Update app/main.py

app/main.py

@@ -3,7 +3,9 @@ import re
 import logging
 import traceback
 import time
+import asyncio
 from typing import List, Optional, Dict
+from concurrent.futures import ThreadPoolExecutor
 from dotenv import load_dotenv
 
 from fastapi import FastAPI, HTTPException
@@ -13,9 +15,9 @@ import uvicorn
 # Load environment variables
 load_dotenv()
 
+# Import internal services
+# Note: Ensure these modules exist in your /app directory
 from app.predictor import classifier, guide_generator, reviewer
-# Note: AIReviewerService from the first version is typically 
-# the underlying service for the 'reviewer' object in the second.
 
 # 1. Setup Logging
 logging.basicConfig(level=logging.INFO)
@@ -76,7 +78,6 @@ async def classify_file(request: FileRequest):
     try:
         result = classifier.predict(request.fileName, request.content)
         
-        # Cache embedding if repoName is provided
         if request.repoName:
             if request.repoName not in REPO_CACHE:
                 REPO_CACHE[request.repoName] = {}
@@ -90,7 +91,6 @@ async def classify_file(request: FileRequest):
         }
     except Exception as e:
         logger.error(f"Classify failed: {e}")
-        traceback.print_exc()
         raise HTTPException(status_code=500, detail=str(e))
 
 @app.post("/review-batch-code")
@@ -101,7 +101,6 @@ async def review_batch_code(request: BatchReviewRequest):
         total_files = len(reviews)
         total_vulns = sum(len(r.get("vulnerabilities", [])) for r in reviews)
         
-        # Calculate Average Maintainability
         m_scores = [r.get("metrics", {}).get("maintainability", 8) for r in reviews]
         avg_maint = sum(m_scores) / max(total_files, 1)
 
@@ -117,37 +116,55 @@ async def review_batch_code(request: BatchReviewRequest):
 
 @app.post("/repo-dashboard-stats")
 async def get_dashboard_stats(request: BatchReviewRequest):
-    """Aggregated stats for frontend dashboards including health and API sniffing."""
+    """Aggregated stats using parallel processing for speed and review-first logic."""
     try:
-        raw_reviews = reviewer.service.review_batch_code(request.files)
-        
-        # 1. Security Count
-        total_vulns = sum(len(r.get("vulnerabilities", [])) for r in raw_reviews)
-        
-        # 2. Performance/Maintainability Ratio
-        scores = [r.get("metrics", {}).get("maintainability", 8) for r in raw_reviews]
-        avg_maintainability = (sum(scores) / len(scores)) * 10 if scores else 0
-        
-        # 3. API Sniffing (Regex)
-        found_apis = []
-        for f in request.files:
-            if f.content:
-                matches = re.findall(r'(?:get|post|put|delete|patch)\([\'"]\/(.*?)[\'"]', f.content.lower())
+        # 1. Run the heavy AI review process in a thread pool to prevent blocking
+        loop = asyncio.get_event_loop()
+        with ThreadPoolExecutor() as executor:
+            raw_reviews = await loop.run_in_executor(
+                executor, reviewer.service.review_batch_code, request.files
+            )
+
+        # 2. Extract Data from the Review Results
+        total_vulns = 0
+        maint_scores = []
+        found_apis = set()
+        api_regex = re.compile(r'(?:get|post|put|delete|patch)\([\'"]\/(.*?)[\'"]', re.IGNORECASE)
+
+        for i, review in enumerate(raw_reviews):
+            # Security Count from Review
+            vulns = review.get("vulnerabilities", [])
+            total_vulns += len(vulns)
+            
+            # Maintainability Score
+            m_score = review.get("metrics", {}).get("maintainability", 8)
+            maint_scores.append(m_score)
+            
+            # API Sniffing
+            content = request.files[i].content
+            if content:
+                matches = api_regex.findall(content)
                 for match in matches:
-                    found_apis.append(f"/{match}")
+                    found_apis.add(f"/{match}")
 
-        # 4. Repo Health Calculation
-        health_score = max(10, 100 - (total_vulns * 10))
+        # 3. Aggregation & Calculations
+        num_files = len(maint_scores)
+        avg_maint = (sum(maint_scores) / num_files) if num_files > 0 else 0
+        
+        # Health Calculation: Base 100, -12 per vulnerability
+        health_score = max(10, 100 - (total_vulns * 12))
 
         return {
             "repo_health": health_score,
-            "health_label": "Excellent Health" if health_score > 80 else "Needs Review",
+            "health_label": "Excellent" if health_score > 85 else "Good" if health_score > 60 else "Critical",
             "security_issues": total_vulns,
-            "performance_ratio": f"{int(avg_maintainability)}%",
-            "exposed_apis": list(set(found_apis))[:10]
+            "performance_ratio": f"{int(avg_maint * 10)}%",
+            "exposed_apis": list(found_apis)[:12],
+            "total_files_processed": num_files
         }
     except Exception as e:
         logger.error(f"Dashboard stats failed: {e}")
+        traceback.print_exc()
         raise HTTPException(status_code=500, detail="Failed to aggregate repository stats")
 
 @app.post("/analyze-file")
@@ -195,8 +212,6 @@ async def semantic_search(request: SearchRequest):
 async def chat(request: ChatRequest):
     """RAG-based chat using provided file context."""
     start_time = time.time()
-    logger.info(f"Received Chat Request for {request.repoName}")
-    
     try:
         from app.core.model_loader import llm_engine
         
@@ -208,21 +223,12 @@ async def chat(request: ChatRequest):
         prompt = f"""
         You are "GitGud AI", an expert software architect.
         Repository: "{request.repoName}"
-
-        INSTRUCTIONS:
-        1. Use the provided CONTEXT to answer. 
-        2. If context is missing, state: "I am using general knowledge as I don't have specific snippets for this."
-        3. Use markdown for code.
-
-        CONTEXT:
-        {context_str if has_context else "[(NO CODE SNIPPETS PROVIDED)]"}
-
-        USER QUESTION:
-        {request.query}
+        CONTEXT: {context_str if has_context else "(NO CODE PROVIDED)"}
+        USER QUESTION: {request.query}
         """
         response = llm_engine.generate_text(prompt)
         
-        logger.info(f"Chat response generated in {time.time() - start_time:.2f}s")
+        logger.info(f"Chat generated in {time.time() - start_time:.2f}s")
         return {"response": response}
     except Exception as e:
         traceback.print_exc()
@@ -240,6 +246,5 @@ async def generate_guide(request: GuideRequest):
 
 # 5. Application Entry Point
 if __name__ == "__main__":
-    # Note: Using 7860 for HF Spaces compatibility, change to 8000 if preferred for local dev
     port = int(os.environ.get("PORT", 7860))
     uvicorn.run(app, host="0.0.0.0", port=port)