, checked no flags at all. (#76, #81)

The flag resolver now lives in feature_flags.resolve_flags() rather than as a
local helper in upload.py. Three copies of that logic is how #75 happened.

PATCH /calls/{id}/transcript now refuses with 409 when correlation is off.
That route wipes tags, severity, location, units, embedding and unlinks the
call from every incident before queueing re-extraction. Gating extraction
alone would have made it destructive-only in the standing flags-off
configuration: the call left blank and orphaned forever, with the route still
answering 200. The wipe and the rebuild are one transaction in intent, so it
refuses before the first write.

Also: the summarizer's stale-incident sweep is no longer behind
summaries_enabled. It is pure Firestore with no model call in it, and gating
it meant nothing auto-resolved while AI was off - so every incident stayed
active forever and the candidate set every correlation reads kept growing.

transcript_correction_enabled is documented as NOT a pure cost lever. The
corrector is also the noise gate that sets not_speech; with it off, recogniser
noise reaches extraction as a real transcript, comes back thin, and
auto-attaches. Never open an evaluation window with correction off and
correlation on.

14 tests added covering flag precedence, both pipeline paths, the 409, the
correction gate and the summarizer no-op. Suite: 264 passed.

Refs #75, #76, #81, #45.
2026-08-27 02:49:09 -04:00

186 lines
7.0 KiB
Python

import uuid
from datetime import datetime, timezone
from typing import Optional
from fastapi import APIRouter, BackgroundTasks, HTTPException, Depends
from app.models import IncidentCreate, IncidentUpdate
from app.internal import firestore as fstore
from app.internal.auth import (
require_admin_token,
require_service_or_firebase_token,
resolve_caller_org_id,
summarize_limiter,
)
router = APIRouter(prefix="/incidents", tags=["incidents"])
@router.get("")
async def list_incidents(
status: Optional[str] = None,
type: Optional[str] = None,
decoded: dict = Depends(require_service_or_firebase_token),
):
filters = {}
if status:
filters["status"] = status
if type:
filters["type"] = type
org_id = await resolve_caller_org_id(decoded)
if org_id is not None:
filters["org_id"] = org_id
return await fstore.collection_list("incidents", **filters)
@router.post("/summarize")
async def summarize_all_stale(
background_tasks: BackgroundTasks,
_: dict = Depends(require_admin_token),
):
"""Immediately run the summarizer pass on all stale incidents (don't wait for the next interval)."""
from app.internal.summarizer import _run_summary_pass
background_tasks.add_task(_run_summary_pass)
return {"ok": True}
@router.get("/{incident_id}")
async def get_incident(incident_id: str, decoded: dict = Depends(require_service_or_firebase_token)):
doc = await fstore.doc_get("incidents", incident_id)
if not doc:
raise HTTPException(404, f"Incident '{incident_id}' not found.")
org_id = await resolve_caller_org_id(decoded)
if org_id is not None and doc.get("org_id") != org_id:
raise HTTPException(404, f"Incident '{incident_id}' not found.")
return doc
@router.post("")
async def create_incident(body: IncidentCreate, _: dict = Depends(require_admin_token)):
now = datetime.now(timezone.utc).isoformat()
incident_id = str(uuid.uuid4())
doc = {
"incident_id": incident_id,
"title": body.title,
"type": body.type,
"status": body.status,
"location": body.location,
"call_ids": body.call_ids,
"summary": body.summary,
"tags": body.tags,
"started_at": now,
"updated_at": now,
}
await fstore.doc_set("incidents", incident_id, doc, merge=False)
return doc
@router.put("/{incident_id}")
async def update_incident(incident_id: str, body: IncidentUpdate, _: dict = Depends(require_admin_token)):
doc = await fstore.doc_get("incidents", incident_id)
if not doc:
raise HTTPException(404, f"Incident '{incident_id}' not found.")
updates = body.model_dump(exclude_none=True)
updates["updated_at"] = datetime.now(timezone.utc).isoformat()
await fstore.doc_update("incidents", incident_id, updates)
return {**doc, **updates}
@router.delete("/{incident_id}")
async def delete_incident(incident_id: str, _: dict = Depends(require_admin_token)):
doc = await fstore.doc_get("incidents", incident_id)
if not doc:
raise HTTPException(404, f"Incident '{incident_id}' not found.")
await fstore.doc_delete("incidents", incident_id)
return {"ok": True}
@router.post("/{incident_id}/summarize")
async def summarize_incident(
incident_id: str,
background_tasks: BackgroundTasks,
decoded: dict = Depends(require_service_or_firebase_token),
):
"""Immediately run the summarizer for a specific incident."""
from app.internal.summarizer import _summarize_incident
from app.internal.feature_flags import get_flags
inc = await fstore.doc_get("incidents", incident_id)
if not inc:
raise HTTPException(404, f"Incident '{incident_id}' not found.")
flags = await get_flags()
if not flags["summaries_enabled"]:
return {"ok": False, "incident_id": incident_id, "summaries_enabled": False}
# Rate limit by incident ID to prevent repeated expensive LLM calls
summarize_limiter.check(incident_id)
background_tasks.add_task(_summarize_incident, inc)
return {"ok": True, "incident_id": incident_id, "summaries_enabled": True}
@router.post("/{incident_id}/calls/{call_id}")
async def link_call_to_incident(incident_id: str, call_id: str, _: dict = Depends(require_admin_token)):
"""Manually attach a call to an incident (the /calls page's attribution action)."""
doc = await fstore.doc_get("incidents", incident_id)
if not doc:
raise HTTPException(404, f"Incident '{incident_id}' not found.")
call = await fstore.doc_get("calls", call_id)
if not call:
raise HTTPException(404, f"Call '{call_id}' not found.")
call_ids = list(doc.get("call_ids") or [])
if call_id not in call_ids:
call_ids.append(call_id)
await fstore.doc_update("incidents", incident_id, {
"call_ids": call_ids,
"updated_at": datetime.now(timezone.utc).isoformat(),
# A manually attached call changes what the incident is about.
"summary_stale": True,
})
# incident_ids is the canonical link — it is what the correlator writes and
# what the frontend queries with array-contains. This route only ever set
# the legacy scalar incident_id, so a manually attached call stayed
# invisible on the incident's own page.
incident_ids = list(call.get("incident_ids") or ([call["incident_id"]] if call.get("incident_id") else []))
if incident_id not in incident_ids:
incident_ids.append(incident_id)
await fstore.doc_update("calls", call_id, {
"incident_ids": incident_ids,
"incident_id": incident_id,
"corr_path": "manual",
})
return {"ok": True, "incident_ids": incident_ids}
@router.delete("/{incident_id}/calls/{call_id}")
async def unlink_call_from_incident(incident_id: str, call_id: str, _: dict = Depends(require_admin_token)):
"""
Detach a call from an incident — the other half of manual attribution.
An incident left with no calls is resolved rather than deleted, matching
what calls.py's transcript correction does when it empties one.
"""
doc = await fstore.doc_get("incidents", incident_id)
if not doc:
raise HTTPException(404, f"Incident '{incident_id}' not found.")
remaining = [c for c in (doc.get("call_ids") or []) if c != call_id]
updates: dict = {
"call_ids": remaining,
"updated_at": datetime.now(timezone.utc).isoformat(),
"summary_stale": True,
}
if not remaining:
updates["status"] = "resolved"
updates["resolved_at"] = datetime.now(timezone.utc).isoformat()
await fstore.doc_update("incidents", incident_id, updates)
call = await fstore.doc_get("calls", call_id)
if call:
incident_ids = [
i for i in (call.get("incident_ids") or ([call["incident_id"]] if call.get("incident_id") else []))
if i != incident_id
]
await fstore.doc_update("calls", call_id, {
"incident_ids": incident_ids,
"incident_id": incident_ids[0] if incident_ids else None,
})
return {"ok": True, "incident_emptied": not remaining}