Update pages/DATA CATALOG.py

pages/DATA CATALOG.py

@@ -59,89 +59,90 @@ st.markdown("""
                }
     </style>
     """, unsafe_allow_html=True)
+
+def clear_cache():
+    if 'rdf' in st.session_state:
+        st.session_state.pop('rdf')
+
+def create_er_diagram(df):
+    G = nx.DiGraph()  # Directed graph
+
+    # Dictionary to hold table columns
+    table_columns = {}
+
+    # Add nodes and edges to the graph
+    for _, row in df.iterrows():
+        parent_table = row['PARENT TABLE']
+        child_table = row['CHILD TABLE']
+        parent_pk = row['PARENT TABLE RELATIONSHIP COLUMN']
+        child_fk = row['CHILD TABLE RELATIONSHIP COLUMN']
+        cardinality = row.get('CARDINALITY', '1:N')
+
+        # Add columns to tables
+        if parent_table not in table_columns:
+            table_columns[parent_table] = []
+        table_columns[parent_table].append(parent_pk)
+
+        if child_table not in table_columns:
+            table_columns[child_table] = []
+        table_columns[child_table].append(child_fk)
+
+        # Add nodes and edges
+        G.add_node(parent_table)
+        G.add_node(child_table)
+        G.add_edge(parent_table, child_table, label=f'{parent_pk} -> {child_fk}\n{cardinality}')
+
+    return G, table_columns
+
+def draw_er_diagram(G, table_columns):
+    pos = nx.spring_layout(G, k=1.5, iterations=50)  # Use a layout that spreads out nodes
+
+    plt.figure(figsize=(8, 8))
+    nx.draw(G, pos, with_labels=False, node_size=2500, node_color='lightblue', edge_color='gray', font_size=8, font_weight='bold', arrows=True)
+
+    # Draw node labels (table names in bold)
+    for node, (x, y) in pos.items():
+        plt.text(x, y + 0.13, node, fontsize=7, fontweight='bold', ha='center', va='center')
+
+    # Draw column names
+    for node, columns in table_columns.items():
+        x, y = pos[node]
+        column_text = '\n'.join(columns)
+        plt.text(x, y, column_text, fontsize=6, ha='center', va='center')
+
+    # Draw edge labels
+    edge_labels = nx.get_edge_attributes(G, 'label')
+    nx.draw_networkx_edge_labels(G, pos, edge_labels=edge_labels, font_size=6)
+    st.subheader("Schematic Representation")
+    with st.container(border=True, height= 350):
+        st.pyplot(plt)
+    img_bytes = BytesIO()
+    plt.savefig(img_bytes, format='png')
+    img_bytes.seek(0)
+    return img_bytes
+
+def cardinality(parent_df, child_df, parent_column, child_column):
+    # Check uniqueness of parent primary key
+    is_parent_unique = parent_df[parent_column].is_unique
+
+    # Check uniqueness of child foreign key
+    is_child_unique = child_df[child_column].is_unique
+
+    # Determine cardinality
+    if is_parent_unique and is_child_unique:
+        return '1:1'
+    elif is_parent_unique and not is_child_unique:
+        return '1:N'
+    elif not is_parent_unique and is_child_unique:
+        return 'N:1'
+    else:
+        return 'N:N'
+
 ######
 def main():
     # st.title('PAGE TITLE')  # Change this for each page
     sidebar()
 ########
-    def clear_cache():
-        if 'rdf' in st.session_state:
-            st.session_state.pop('rdf')
-    
-    def create_er_diagram(df):
-        G = nx.DiGraph()  # Directed graph
-    
-        # Dictionary to hold table columns
-        table_columns = {}
-    
-        # Add nodes and edges to the graph
-        for _, row in df.iterrows():
-            parent_table = row['PARENT TABLE']
-            child_table = row['CHILD TABLE']
-            parent_pk = row['PARENT TABLE RELATIONSHIP COLUMN']
-            child_fk = row['CHILD TABLE RELATIONSHIP COLUMN']
-            cardinality = row.get('CARDINALITY', '1:N')
-    
-            # Add columns to tables
-            if parent_table not in table_columns:
-                table_columns[parent_table] = []
-            table_columns[parent_table].append(parent_pk)
-    
-            if child_table not in table_columns:
-                table_columns[child_table] = []
-            table_columns[child_table].append(child_fk)
-    
-            # Add nodes and edges
-            G.add_node(parent_table)
-            G.add_node(child_table)
-            G.add_edge(parent_table, child_table, label=f'{parent_pk} -> {child_fk}\n{cardinality}')
-    
-        return G, table_columns
-    
-    def draw_er_diagram(G, table_columns):
-        pos = nx.spring_layout(G, k=1.5, iterations=50)  # Use a layout that spreads out nodes
-    
-        plt.figure(figsize=(8, 8))
-        nx.draw(G, pos, with_labels=False, node_size=2500, node_color='lightblue', edge_color='gray', font_size=8, font_weight='bold', arrows=True)
-    
-        # Draw node labels (table names in bold)
-        for node, (x, y) in pos.items():
-            plt.text(x, y + 0.13, node, fontsize=7, fontweight='bold', ha='center', va='center')
-    
-        # Draw column names
-        for node, columns in table_columns.items():
-            x, y = pos[node]
-            column_text = '\n'.join(columns)
-            plt.text(x, y, column_text, fontsize=6, ha='center', va='center')
-    
-        # Draw edge labels
-        edge_labels = nx.get_edge_attributes(G, 'label')
-        nx.draw_networkx_edge_labels(G, pos, edge_labels=edge_labels, font_size=6)
-        st.subheader("Schematic Representation")
-        with st.container(border=True, height= 350):
-            st.pyplot(plt)
-        img_bytes = BytesIO()
-        plt.savefig(img_bytes, format='png')
-        img_bytes.seek(0)
-        return img_bytes
-    
-    def cardinality(parent_df, child_df, parent_column, child_column):
-        # Check uniqueness of parent primary key
-        is_parent_unique = parent_df[parent_column].is_unique
-    
-        # Check uniqueness of child foreign key
-        is_child_unique = child_df[child_column].is_unique
-    
-        # Determine cardinality
-        if is_parent_unique and is_child_unique:
-            return '1:1'
-        elif is_parent_unique and not is_child_unique:
-            return '1:N'
-        elif not is_parent_unique and is_child_unique:
-            return 'N:1'
-        else:
-            return 'N:N'
-    
     #st.title('AUTOMATED DATA CATALOGUE')
     st.subheader('SELECT SOURCE')
     selectcol11, selectcol12 = st.columns(2)