I have made streamlit under snowflake native apps, and the streamlit is working fine locally, but while going through snowflake native apps, it is not working, like it is not able to use the desired database that I wanted.
Here is the app.py:
import streamlit as stimport pandas as pdfrom snowflake.snowpark import Sessionfrom snowflake.snowpark.context import get_active_sessiondef get_snowflake_session() -> Session: try: # Attempt to get the active session provided by the environment session = get_active_session() return session except Exception as e: st.error(f"Error getting active Snowflake session: {e}") return Nonedef show_data_page(): st.title("Snowflake Data Viewer") # Ensure credentials are stored in session_state if 'user' in st.session_state: try: session = get_snowflake_session() if session: st.success("Successfully connected to Snowflake") # Display current database and schema st.write(f"Database: {st.session_state['database']}") st.write(f"Schema: {st.session_state['schema']}") # Perform a query to get tables in the specified schema query1 = f""" USE DATABASE {st.session_state['database']};""" query2 = f""" SELECT TABLE_NAME FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'PUBLIC';""" st.write(f"Executing query: {query1}") st.write(f"Executing next query: {query2}") try: # Execute the SQL query and retrieve the result directly into a Pandas DataFrame st.write("Database Selection Query") # session.sql(query1)._execute_and_get_query_id() # Convert the result to a Pandas DataFrame st.write("Table Creation:") tables_df = session.sql(query2).to_pandas() st.write("Table:") st.write(tables_df) # tables_df = result.to_pandas() # st.write("Tables in Schema:", tables_df) # Allow the user to select a table and display its contents table_name = st.selectbox("Select a table to view data", tables_df['TABLE_NAME']) if table_name: table_query = f"SELECT * FROM {st.session_state['database']}.{st.session_state['schema']}.{table_name} LIMIT 10" st.write(f"Executing query: {table_query}") table_df = session.sql(table_query).to_pandas() st.write(f"Data in {table_name}:", table_df) except Exception as e: st.error(f"Error executing query: {e}") st.write("Debug Info: Could not retrieve table names or table data.") session.close() except Exception as e: st.error(f"Error using Snowflake session: {e}") else: st.warning("You are not logged in. Please log in first.")if __name__ == "__main__": # Streamlit sidebar inputs for Snowflake credentials st.sidebar.header("Snowflake Credentials") user = st.sidebar.text_input("User") password = st.sidebar.text_input("Password", type="password") account = st.sidebar.text_input("Account") warehouse = st.sidebar.text_input("Warehouse") database = st.sidebar.text_input("Database") schema = st.sidebar.text_input("Schema") if st.sidebar.button("Connect to Snowflake"): if user and password and account and warehouse and database and schema: st.session_state['user'] = user st.session_state['password'] = password st.session_state['account'] = account st.session_state['warehouse'] = warehouse st.session_state['database'] = database st.session_state['schema'] = schema show_data_page() else: st.warning("Please enter all Snowflake credentials") else: st.warning("Please enter your Snowflake credentials and click the button to connect.") # Show data page if connected if 'user' in st.session_state: show_data_page()But in this the query part, when I mention a database or the session database it is throwing an error as database not exists or unauthorized, but it is showing snowflake is connected. And the same query is working under snowflake worksheets.
I was trying to do a login with the credentials and accept it and show the tables from the credentials, but it is not working.
Edit: I tried to grant permissions using worksheets and it is working, have given the full details in comments section, but don't know whether it is the permanent or correct solution or not, can someone clarify it?
using this doc