Update app.py

app.py

@@ -17,6 +17,10 @@ import pytz
 import re
 import pandas as pd
 import base64
+import io
+import openpyxl
+from openpyxl.utils.dataframe import dataframe_to_rows
+from openpyxl.worksheet.hyperlink import Hyperlink
 
 # --- Constants ---
 MELBOURNE_TIMEZONE = 'Australia/Melbourne'
@@ -213,6 +217,50 @@ def get_table_download_link(df, filename="citation_data.csv"):
     href = f'<a href="data:file/csv;base64,{b64}" download="{filename}">Download Citation Data as CSV</a>'
     return href
 
+def get_excel_download_link(df, filename="citation_data.xlsx"):
+    output = io.BytesIO()
+    wb = openpyxl.Workbook()
+    ws = wb.active
+    # Write headers
+    headers = df.columns.tolist()
+    ws.append(headers)
+
+    # Write data rows
+    for index, row in df.iterrows():
+        row_data = []
+        cell_positions = {}  # track cell positions for hyperlink assignment
+        urls = {}            # store URLs per column
+
+        for col_idx, col in enumerate(headers):
+            value = row[col]
+            if col in ["Citation", "SCC Index"]:
+                # Extract URL and display text from HTML anchor tag
+                match = re.search(r'<a href="([^"]+)"[^>]*>([^<]+)</a>', str(value))
+                if match:
+                    link_url, display_text = match.groups()
+                    row_data.append(display_text)
+                    # Position where this cell will be written (next row after append)
+                    cell_positions[col] = (ws.max_row + 1, col_idx + 1)
+                    urls[col] = link_url
+                else:
+                    row_data.append(value)
+            else:
+                row_data.append(value)
+
+        ws.append(row_data)
+
+        # Apply hyperlinks after appending row
+        for col, (r, c) in cell_positions.items():
+            cell = ws.cell(row=r, column=c)
+            cell.hyperlink = urls[col]
+            cell.hyperlink.tooltip = "Click to visit source"
+            cell.style = "Hyperlink"
+
+    wb.save(output)
+    b64 = base64.b64encode(output.getvalue()).decode()
+    href = f'<a href="data:application/vnd.openxmlformats-officedocument.spreadsheetml.sheet;base64,{b64}" download="{filename}">Download citation data as Excel</a>'
+    return href
+
 # --- Live Clock JavaScript ---
 def live_clock():
     return """
@@ -413,27 +461,62 @@ with tabs[0]:
             citation_link_end = f'<a href="{source_url}#:~:text={encode_text_fragment(selected_fragment)}" data-hash="{scc_hash}">({author_name}, {publication_year})</a>'
             metadata_link = format_metadata_html(source_url, author_name, publication_year, scc_hash, username, task_name, current_date, current_time)
 
-            col_html1, col_html2 = st.columns(2)
+            # --- Persistent Table with Clickable SCC Hash ---
 
+            # First, ensure session state is initialized for the citation DataFrame
+            if 'citation_df' not in st.session_state:
+                st.session_state.citation_df = pd.DataFrame(columns=[
+                    "Username", "Task Name", "Time", "Date",
+                    "Citation", "SCC Index", "Annotated Text"
+                ])
+    
+            # Create clickable HTML for SCC Index (full metadata link)
+            clickable_index = metadata_link
+    
+            # Create new row data
+            new_row = {
+                "Username": username,
+                "Task Name": task_name,
+                "Time": current_time,
+                "Date": current_date,
+                "Citation": citation_link_start,
+                "SCC Index": clickable_index,
+                "Annotated Text": annotated_text
+            }
+    
+            # Append the new row to the session state DataFrame
+            new_df = pd.DataFrame([new_row])
+            st.session_state.citation_df = pd.concat([st.session_state.citation_df, new_df], ignore_index=True)
+    
+            # Get the accumulated DataFrame for display and download
+            df = st.session_state.citation_df
+    
+            col_html1, col_html2 = st.columns(2)
+    
             # HTML Citation - Start of Text
             with col_html1:
                 st.markdown("### Citation (Start of Text)")
                 st.markdown('<div class="rendered-citation">', unsafe_allow_html=True)
                 st.markdown(citation_link_start, unsafe_allow_html=True)
                 st.markdown('</div>', unsafe_allow_html=True)
-
+    
             # HTML Citation - End of Text
             with col_html2:
                 st.markdown("### Citation (End of Text)")
                 st.markdown('<div class="rendered-citation">', unsafe_allow_html=True)
                 st.markdown(citation_link_end, unsafe_allow_html=True)
                 st.markdown('</div>', unsafe_allow_html=True)
-
+    
             # SCC Index
             st.markdown("### SCC Index")
             st.markdown(metadata_link, unsafe_allow_html=True)
     
-    st.markdown('</div>', unsafe_allow_html=True)
+            # Display table after SCC Index
+            st.markdown("### Citation Table")
+            st.markdown(get_excel_download_link(df), unsafe_allow_html=True)
+            st.markdown(df.to_html(classes="citation-table", index=False, escape=False), unsafe_allow_html=True)
+    
+        st.markdown('</div>', unsafe_allow_html=True)
 
 with tabs[1]:
     st.markdown('<div class="tab-content">', unsafe_allow_html=True)