Update app.py

app.py

@@ -6,6 +6,11 @@ import json
 import io
 import subprocess  # To call process_hf_dataset.py
 from database import init_chromadb, store_program, query_programs, load_chromadb_from_hf, DB_NAME
+import logging
+
+# Set up logging
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+logger = logging.getLogger(__name__)
 
 # User-configurable variables
 UPLOAD_DIR = "./uploads"  # Directory for uploads
@@ -40,6 +45,7 @@ def index():
             client = init_chromadb()
             vectors = [part['vector'] for part in parts]
             store_program(client, code_input, sequence, vectors, DB_NAME)
+            logger.info(f"Stored code: {filename}")
         elif 'code' in request.form and request.form['code'].strip():
             code_input = request.form['code']
             filename = request.form.get('filename', 'unnamed.py') or 'unnamed.py'
@@ -49,11 +55,13 @@ def index():
             vectors = [part['vector'] for part in parts]
             client = init_chromadb()
             store_program(client, code_input, sequence, vectors, DB_NAME)
+            logger.info(f"Stored code: {filename}")
         elif 'query_ops' in request.form and request.form['query_ops'].strip():
             # Handle query for operations (category sequence)
             operations = [op.strip() for op in request.form['query_ops'].split(',')]
             client = load_chromadb_from_hf()
             query_results = query_programs(client, operations, DB_NAME)
+            logger.info(f"Queried operations: {operations}")
             return render_template(
                 'results_partial.html',
                 parts=None,
@@ -67,6 +75,7 @@ def index():
             semantic_query = request.form['semantic_query']
             client = load_chromadb_from_hf()
             query_results = query_programs(client, None, DB_NAME, semantic_query=semantic_query)
+            logger.info(f"Queried semantically: {semantic_query}")
             return render_template(
                 'results_partial.html',
                 parts=None,
@@ -82,12 +91,15 @@ def index():
                 client = init_chromadb()
                 try:
                     client.delete_collection(DB_NAME)
-                except:
-                    pass  # Collection may not exist
+                    logger.info(f"Deleted ChromaDB collection: {DB_NAME}")
+                except Exception as e:
+                    logger.warning(f"Failed to delete collection {DB_NAME}: {e}")
                 client.create_collection(DB_NAME)
+                logger.info(f"Created fresh ChromaDB collection: {DB_NAME}")
                 
                 # Process dataset
-                subprocess.run(['python', 'process_hf_dataset.py'], check=True)
+                result = subprocess.run(['python', 'process_hf_dataset.py'], check=True, capture_output=True, text=True)
+                logger.info(f"Process Hugging Face dataset output: {result.stdout}")
                 return render_template(
                     'results_partial.html',
                     parts=None,
@@ -98,11 +110,19 @@ def index():
                     message="Hugging Face dataset processed and stored successfully with fresh database."
                 )
             except subprocess.CalledProcessError as e:
-                return f"Error processing Hugging Face dataset: {e}", 500
+                logger.error(f"Error processing Hugging Face dataset: {e.stderr}")
+                return f"Error processing Hugging Face dataset: {e.stderr}", 500
         elif 'load_dataset' in request.form:
             # Trigger loading of Hugging Face dataset without resetting
             try:
-                subprocess.run(['python', 'process_hf_dataset.py'], check=True)
+                # Check if collection exists, get or create if needed
+                client = init_chromadb()
+                collection = client.get_or_create_collection(DB_NAME)
+                logger.info(f"Using existing or new ChromaDB collection: {DB_NAME}")
+                
+                # Process dataset
+                result = subprocess.run(['python', 'process_hf_dataset.py'], check=True, capture_output=True, text=True)
+                logger.info(f"Load Hugging Face dataset output: {result.stdout}")
                 return render_template(
                     'results_partial.html',
                     parts=None,
@@ -113,16 +133,19 @@ def index():
                     message="Hugging Face dataset loaded and stored successfully."
                 )
             except subprocess.CalledProcessError as e:
-                return f"Error loading Hugging Face dataset: {e}", 500
+                logger.error(f"Error loading Hugging Face dataset: {e.stderr}")
+                return f"Error loading Hugging Face dataset: {e.stderr}", 500
         elif 'reset_db' in request.form:
             # Reset ChromaDB collection (no repopulation with samples)
             try:
                 client = init_chromadb()
                 try:
                     client.delete_collection(DB_NAME)
-                except:
-                    pass  # Collection may not exist
+                    logger.info(f"Deleted ChromaDB collection: {DB_NAME}")
+                except Exception as e:
+                    logger.warning(f"Failed to delete collection {DB_NAME}: {e}")
                 client.create_collection(DB_NAME)
+                logger.info(f"Created fresh ChromaDB collection: {DB_NAME}")
                 return render_template(
                     'results_partial.html',
                     parts=None,
@@ -133,6 +156,7 @@ def index():
                     message="Database reset successfully."
                 )
             except Exception as e:
+                logger.error(f"Error resetting database: {e}")
                 return f"Error resetting database: {e}", 500
 
         if parts:
@@ -148,7 +172,8 @@ def index():
             )
         return 'No file, code, or query provided', 400
 
-    # Initial page load (do not populate with samples automatically)
+    # Initial page load (start empty, no default population)
+    logger.info("Application started, database empty until triggered by buttons")
     return render_template('index.html', parts=None, filename=None, reconstructed_code=None, code_input=None, query_results=None)
 
 @app.route('/export_json', methods=['POST'])