Update app.py

app.py

@@ -130,29 +130,23 @@ def validate_excel_structure(df):
         return False, validation_messages
     
     try:
-        # Check abstract length
-        min_length = df['Abstract'].fillna('').astype(str).str.len().min()
-        if min_length < 50:
-            validation_messages.append("Some abstracts are too short (less than 50 characters)")
-        
-        # Check publication year format
+        # Check publication year format - this is useful for sorting/filtering
         df['Publication Year'] = pd.to_numeric(df['Publication Year'], errors='coerce')
         if df['Publication Year'].isna().any():
-            validation_messages.append("Some publication years are invalid")
+            validation_messages.append("Some publication years are invalid. Please ensure all years are in numeric format (e.g., 2024)")
         else:
             years = df['Publication Year'].dropna()
-            if len(years) > 0:  # Only check if we have valid years
+            if len(years) > 0:
                 if years.min() < 1900 or years.max() > 2025:
                     validation_messages.append("Publication years must be between 1900 and 2025")
-        
-        # Check DOIs (allow empty DOIs)
-        doi_pattern = r'10\.\d{4,}/.+'
-        valid_dois = df['DOI'].fillna('').astype(str).str.contains(doi_pattern, na=True, regex=True)
-        if not valid_dois.all() and len(valid_dois) > 0:
-            validation_messages.append("Some DOIs are in invalid format")
+            
+        # For short abstracts - just show a warning
+        short_abstracts = df['Abstract'].fillna('').astype(str).str.len() < 50
+        if short_abstracts.any():
+            st.warning("ℹ️ Some abstracts are quite short, but will still be processed")
             
     except Exception as e:
-        validation_messages.append(f"Error validating data: {str(e)}")
+        validation_messages.append(f"Error checking data format: {str(e)}")
     
     return len(validation_messages) == 0, validation_messages