Update app.py

app.py

@@ -5,6 +5,7 @@ import urllib.parse
 from datetime import datetime
 import pytz
 import pandas as pd
+import re
 
 # --- Constants ---
 MELBOURNE_TIMEZONE = 'Australia/Melbourne'
@@ -115,6 +116,26 @@ def format_metadata_html(url, author, year, scc_hash, username, task_name, curre
 def check_for_fragment(url):
     return '#:~:text=' in url
 
+def parse_citation(citation_html):
+    # Parse citation HTML to extract author, year, URL, and fragment text
+    match = re.match(r'<a href="([^"]+)#:~:text=([^"]+)" data-hash="[^"]+">([^<]+) \((\d{4})\)</a>', citation_html)
+    if match:
+        url, encoded_fragment, author, year = match.groups()
+        fragment_text = urllib.parse.unquote(encoded_fragment)
+        return author, year, url, fragment_text
+    return None, None, None, None
+
+def parse_metadata_hash(metadata_html):
+    # Parse metadata HTML to extract hash, username, task_name, date, time
+    match = re.match(r'<a href="([^"]+)#:~:text=([^"]+)" data-hash="([^"]+)">[^<]+\(\d{4}\)\. ([^<]+)</a>', metadata_html)
+    if match:
+        url, encoded_metadata, scc_hash, same_hash = match.groups()
+        metadata_parts = urllib.parse.unquote(encoded_metadata).split('—')
+        if len(metadata_parts) == 4:
+            username, task_name, date, time = metadata_parts
+            return scc_hash, username, task_name, date, time
+    return None, None, None, None, None
+
 # --- Live Clock JavaScript ---
 def live_clock():
     return """
@@ -279,15 +300,6 @@ with tabs[0]:
             st.markdown('<div class="hash-display">', unsafe_allow_html=True)
             st.markdown(metadata_link, unsafe_allow_html=True)
             st.markdown('</div>', unsafe_allow_html=True)
-
-    # Guidance on Verifying Citations
-    st.markdown("## Guidance on Verifying Citations")
-    st.markdown("""
-    <div class="info-card">
-    To verify a citation, you can recompute the hash using the original input data and compare it to the embedded hash. 
-    The <strong>'Verify Citation'</strong> tab allows you to do this easily.
-    </div>
-    """, unsafe_allow_html=True)
     
     st.markdown('</div>', unsafe_allow_html=True)
 
@@ -296,128 +308,46 @@ with tabs[1]:
     st.header("Verify Citation")
     st.markdown("""
     <div class="info-card">
-    Enter the citation details below to recompute and verify the hash. This ensures the citation hasn't been tampered with or fabricated.
+    Paste the generated citation and hash with their embedded links below to verify the citation's authenticity.
     </div>
     """, unsafe_allow_html=True)
 
-    # Initialize session state for storing verified hashes
-    if 'verified_hashes' not in st.session_state:
-        st.session_state.verified_hashes = []
+    # Input fields for citation and hash
+    citation_input = st.text_area("Paste Citation (with embedded link)", help="Paste the citation HTML, e.g., <a href='URL'>Author (Year)</a>", placeholder='<a href="https://example.com#:~:text=fragment" data-hash="hash">Author (Year)</a>')
+    hash_input = st.text_area("Paste Hash (with embedded link)", help="Paste the hash HTML, e.g., <a href='URL'>Author (Year). Hash</a>", placeholder='<a href="https://example.com#:~:text=metadata" data-hash="hash">Author (Year). Hash</a>')
 
-    # User Information Section
-    st.subheader("User Information")
-    col1, col2 = st.columns(2)
-    with col1:
-        verify_username = st.text_input("Username (for verification)", placeholder="e.g., john_doe")
-    with col2:
-        verify_task_name = st.text_input("Task Name (for verification)", placeholder="e.g., Literature Review Assignment")
-
-    # Citation Info Section
-    st.subheader("Citation Info")
-    col3, col4 = st.columns(2)
-    with col3:
-        verify_author_name = st.text_input("Author(s) Name (for verification)", placeholder="e.g., Smith or Smith et al.")
-    with col4:
-        verify_publication_year = st.text_input("Publication Year (for verification)", placeholder="e.g., 2023")
-    
-    col5, col6 = st.columns(2)
-    with col5:
-        verify_source_url = st.text_input("Source URL (for verification)", placeholder="https://example.com/article")
-    with col6:
-        verify_annotated_text = st.text_input("Annotated Text (for verification)", placeholder="e.g., Thermal comfort thresholds...")
-    
-    col7, col8 = st.columns(2)
-    with col7:
-        verify_date = st.text_input("Date (YYYY-MM-DD) (for verification)", placeholder="e.g., 2025-01-08")
-    with col8:
-        verify_time = st.text_input("Time (HH:MM:SS) (for verification)", placeholder="e.g., 14:30:25")
-    
-    expected_hash = st.text_input("Expected Hash (from the citation)", placeholder="Enter the full hash from the citation")
-
-    verify_button = st.button("Verify Hash", type="primary", use_container_width=True)
+    verify_button = st.button("Verify Citation", type="primary", use_container_width=True)
 
     if verify_button:
-        if not all([verify_username, verify_task_name, verify_author_name, verify_publication_year, 
-                   verify_source_url, verify_annotated_text, verify_date, verify_time, expected_hash]):
-            st.error("Please fill in all fields before verifying the hash.")
+        if not (citation_input and hash_input):
+            st.error("Please paste both the citation and hash before verifying.")
         else:
-            recomputed_hash = generate_citation_hash(
-                verify_author_name, verify_publication_year, verify_source_url,
-                verify_annotated_text, verify_annotated_text, verify_username, verify_task_name,
-                verify_date, verify_time
-            )
-
-            if recomputed_hash == expected_hash:
-                st.markdown("""
-                <div class="success-box">
-                    <strong>Hash verified successfully!</strong> The citation is authentic and hasn't been tampered with.
-                </div>
-                """, unsafe_allow_html=True)
-                
-                st.session_state.verified_hashes.append({
-                    "Author": verify_author_name,
-                    "Year": verify_publication_year,
-                    "URL": verify_source_url,
-                    "Fragment text": verify_annotated_text,
-                    "Cited text": verify_annotated_text,
-                    "Username": verify_username,
-                    "Task name": verify_task_name,
-                    "Date": verify_date,
-                    "Time": verify_time,
-                    "Original Hash": expected_hash,
-                    "Recomputed Hash": recomputed_hash,
-                    "Status": "Verified"
-                })
+            # Parse citation
+            author, year, url, fragment_text = parse_citation(citation_input)
+            # Parse hash and metadata
+            scc_hash, username, task_name, date, time = parse_metadata_hash(hash_input)
+
+            if not all([author, year, url, fragment_text, scc_hash, username, task_name, date, time]):
+                st.error("Invalid citation or hash format. Please ensure both inputs are correctly formatted HTML with embedded links.")
             else:
-                st.markdown("""
-                <div class="warning-box">
-                    <strong>Hash verification failed!</strong> The citation may have been altered or is not authentic.
-                </div>
-                """, unsafe_allow_html=True)
-                
-                st.session_state.verified_hashes.append({
-                    "Author": verify_author_name,
-                    "Year": verify_publication_year,
-                    "URL": verify_source_url,
-                    "Fragment text": verify_annotated_text,
-                    "Cited text": verify_annotated_text,
-                    "Username": verify_username,
-                    "Task name": verify_task_name,
-                    "Date": verify_date,
-                    "Time": verify_time,
-                    "Original Hash": expected_hash,
-                    "Recomputed Hash": recomputed_hash,
-                    "Status": "Failed"
-                })
-
-    if st.session_state.verified_hashes:
-        st.markdown("## Verification History")
-        df = pd.DataFrame(st.session_state.verified_hashes)
-        
-        st.markdown('<div class="verification-table">', unsafe_allow_html=True)
-        st.dataframe(df, use_container_width=True)
-        st.markdown('</div>', unsafe_allow_html=True)
-
-        # Download as CSV
-        @st.cache_data
-        def convert_df_to_csv(df):
-            return df.to_csv(index=False).encode('utf-8')
-
-        csv = convert_df_to_csv(df)
-
-        st.download_button(
-            label="Download Verification History as CSV",
-            data=csv,
-            file_name="scc_verification_history.csv",
-            mime="text/csv",
-            use_container_width=True
-        )
-        
-        # Clear history button
-        if st.button("Clear Verification History", type="secondary"):
-            st.session_state.verified_hashes = []
-            st.experimental_rerun()
-    
+                # Recompute hash
+                recomputed_hash = generate_citation_hash(
+                    author, year, url, fragment_text, fragment_text, username, task_name, date, time
+                )
+
+                if recomputed_hash == scc_hash:
+                    st.markdown("""
+                    <div class="success-box">
+                        <strong>Hash verified successfully!</strong> The citation is authentic and hasn't been tampered with.
+                    </div>
+                    """, unsafe_allow_html=True)
+                else:
+                    st.markdown("""
+                    <div class="warning-box">
+                        <strong>Hash verification failed!</strong> The citation may have been altered or is not authentic.
+                    </div>
+                    """, unsafe_allow_html=True)
+
     st.markdown('</div>', unsafe_allow_html=True)
 
 # Footer