Update app.py

app.py

@@ -97,8 +97,10 @@ def normalise_hyphens(text):
     return text.replace('\u2011', '-').replace('\u2013', '-').replace('\u2014', '-')
 
 def encode_text_fragment(text):
-    # Encode text for W3C Text Fragments, preserving only hyphens
-    # En dashes (–) and em dashes (—) are encoded as %E2%80%93 and %E2%80%94
+    # Encode text for W3C Text Fragments, preserving only regular hyphens (U+002D)
+    # Non-breaking hyphens (U+2011) are encoded as %E2%80%91
+    # En dashes (U+2013) are encoded as %E2%80%93
+    # Em dashes (U+2014) are encoded as %E2%80%94
     return urllib.parse.quote(text, safe='-')
 
 def generate_citation_hash(author, year, url, fragment_text, cited_text, username, task_name, current_date, current_time):