Update app.py

app.py

@@ -68,7 +68,7 @@ def paraphrase(
 
 def remove_punctuations(text):
     # Remove punctuations while preserving hyphenated words
-    return re.sub(r'(?<!\w)-|-(?!\w)', ' ', re.sub(r'[^\w\s,-]', '', text))
+    return re.sub(r'(?<!\w)-|-(?!\w)', ' ', re.sub(r'[^\w\s-]', '', text))
 
 def tokenize(sentence):
     # Remove punctuations using the updated function and tokenize the sentence into words
@@ -293,6 +293,18 @@ for sentence in highlighted_sentence_list:
 
 
 
+highlighted_main_sentence = main_sentence
+
+# Iterate through each paraphrase and apply different colors to combined words
+for i, combined_words in enumerate(combined_words_list):
+    color = colors[i % len(colors)]  # Get color for this paraphrase
+    # Highlight combined words from this paraphrase with the corresponding color
+    for substring in combined_words:
+        highlighted_main_sentence = highlighted_main_sentence.replace(substring, f'<span style="background-color: {color}; color: white;">{substring}</span>')
+
+st.markdown("\nHighlighted Main Sentence with LCS from All Paraphrases:")
+st.write(highlighted_main_sentence, unsafe_allow_html=True)
+