Update app.py

app.py

@@ -30,7 +30,16 @@ logger = setup_logging('app')
 nltk.download('punkt')
 nltk.download('punkt_tab')
 nltk.download('stopwords')
-    
+
+# Create directories if they don't exist
+def create_dirs_if_needed():
+    """Create the necessary directories if they don't exist."""
+    os.makedirs('/tmp/data', exist_ok=True)
+    os.makedirs('/tmp/db', exist_ok=True)
+
+# Call the function at the start of your app
+create_dirs_if_needed()
+
 # Load environment variables
 load_dotenv()
 
@@ -500,12 +509,12 @@ def is_query_relevant(question, source_documents, threshold=0.1):
     except Exception as e:
         logger.error(f"Error checking query relevance: {str(e)}", exc_info=True)
         return False
-
 def get_pdf_details(filename, page_number):
     """Get details of a specific PDF page."""
     logger.info(f"Processing PDF details for file: {filename}, page: {page_number}")
     try:
-        data_path = "/home/bacancy/Documents/Company/Smart PDF Search/data"
+        # Update the paths to point to /tmp for Hugging Face Space
+        data_path = '/tmp/data'
         file_path = os.path.join(data_path, filename)
         
         # Open the PDF
@@ -565,7 +574,8 @@ def get_romanized_text(filename):
     """Get romanized text from a PDF."""
     logger.info(f"Processing romanized text for file: {filename}")
     try:
-        data_path = "/home/bacancy/Documents/Company/Smart PDF Search/data"
+        # Update the paths to point to /tmp for Hugging Face Space
+        data_path = '/tmp/data'
         file_path = os.path.join(data_path, filename)
         
         # Open the PDF
@@ -607,6 +617,9 @@ def get_romanized_text(filename):
     
 def main():
     logger.info("Starting Smart PDF Search application")
+
+    # Ensure directories are created before file processing starts
+    create_dirs_if_needed()
     
     # Detect page from query parameters
     query_params = st.query_params