Merge pull request 'feat(ops): דף /operations — כל מה שרץ ברקע' (#149) from worktree-ops-dashboard into main

web/app.py

@@ -6009,6 +6009,94 @@ async def digest_queue_pending(limit: int = 20):
     return {"items": items, "count": len(items)}
 
 
+# ── Operations dashboard (/operations) ────────────────────────────────────
+# One snapshot of everything running in the background that downloads or
+# analyses: the host pm2 services/crons + the DB-backed pipelines & queues.
+_COURT_FETCH_SERVICE_URL = os.environ.get(
+    "COURT_FETCH_SERVICE_URL", "http://10.0.1.1:8771"
+)
+
+
+async def _ops_pm2_services() -> dict:
+    """Proxy the host court-fetch-service /pm2 (host-only capability)."""
+    try:
+        async with httpx.AsyncClient(timeout=8.0) as client:
+            r = await client.get(f"{_COURT_FETCH_SERVICE_URL}/pm2")
+        if r.status_code == 200:
+            return {"services": r.json().get("services", []), "error": None}
+        return {"services": [], "error": f"pm2 bridge {r.status_code}"}
+    except Exception as e:  # host service down / unreachable
+        return {"services": [], "error": f"לא ניתן להגיע לשירות-המארח: {e}"}
+
+
+@app.get("/api/operations")
+async def operations_snapshot():
+    """Everything running in the background: services + pipelines/queues."""
+    pool = await db.get_pool()
+    async with pool.acquire() as conn:
+        async def counts(sql: str) -> dict:
+            return {r[0]: r[1] for r in await conn.fetch(sql)}
+
+        court_fetch = await counts(
+            "SELECT status, count(*) FROM court_fetch_jobs GROUP BY 1"
+        )
+        court_recent = [
+            dict(r) for r in await conn.fetch(
+                "SELECT case_number_norm, citation_raw, tier, status, error, "
+                "updated_at FROM court_fetch_jobs ORDER BY updated_at DESC LIMIT 15"
+            )
+        ]
+        meta = await counts(
+            "SELECT coalesce(metadata_extraction_status,'unknown'), count(*) "
+            "FROM case_law GROUP BY 1"
+        )
+        meta_queued = await conn.fetchval(
+            "SELECT count(*) FROM case_law WHERE metadata_extraction_requested_at IS NOT NULL"
+        )
+        hal_ext = await counts(
+            "SELECT coalesce(halacha_extraction_status,'unknown'), count(*) "
+            "FROM case_law GROUP BY 1"
+        )
+        hal_queued = await conn.fetchval(
+            "SELECT count(*) FROM case_law WHERE halacha_extraction_requested_at IS NOT NULL"
+        )
+        review = await counts("SELECT review_status, count(*) FROM halachot GROUP BY 1")
+        missing = await counts("SELECT status, count(*) FROM missing_precedents GROUP BY 1")
+        digests_total = await conn.fetchval("SELECT count(*) FROM digests")
+        digests_linked = await conn.fetchval(
+            "SELECT count(*) FROM digests WHERE linked_case_law_id IS NOT NULL"
+        )
+        ingested_recent = [
+            dict(r) for r in await conn.fetch(
+                "SELECT case_number, court, source_url, created_at FROM case_law "
+                "WHERE source_url LIKE '%court.gov.il%' ORDER BY created_at DESC LIMIT 12"
+            )
+        ]
+
+    pm2 = await _ops_pm2_services()
+
+    def _iso(rows: list[dict]) -> list[dict]:
+        for d in rows:
+            for k, v in list(d.items()):
+                if hasattr(v, "isoformat"):
+                    d[k] = v.isoformat()
+        return rows
+
+    return {
+        "services": pm2["services"],
+        "services_error": pm2["error"],
+        "pipelines": {
+            "court_fetch": {"by_status": court_fetch, "recent": _iso(court_recent)},
+            "metadata_extraction": {"by_status": meta, "queued": meta_queued},
+            "halacha_extraction": {"by_status": hal_ext, "queued": hal_queued},
+            "halacha_review": {"by_status": review},
+            "missing_precedents": {"by_status": missing},
+            "digests": {"total": digests_total, "linked": digests_linked},
+            "ingested_recent": _iso(ingested_recent),
+        },
+    }
+
+
 @app.get("/api/digests/{digest_id}")
 async def digest_get(digest_id: str):
     try: