added streamlit analytics for poems, connections, and voting

.gitignore

@@ -0,0 +1 @@
+.venv

app.py

@@ -0,0 +1,145 @@
+import streamlit as st
+import pandas as pd
+import psycopg2
+from psycopg2 import Error
+
+# Database connection configuration
+DB_PARAMS = {
+    "host": "aws-0-us-east-2.pooler.supabase.com",
+    "port": "5432",
+    "database": "postgres",
+    "user": "postgres.reeiesvuumiuorjgnmad",
+    "password": "S2ef4ypnmYIWYFK4"
+}
+
+def get_db_connection():
+    try:
+        connection = psycopg2.connect(**DB_PARAMS)
+        return connection
+    except Error as e:
+        st.error(f"Error connecting to database: {e}")
+        return None
+
+def fetch_data():
+    conn = get_db_connection()
+    if conn is None:
+        return None, None
+
+    try:
+        # Create a cursor object to execute PostgreSQL commands
+        cur = conn.cursor()
+
+        # Query for connections data with poem linkage
+        connections_query = """
+        SELECT c.symbol1, c.symbol2, c.reasoning, c.up_votes, c.down_votes, c.version, c.poem_id
+        FROM connections c
+        """
+        connections_df = pd.read_sql_query(connections_query, conn)
+
+        # Query for total number of poems
+        poems_query = """
+        SELECT COUNT(*) as poem_count
+        FROM poems
+        """
+        poems_df = pd.read_sql_query(poems_query, conn)
+
+        cur.close()
+        conn.close()
+        return connections_df, poems_df
+
+    except Error as e:
+        st.error(f"Error fetching data: {e}")
+        return None, None
+
+def calculate_analytics(connections_df, poems_df):
+    if connections_df is None or connections_df.empty or poems_df is None:
+        return None, None, None, None
+
+    # Total number of poems from poems table
+    total_poems = poems_df['poem_count'].iloc[0]
+
+    # Total number of connections
+    total_connections = len(connections_df)
+
+    # Calculate vote ratio
+    connections_df['vote_ratio'] = connections_df.apply(
+        lambda row: row['up_votes'] / (row['up_votes'] + row['down_votes'])
+        if (row['up_votes'] + row['down_votes']) > 0 else 0,
+        axis=1
+    )
+
+    # Group by poem_id for connections per poem
+    connections_per_poem = connections_df.groupby('poem_id').agg({
+        'symbol1': 'count',
+        'up_votes': 'sum',
+        'down_votes': 'sum',
+        'vote_ratio': 'mean'
+    }).rename(columns={'symbol1': 'connection_count'})
+
+    # Group by version
+    connections_by_version = connections_df.groupby('version').agg({
+        'up_votes': 'sum',
+        'down_votes': 'sum',
+        'vote_ratio': 'mean',
+        'symbol1': 'count'
+    }).rename(columns={'symbol1': 'connections_count'})
+
+    return total_poems, total_connections, connections_per_poem, connections_by_version
+
+def main():
+    st.title("Connections Analytics Dashboard")
+
+    # Fetch and process data
+    connections_df, poems_df = fetch_data()
+    if connections_df is None or poems_df is None:
+        st.error("Failed to load data from database")
+        return
+
+    total_poems, total_connections, connections_per_poem, connections_by_version = calculate_analytics(connections_df, poems_df)
+
+    if total_poems is None:
+        st.error("No data available to analyze")
+        return
+
+    # Display basic stats
+    st.header("Overview")
+    col1, col2 = st.columns(2)
+    with col1:
+        st.metric("Total Number of Poems", total_poems)
+    with col2:
+        st.metric("Total Number of Connections", total_connections)
+
+    # Connections per poem table
+    st.header("Connections per Poem")
+    st.dataframe(
+        connections_per_poem.reset_index(),
+        column_config={
+            "poem_id": "Poem ID",
+            "connection_count": "Number of Connections",
+            "up_votes": "Up Votes",
+            "down_votes": "Down Votes",
+            "vote_ratio": st.column_config.NumberColumn(
+                "Vote Ratio",
+                format="%.2f"
+            )
+        }
+    )
+
+    # Connections by version table
+    st.header("Connections by Version")
+    st.dataframe(
+        connections_by_version.reset_index(),
+        column_config={
+            "version": "Version",
+            "up_votes": "Up Votes",
+            "down_votes": "Down Votes",
+            "vote_ratio": st.column_config.NumberColumn(
+                "Vote Ratio",
+                format="%.2f"
+            ),
+            "connections_count": "Number of Connections"
+        }
+    )
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,3 @@
+streamlit
+pandas
+psycopg2-binary