Implementation:Eventual Inc Daft DataFrame Sort

Knowledge Sources	Daft Daft Docs
Domains	Data_Engineering, Data_Analysis
Last Updated	2026-02-08 00:00 GMT

Overview

Concrete tool for globally sorting DataFrame rows by column values provided by the Daft library.

Description

The sort method on a Daft DataFrame performs a global sort of all rows based on one or more columns or expressions. Each sort column can have an independent descending flag and null placement configuration. When nulls_first is not specified, it defaults to matching the desc parameter (nulls first when descending, nulls last when ascending). Since this is a global sort across all partitions, it requires an expensive repartition operation and can be slow on large datasets.

Usage

Use this method on a DataFrame when you need fully ordered results for display, top-N queries, reporting, or operations that require sorted input.

Code Reference

Source Location

Repository: Daft
File: daft/dataframe/dataframe.py
Lines: L2548-2642

Signature

def sort(
    self,
    by: ColumnInputType | list[ColumnInputType],
    desc: bool | list[bool] = False,
    nulls_first: bool | list[bool] | None = None,
) -> "DataFrame"

Import

# Method on DataFrame, no separate import needed
sorted_df = df.sort("col", desc=True)

I/O Contract

Inputs

Name	Type	Required	Description
by	list[ColumnInputType]	Yes	Column(s) to sort by; can be string column names, expressions, or a list of either
desc	list[bool]	No	Sort descending; can be a single bool or a list matching the number of sort columns; defaults to False
nulls_first	list[bool] \| None	No	Place nulls first; defaults to None (nulls treated as greatest value, matching desc behavior)

Outputs

Name	Type	Description
return	DataFrame	A new DataFrame with rows sorted according to the specified columns and directions

Usage Examples

Basic Usage

import daft

df = daft.from_pydict({"x": [3, 2, 1], "y": [6, 4, 5]})

# Sort by expression (x + y) ascending
sorted_df = df.sort(df["x"] + df["y"])
sorted_df.show()

# Multi-column sort with different directions
df = daft.from_pydict({"x": [1, 2, 1, 2], "y": [9, 8, 7, 6]})
sorted_df = df.sort(["x", "y"], [True, False])
sorted_df.show()

# Sort with explicit null positioning
df = daft.from_pydict({"x": [1, 2, None], "y": [9, 8, None]})
sorted_df = df.sort(["x", "y"], [True, False], nulls_first=[True, True])
sorted_df.show()

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment