Update app.py

app.py

@@ -21,7 +21,7 @@ for module_name, pip_name in REQUIRED_PACKAGES.items():
         print(f"✅ {pip_name} установлен!")
 
 # ============================================
-# 👑 TOMIRIS SPACE 29 v2.1 — MACRO SURPRISE ENGINE (Hub-Connected)
+# 👑 TOMIRIS SPACE 29 v2.2 — MACRO SURPRISE ENGINE (АВТО-ОТПРАВКА В HUB)
 # ============================================
 import os, time, json, logging, asyncio
 from typing import Dict, Any, List, Optional
@@ -41,14 +41,28 @@ HUB_URL = os.getenv("SPACE17_URL", "https://tomiris-ai-name5-5.hf.space")
 FRED_KEY = os.getenv("FRED_KEY", "faa11c8e2e4beee08c5b966e8b63a513")
 NEWSAPI_KEY = os.getenv("NEWSAPI_KEY", "948c7816beea47baa23b054592472d0e")
 
+# Интервал авто-отправки (секунды)
+AUTO_SEND_INTERVAL = int(os.getenv("AUTO_SEND_INTERVAL", "3600"))  # раз в час — макро-данные медленные
+
 # Консенсус-прогнозы (обновлять ежемесячно)
 CONSENSUS = {
-    "CPI_YOY": 3.2, "CORE_CPI_YOY": 3.5, "UNEMPLOYMENT": 4.0, "NFP": 180000,
-    "GDP_QOQ": 2.0, "FED_RATE": 4.25, "ISM_MANUF": 49.0, "ISM_SERVICES": 52.0,
-    "RETAIL_SALES": 0.3, "DURABLE_GOODS": 0.5
+    "CPI_YOY": 3.2,
+    "CORE_CPI_YOY": 3.5,
+    "UNEMPLOYMENT": 4.0,
+    "NFP": 180000,
+    "GDP_QOQ": 2.0,
+    "FED_RATE": 4.25,
+    "ISM_MANUF": 49.0,
+    "ISM_SERVICES": 52.0,
+    "RETAIL_SALES": 0.3,
+    "DURABLE_GOODS": 0.5
+}
+
+CACHE_TTL = {
+    "fred": 3600,
+    "news": 1800
 }
 
-CACHE_TTL = {"fred": 3600, "news": 1800}
 HISTORY_FILE = "surprise_history.json"
 
 # ================= HTTP КЛИЕНТ =================
@@ -77,22 +91,30 @@ async def fetch_fred_series(series_id: str, months: int = 13) -> List[Dict]:
     cache_key = f"fred_{series_id}_{months}"
     if cache_key in cache_store and time.time() - cache_times.get(cache_key, 0) < CACHE_TTL["fred"]:
         return cache_store[cache_key]
+
     try:
         r = await http_client.get(
-            f"https://api.stlouisfed.org/fred/series/observations?series_id={series_id}&api_key={FRED_KEY}&file_type=json&sort_order=desc&limit={months}")
+            f"https://api.stlouisfed.org/fred/series/observations?series_id={series_id}&api_key={FRED_KEY}&file_type=json&sort_order=desc&limit={months}"
+        )
         if r.status_code == 200:
             data = r.json()
-            values = [{'date': obs['date'], 'value': float(obs['value'])} for obs in data.get('observations', []) if obs['value'] != '.']
+            values = [
+                {'date': obs['date'], 'value': float(obs['value'])}
+                for obs in data.get('observations', [])
+                if obs['value'] != '.'
+            ]
             cache_store[cache_key] = values
             cache_times[cache_key] = time.time()
             return values
     except Exception as e:
         logger.warning(f"FRED {series_id}: {e}")
+
     return []
 
 def get_yoy_change(data: List[Dict], current_month: str) -> Optional[float]:
     current_val = None
     prev_val = None
+
     for item in data:
         date = item['date']
         if date == current_month:
@@ -100,15 +122,24 @@ def get_yoy_change(data: List[Dict], current_month: str) -> Optional[float]:
         year_ago = str(int(date[:4]) - 1) + date[4:]
         if date == year_ago and date[:7] == current_month[:7]:
             prev_val = item['value']
+
     if current_val and prev_val:
         return ((current_val - prev_val) / prev_val) * 100
+
     return None
 
 # ================= АНАЛИЗ СЮРПРИЗА =================
 def calc_surprise(actual: float, consensus: float) -> Dict:
     if consensus == 0:
-        return {"surprise_pct": 0, "level": "IN_LINE", "impact": 0, "direction": "NEUTRAL"}
+        return {
+            "surprise_pct": 0,
+            "level": "IN_LINE",
+            "impact": 0,
+            "direction": "NEUTRAL"
+        }
+
     surprise_pct = ((actual - consensus) / abs(consensus)) * 100
+
     if abs(surprise_pct) > 100:
         level, impact = "EXTREME_SURPRISE", 30
     elif abs(surprise_pct) > 50:
@@ -119,33 +150,73 @@ def calc_surprise(actual: float, consensus: float) -> Dict:
         level, impact = "MINOR_SURPRISE", 5
     else:
         level, impact = "IN_LINE", 0
+
     direction = "POSITIVE" if surprise_pct > 0 else "NEGATIVE" if surprise_pct < 0 else "NEUTRAL"
-    return {"surprise_pct": round(surprise_pct, 2), "level": level, "impact": impact, "direction": direction}
+
+    return {
+        "surprise_pct": round(surprise_pct, 2),
+        "level": level,
+        "impact": impact,
+        "direction": direction
+    }
 
 # ================= FOMC SURPRISE (через новости) =================
 async def fetch_fomc_surprise() -> Dict:
     if not NEWSAPI_KEY:
-        return {"indicator": "FOMC", "impact": 0, "direction": "NEUTRAL"}
+        return {
+            "indicator": "FOMC",
+            "impact": 0,
+            "direction": "NEUTRAL"
+        }
+
     try:
         r = await http_client.get(
-            f"https://newsapi.org/v2/everything?q=fomc+fed+rate+decision&pageSize=10&apiKey={NEWSAPI_KEY}")
+            f"https://newsapi.org/v2/everything?q=fomc+fed+rate+decision&pageSize=10&apiKey={NEWSAPI_KEY}"
+        )
         if r.status_code == 200:
             articles = r.json().get('articles', [])
-            hawk = sum(1 for a in articles if any(w in (a.get('title','')+a.get('description','')).lower() for w in ['hawkish','raise','tighten','surprise hike']))
-            dove = sum(1 for a in articles if any(w in (a.get('title','')+a.get('description','')).lower() for w in ['dovish','cut','ease','surprise cut']))
-            if hawk > dove*2:
-                return {"indicator": "FOMC", "impact": -15, "direction": "HAWKISH", "hawkish": hawk, "dovish": dove}
-            elif dove > hawk*2:
-                return {"indicator": "FOMC", "impact": 15, "direction": "DOVISH", "hawkish": hawk, "dovish": dove}
+            hawk = sum(
+                1 for a in articles
+                if any(w in (a.get('title', '') + a.get('description', '')).lower()
+                       for w in ['hawkish', 'raise', 'tighten', 'surprise hike'])
+            )
+            dove = sum(
+                1 for a in articles
+                if any(w in (a.get('title', '') + a.get('description', '')).lower()
+                       for w in ['dovish', 'cut', 'ease', 'surprise cut'])
+            )
+
+            if hawk > dove * 2:
+                return {
+                    "indicator": "FOMC",
+                    "impact": -15,
+                    "direction": "HAWKISH",
+                    "hawkish": hawk,
+                    "dovish": dove
+                }
+            elif dove > hawk * 2:
+                return {
+                    "indicator": "FOMC",
+                    "impact": 15,
+                    "direction": "DOVISH",
+                    "hawkish": hawk,
+                    "dovish": dove
+                }
     except:
         pass
-    return {"indicator": "FOMC", "impact": 0, "direction": "NEUTRAL"}
+
+    return {
+        "indicator": "FOMC",
+        "impact": 0,
+        "direction": "NEUTRAL"
+    }
 
 # ================= ГЛАВНЫЙ АНАЛИЗ =================
 async def analyze_macro_surprises() -> Dict:
     today = datetime.now(timezone.utc)
     current_month_str = today.strftime("%Y-%m")
 
+    # Загружаем все данные параллельно
     cpi_data = await fetch_fred_series("CPIAUCSL", 13)
     core_cpi_data = await fetch_fred_series("CPILFESL", 13)
     unemp_data = await fetch_fred_series("UNRATE", 6)
@@ -158,51 +229,95 @@ async def analyze_macro_surprises() -> Dict:
 
     surprises = []
     total_score = 0.0
-    weights = {"CPI_YOY": 0.30, "NFP": 0.25, "FOMC": 0.20, "ISM_MANUF": 0.10, "GDP_QOQ": 0.10, "RETAIL_SALES": 0.05}
+    weights = {
+        "CPI_YOY": 0.30,
+        "NFP": 0.25,
+        "FOMC": 0.20,
+        "ISM_MANUF": 0.10,
+        "GDP_QOQ": 0.10,
+        "RETAIL_SALES": 0.05
+    }
 
+    # CPI YoY
     cpi_yoy = get_yoy_change(cpi_data, current_month_str)
     if cpi_yoy is not None:
         s = calc_surprise(cpi_yoy, CONSENSUS["CPI_YOY"])
         impact = s['impact'] * (1 if s['direction'] == 'NEGATIVE' else -0.5)
         total_score += impact * weights["CPI_YOY"]
-        surprises.append({"indicator": "CPI_YOY", "actual": round(cpi_yoy, 2), "consensus": CONSENSUS["CPI_YOY"], "surprise": s})
+        surprises.append({
+            "indicator": "CPI_YOY",
+            "actual": round(cpi_yoy, 2),
+            "consensus": CONSENSUS["CPI_YOY"],
+            "surprise": s
+        })
 
+    # NFP
     if nfp_data:
         nfp_actual = nfp_data[0]['value']
         s = calc_surprise(nfp_actual, CONSENSUS["NFP"])
         impact = s['impact'] * (1 if s['direction'] == 'POSITIVE' else -1)
         total_score += impact * weights["NFP"]
-        surprises.append({"indicator": "NFP", "actual": int(nfp_actual), "consensus": CONSENSUS["NFP"], "surprise": s})
+        surprises.append({
+            "indicator": "NFP",
+            "actual": int(nfp_actual),
+            "consensus": CONSENSUS["NFP"],
+            "surprise": s
+        })
 
+    # GDP QoQ
     if gdp_data:
         gdp_actual = gdp_data[0]['value']
         s = calc_surprise(gdp_actual, CONSENSUS["GDP_QOQ"])
         impact = s['impact'] * (1 if s['direction'] == 'POSITIVE' else -1)
         total_score += impact * weights["GDP_QOQ"]
-        surprises.append({"indicator": "GDP_QOQ", "actual": round(gdp_actual, 2), "consensus": CONSENSUS["GDP_QOQ"], "surprise": s})
+        surprises.append({
+            "indicator": "GDP_QOQ",
+            "actual": round(gdp_actual, 2),
+            "consensus": CONSENSUS["GDP_QOQ"],
+            "surprise": s
+        })
 
+    # ISM Manufacturing
     if ism_data:
         ism_actual = ism_data[0]['value']
         s = calc_surprise(ism_actual, CONSENSUS["ISM_MANUF"])
         impact = s['impact'] * (1 if s['direction'] == 'POSITIVE' else -1)
         total_score += impact * weights["ISM_MANUF"]
-        surprises.append({"indicator": "ISM_MANUF", "actual": round(ism_actual, 2), "consensus": CONSENSUS["ISM_MANUF"], "surprise": s})
+        surprises.append({
+            "indicator": "ISM_MANUF",
+            "actual": round(ism_actual, 2),
+            "consensus": CONSENSUS["ISM_MANUF"],
+            "surprise": s
+        })
 
+    # Retail Sales
     if retail_data:
         retail_actual = retail_data[0]['value']
         s = calc_surprise(retail_actual, CONSENSUS["RETAIL_SALES"])
         impact = s['impact'] * (1 if s['direction'] == 'POSITIVE' else -1)
         total_score += impact * weights["RETAIL_SALES"]
-        surprises.append({"indicator": "RETAIL_SALES", "actual": round(retail_actual, 2), "consensus": CONSENSUS["RETAIL_SALES"], "surprise": s})
+        surprises.append({
+            "indicator": "RETAIL_SALES",
+            "actual": round(retail_actual, 2),
+            "consensus": CONSENSUS["RETAIL_SALES"],
+            "surprise": s
+        })
 
+    # FOMC Surprise
     if fomc_surprise['impact'] != 0:
         total_score += fomc_surprise['impact'] * weights["FOMC"]
-        surprises.append({"indicator": "FOMC", "signal": fomc_surprise['direction'], "impact": fomc_surprise['impact']})
+        surprises.append({
+            "indicator": "FOMC",
+            "signal": fomc_surprise['direction'],
+            "impact": fomc_surprise['impact']
+        })
 
+    # Нормализация
     total_score = max(-50, min(50, total_score))
     surprise_index = 50 + total_score
     surprise_index = max(0, min(100, surprise_index))
 
+    # Определение режима
     if surprise_index > 65:
         regime, direction = "RISK_ON", "LONG"
         confidence = surprise_index / 100
@@ -213,6 +328,7 @@ async def analyze_macro_surprises() -> Dict:
         regime, direction = "NEUTRAL", "WAIT"
         confidence = 0.0
 
+    # Сохраняем в историю
     SURPRISE_HISTORY.append({
         "timestamp": today.isoformat(),
         "surprise_index": round(surprise_index, 2),
@@ -232,15 +348,19 @@ async def analyze_macro_surprises() -> Dict:
 
 # ================= ОТПРАВКА В HUB =================
 async def send_signal_to_hub(symbol: str, direction: str, confidence: float):
+    """Отправка сигнала в Space 17 (Data Hub)."""
     try:
-        await http_client.post(f"{HUB_URL}/signal", json={
+        resp = await http_client.post(f"{HUB_URL}/signal", json={
             "space": "space_29_macro_surprise",
             "symbol": symbol,
             "direction": direction,
             "confidence": confidence,
             "raw": json.dumps({"source": "space_29_macro_surprise"})
-        })
-        logger.info(f"📤 {symbol}: {direction} conf={confidence:.3f} отправлен в Hub")
+        }, timeout=10)
+        if resp.status_code == 200:
+            logger.info(f"📤 {symbol}: {direction} conf={confidence:.3f} отправлен в Hub")
+        else:
+            logger.warning(f"Hub вернул {resp.status_code}: {resp.text[:100]}")
     except Exception as e:
         logger.error(f"Ошибка отправки в Hub: {e}")
 
@@ -257,7 +377,13 @@ async def get_macro_surprise_signal() -> Dict[str, Any]:
     result = {
         "space": "space_29_macro_surprise",
         "timestamp": int(time.time()),
-        "signals": {sym: {"direction": analysis['direction'], "confidence": analysis['confidence']} for sym in SYMBOLS},
+        "signals": {
+            sym: {
+                "direction": analysis['direction'],
+                "confidence": analysis['confidence']
+            }
+            for sym in SYMBOLS
+        },
         "surprise_analysis": analysis,
         "latency_ms": latency
     }
@@ -265,19 +391,45 @@ async def get_macro_surprise_signal() -> Dict[str, Any]:
     logger.info(f"📈 Macro Surprise: Index={analysis['surprise_index']:.1f} Regime={analysis['market_regime']}")
     return result
 
+# ================= АВТО-ОТПРАВКА ПО ТАЙМЕРУ =================
+async def auto_send_loop():
+    """🔥 Фоновая задача: каждый час анализирует макро-сюрпризы и шлёт сигналы в Hub."""
+    logger.info(f"🔄 Авто-отправка Macro Surprise запущена (интервал {AUTO_SEND_INTERVAL}с)")
+    # Первый запуск через 30 секунд после старта
+    await asyncio.sleep(30)
+    while True:
+        try:
+            logger.info("📈 Macro Surprise авто-анализ...")
+            await get_macro_surprise_signal()
+            logger.info("✅ Macro Surprise авто-отправка завершена")
+        except Exception as e:
+            logger.error(f"Ошибка в авто-отправке: {e}")
+        
+        await asyncio.sleep(AUTO_SEND_INTERVAL)
+
 # ================= FASTAPI =================
-app = FastAPI(title="Tomiris Space 29 v2.1 — Macro Surprise Engine (Hub)")
+app = FastAPI(title="Tomiris Space 29 v2.2 — Macro Surprise Engine (Auto-Hub)")
 
 @app.on_event("startup")
-async def startup(): pass
+async def startup():
+    # Запускаем ф��новую авто-отправку
+    asyncio.create_task(auto_send_loop())
+    logger.info("🚀 Space 29 v2.2 запущен с авто-отправкой в Hub")
 
 @app.on_event("shutdown")
-async def shutdown(): await http_client.aclose()
+async def shutdown():
+    await http_client.aclose()
 
 @app.get("/health")
 async def health():
-    return {"status": "operational", "version": "2.1", "hub_connected": True,
-            "indicators": list(CONSENSUS.keys()), "history_length": len(SURPRISE_HISTORY)}
+    return {
+        "status": "operational",
+        "version": "2.2",
+        "hub_url": HUB_URL,
+        "auto_send_interval": AUTO_SEND_INTERVAL,
+        "indicators": list(CONSENSUS.keys()),
+        "history_length": len(SURPRISE_HISTORY)
+    }
 
 @app.get("/consilium")
 async def consilium():
@@ -295,8 +447,13 @@ async def fomc():
 async def history(limit: int = 50):
     return list(SURPRISE_HISTORY)[-limit:]
 
+@app.get("/send_now")
+async def send_now():
+    """Ручной триггер отправки."""
+    return await get_macro_surprise_signal()
+
 if __name__ == "__main__":
     import uvicorn
     uvicorn.run(app, host="0.0.0.0", port=7860)
 
-print("🚀 SPACE 29 v2.1 — MACRO SURPRISE ENGINE (Hub-Connected) ЗАПУЩЕН!")
+print("🚀 SPACE 29 v2.2 — MACRO SURPRISE ENGINE (АВТО-ОТПРАВКА В HUB) ЗАПУЩЕН!")