Sort rows
Sort rows — sortRows
Sort by one or more columns.
Use cases
- Order results by amount, date, or rank for presentation or export.
- Set up row order before Limit rows (top-N) or a Window function.
What it does
Reorders rows by the chosen keys. Multiple keys resolve ties: the first column is primary, subsequent ones are tie-breakers.
Before
| region | amount |
|---|---|
| North | 80 |
| South | 150 |
| East | 30 |
| West | 120 |
4 rows · 2 cols
Sort rows (columns=[amount], ascending=false)
After
| region | amount |
|---|---|
| South | 150 |
| West | 120 |
| North | 80 |
| East | 30 |
4 rows · 2 cols
Configuration
| Config key | Type | Required | Description |
|---|---|---|---|
columns | string[] | Yes | Sort keys (in priority order) |
ascending | bool | bool[] | No | Default true; pass a list for per-column direction |
na_position | string | No | last (default) or first |
Generated Python code
df_2 = df_1.sort_values('amount', ascending=False)
Tips & common mistakes
- Per-column direction: pass an
ascendinglist the same length ascolumns(e.g. sort by region ascending, amount descending). - Window functions order within their own
order_by, so you don't need a separate sort just to feed one.