Update app.py

app.py

@@ -110,15 +110,37 @@ def split_text(text, max_tokens=500):
 
     return chunks
 
+def extract_year(text):
+    year_match = re.search(r'\b(\d{4})\b', text)
+    return year_match.group(1) if year_match else None
+
 def create_bibtex_entry(data):
-    bibtex = "@article{idnothing,\n"
+    # Determine the entry type
+    if 'Journal' in data:
+        entry_type = 'article'
+    elif 'Booktitle' in data:
+        entry_type = 'incollection'
+    else:
+        entry_type = 'book'
+
+    # Extract year from 'None' if it exists
+    none_content = data.pop('None', '')
+    year = extract_year(none_content)
+    if year and 'Year' not in data:
+        data['Year'] = year
+
+    # Create BibTeX ID
+    author_words = data.get('Author', '').split()
+    first_author = author_words[0] if author_words else 'Unknown'
+    bibtex_id = f"{first_author}{year}" if year else first_author
+
+    bibtex = f"@{entry_type}{{{bibtex_id},\n"
     for key, value in data.items():
-        if key != 'None' and value.strip():
+        if value.strip():
             bibtex += f"  {key.lower()} = {{{value.strip()}}},\n"
     bibtex = bibtex.rstrip(',\n') + "\n}"
     return bibtex
 
-
 def transform_chunks(marianne_segmentation):
     marianne_segmentation = pd.DataFrame(marianne_segmentation)
     marianne_segmentation = marianne_segmentation[marianne_segmentation['entity_group'] != 'separator']
@@ -153,8 +175,6 @@ def transform_chunks(marianne_segmentation):
     
     final_html = '\n'.join(html_output)
     return final_html, bibtex_entry
-    
-
 
 # Class to encapsulate the Falcon chatbot
 class MistralChatBot: