more logs

src/legisqa_local/config/settings.py

@@ -74,6 +74,8 @@ def setup_chromadb():
             # Update environment variable to point to persistent storage
             os.environ["CHROMA_PERSIST_DIRECTORY"] = persistent_chroma_path
             logger.info(f"Updated CHROMA_PERSIST_DIRECTORY to: {persistent_chroma_path}")
+            # Inspect the existing ChromaDB
+            inspect_chromadb(persistent_chroma_path)
             return persistent_chroma_path
         
         # ChromaDB not found in persistent storage, try to download from HF Dataset
@@ -92,6 +94,8 @@ def setup_chromadb():
                 # Update environment variable to point to persistent storage
                 os.environ["CHROMA_PERSIST_DIRECTORY"] = persistent_chroma_path
                 logger.info(f"✅ ChromaDB download successful! Updated path to: {persistent_chroma_path}")
+                # Inspect the downloaded ChromaDB
+                inspect_chromadb(persistent_chroma_path)
                 return persistent_chroma_path
             else:
                 logger.error("❌ ChromaDB download from HF Dataset failed!")
@@ -109,6 +113,8 @@ def setup_chromadb():
     if os.path.exists(chroma_path):
         if os.listdir(chroma_path):
             logger.info(f"✅ ChromaDB found at {chroma_path}")
+            # Inspect the fallback ChromaDB
+            inspect_chromadb(chroma_path)
             return chroma_path
         else:
             logger.warning(f"ChromaDB directory exists but is empty: {chroma_path}")
@@ -172,6 +178,67 @@ def download_chromadb_from_hf_dataset(dataset_repo: str, local_path: str) -> boo
         logger.error(f"Exception type: {type(e).__name__}")
         return False
 
+def inspect_chromadb(chroma_path: str):
+    """Inspect ChromaDB collection to verify it's working correctly"""
+    logger.info("🔍 === ChromaDB Collection Inspection ===")
+    
+    try:
+        from chromadb import PersistentClient
+        
+        # Get collection name from config
+        chroma_config = get_chroma_config()
+        collection_name = chroma_config["collection_name"]
+        
+        logger.info(f"📋 Collection name: {collection_name}")
+        logger.info(f"📁 ChromaDB path: {chroma_path}")
+        
+        # Create ChromaDB client
+        client = PersistentClient(path=chroma_path)
+        logger.info("✅ ChromaDB client created successfully")
+        
+        # List all collections
+        collections = client.list_collections()
+        logger.info(f"📚 Available collections: {[c.name for c in collections]}")
+        
+        # Get the specific collection
+        if collection_name in [c.name for c in collections]:
+            collection = client.get_collection(name=collection_name)
+            logger.info(f"✅ Collection '{collection_name}' found")
+            
+            # Get collection count
+            count = collection.count()
+            logger.info(f"📊 Collection count: {count} items")
+            
+            if count > 0:
+                # Get a sample item
+                logger.info("🔍 Fetching sample items...")
+                sample = collection.get(limit=3, include=["documents", "metadatas", "ids"])
+                
+                logger.info(f"📝 Sample IDs: {sample['ids']}")
+                
+                if sample['documents']:
+                    logger.info(f"📄 Sample document (first 200 chars):")
+                    logger.info(f"   {sample['documents'][0][:200]}...")
+                
+                if sample['metadatas']:
+                    logger.info(f"🏷️  Sample metadata:")
+                    for i, metadata in enumerate(sample['metadatas'][:2]):
+                        logger.info(f"   Item {i}: {metadata}")
+            else:
+                logger.warning("⚠️  Collection is empty!")
+                
+        else:
+            logger.error(f"❌ Collection '{collection_name}' not found!")
+            logger.error(f"Available collections: {[c.name for c in collections]}")
+            
+    except ImportError as e:
+        logger.error(f"❌ Import error during ChromaDB inspection: {e}")
+    except Exception as e:
+        logger.error(f"❌ Error inspecting ChromaDB: {e}")
+        logger.error(f"Exception type: {type(e).__name__}")
+    
+    logger.info("🔍 === ChromaDB Inspection Complete ===")
+
 # Embedding model configuration
 EMBEDDING_MODEL = "sentence-transformers/static-retrieval-mrl-en-v1"
 EMBEDDING_DEVICE = "cpu"