fix: Handle duplicate IDs in rolodex import gracefully

- Added duplicate tracking within import session (seen_in_import set)
- Skip records that already exist in database
- Added fallback to row-by-row insert when bulk insert fails
- Track skipped records in result
- Prevents cascade errors after UNIQUE constraint violation
- Gracefully handles legacy data with duplicate IDs

app/import_legacy.py

@@ -591,13 +591,16 @@ def import_rvarlkup(db: Session, file_path: str) -> Dict[str, Any]:
 
 def import_rolodex(db: Session, file_path: str) -> Dict[str, Any]:
     """Import ROLODEX.csv → Rolodex model."""
-    result = {'success': 0, 'errors': [], 'total_rows': 0}
+    result = {'success': 0, 'errors': [], 'total_rows': 0, 'skipped': 0}
     
     try:
         f, encoding = open_text_with_fallbacks(file_path)
         reader = csv.DictReader(f)
         
+        # Track IDs we've seen in this import to handle duplicates
+        seen_in_import = set()
         batch = []
+        
         for row_num, row in enumerate(reader, start=2):
             result['total_rows'] += 1
             
@@ -606,6 +609,17 @@ def import_rolodex(db: Session, file_path: str) -> Dict[str, Any]:
                 if not rolodex_id:
                     continue
                 
+                # Skip if we've already processed this ID in current import
+                if rolodex_id in seen_in_import:
+                    result['skipped'] += 1
+                    continue
+                
+                # Skip if it already exists in database
+                if db.query(Rolodex).filter(Rolodex.id == rolodex_id).first():
+                    result['skipped'] += 1
+                    seen_in_import.add(rolodex_id)
+                    continue
+                
                 record = Rolodex(
                     id=rolodex_id,
                     prefix=clean_string(row.get('Prefix')),
@@ -629,20 +643,47 @@ def import_rolodex(db: Session, file_path: str) -> Dict[str, Any]:
                     memo=clean_string(row.get('Memo'))
                 )
                 batch.append(record)
+                seen_in_import.add(rolodex_id)
                 
                 if len(batch) >= BATCH_SIZE:
+                    try:
                         db.bulk_save_objects(batch)
                         db.commit()
                         result['success'] += len(batch)
                         batch = []
+                    except IntegrityError as ie:
+                        db.rollback()
+                        # Handle any remaining duplicates by inserting one at a time
+                        for record in batch:
+                            try:
+                                db.add(record)
+                                db.commit()
+                                result['success'] += 1
+                            except IntegrityError:
+                                db.rollback()
+                                result['skipped'] += 1
+                        batch = []
                     
             except Exception as e:
                 result['errors'].append(f"Row {row_num}: {str(e)}")
         
+        # Save remaining batch
         if batch:
+            try:
                 db.bulk_save_objects(batch)
                 db.commit()
                 result['success'] += len(batch)
+            except IntegrityError:
+                db.rollback()
+                # Handle any remaining duplicates by inserting one at a time
+                for record in batch:
+                    try:
+                        db.add(record)
+                        db.commit()
+                        result['success'] += 1
+                    except IntegrityError:
+                        db.rollback()
+                        result['skipped'] += 1
         
         f.close()
         logger.info("import_rolodex_complete", **result)