Update context_engine.py

context_engine.py

@@ -1,35 +1,35 @@
-"""
-Nexari Context Engine (UPDATED)
-Author: Piyush
-Improvements:
- - Robust emotion pipeline usage & error handling
- - Safer fallback when model not available
- - Returns compact psychological profile instruction
-"""
+# context_engine.py - UPDATED (replace original)
+from typing import Tuple
+import threading
 
-from transformers import pipeline
+_emotion_classifier = None
+_emotion_lock = threading.Lock()
 
-print(">>> Context: Loading Emotion Analysis Model...")
-# load with top_k=1 by default
-try:
-    emotion_classifier = pipeline("text-classification", model="j-hartmann/emotion-english-distilroberta-base", top_k=1)
-except Exception as e:
-    print(f"Context: Failed to load emotion model: {e}")
-    emotion_classifier = None
+def _load_emotion_model():
+    global _emotion_classifier
+    try:
+        from transformers import pipeline
+        _emotion_classifier = pipeline("text-classification", model="j-hartmann/emotion-english-distilroberta-base", top_k=1)
+    except Exception as e:
+        print(f"Context: Failed to load emotion model: {e}")
+        _emotion_classifier = None
 
-def _safe_emotion_analysis(text):
-    if not emotion_classifier:
-        return ("neutral", 0.0)
+def _ensure_emotion_loaded():
+    if _emotion_classifier is None:
+        with _emotion_lock:
+            if _emotion_classifier is None:
+                _load_emotion_model()
+
+def _safe_emotion_analysis(text: str) -> Tuple[str, float]:
     try:
-        # pipeline may return list of dicts or list-of-lists depending on version
-        res = emotion_classifier(text)
+        _ensure_emotion_loaded()
+        if not _emotion_classifier:
+            return ("neutral", 0.0)
+        res = _emotion_classifier(text)
         if isinstance(res, list) and len(res) > 0:
-            # res could be [{'label':..., 'score':...}] or [['label',score],...]
             first = res[0]
             if isinstance(first, dict):
                 return (first.get('label', 'neutral'), float(first.get('score', 0.0)))
-            elif isinstance(first, list) and len(first) > 0 and isinstance(first[0], dict):
-                return (first[0].get('label', 'neutral'), float(first[0].get('score', 0.0)))
         return ("neutral", 0.0)
     except Exception as e:
         print(f"Emotion analysis error: {e}")
@@ -40,11 +40,10 @@ def get_smart_context(user_text):
     Analyzes the user's 'Vibe' and returns a short persona instruction.
     """
     try:
-        label, confidence = _safe_emotion_analysis(user_text)
-        top_emotion = label.lower()
-        confidence = float(confidence)
-
-        word_count = len(user_text.split()) if user_text else 0
+        label, confidence = _safe_emotion_analysis(user_text or "")
+        top_emotion = label.lower() if label else "neutral"
+        confidence = float(confidence or 0.0)
+        word_count = len((user_text or "").split())
 
         if word_count < 4:
             conversation_mode = "Ping-Pong Mode (Fast)"