Connect to Databend Using Python

Python Driver for Databend

Connect to Databend using Python with our official drivers supporting both synchronous and asynchronous operations.

Quick Start

Choose your preferred approach:

Package	Best For	Installation
databend-driver	Direct database operations, async/await	pip install databend-driver
databend-sqlalchemy	ORM integration, existing SQLAlchemy apps	pip install databend-sqlalchemy

Connection String: See Drivers Overview for DSN format and examples.

---

databend-driver (Recommended)

Features

• ✅ Native Performance: Direct connection to Databend
• ✅ Async/Sync Support: Choose your programming style
• ✅ PEP 249 Compatible: Standard Python DB API
• ✅ Type Safety: Full Python type mapping

Synchronous Usage

from databend_driver import BlockingDatabendClient

# Connect and execute
client = BlockingDatabendClient('<your-dsn>')
cursor = client.cursor()

# DDL: Create table
cursor.execute("CREATE TABLE users (id INT, name STRING)")

# Write: Insert data  
cursor.execute("INSERT INTO users VALUES (?, ?)", (1, 'Alice'))

# Query: Read data
# Query: Read data
cursor.execute("SELECT * FROM users")

# Get column names
# cursor.description returns a list of tuples, where the first element is the column name
print(f"Columns: {[desc[0] for desc in cursor.description]}")

for row in cursor.fetchall():
    # row is a databend_driver.Row object
    # Access by column name
    print(f"id: {row['id']}, name: {row['name']}")

cursor.close()

Working with Row Objects

The Row object supports multiple access patterns and methods:

for row in cursor.fetchall():
    # 1. Access by column name (Recommended)
    print(f"Name: {row['name']}")
    
    # 2. Access by index
    print(f"First column: {row[0]}")
    
    # 3. Convert to tuple
    print(f"Values: {row.values()}")
    
    # 4. Explicit methods
    print(row.get_by_field('name'))
    print(row.get_by_index(0))

Asynchronous Usage

import asyncio
from databend_driver import AsyncDatabendClient

async def main():
    client = AsyncDatabendClient('<your-dsn>')
    conn = await client.get_conn()
    
    # DDL: Create table
    await conn.exec("CREATE TABLE users (id INT, name STRING)")
    
    # Write: Insert data
    await conn.exec("INSERT INTO users VALUES (?, ?)", (1, 'Alice'))
    
    # Query: Read data
    rows = await conn.query_iter("SELECT * FROM users")
    async for row in rows:
        print(row.values())
    
    await conn.close()

asyncio.run(main())

---

databend-sqlalchemy

For SQLAlchemy ORM integration:

from sqlalchemy import create_engine, text

engine = create_engine('<your-dsn>')
with engine.connect() as conn:
    result = conn.execute(text("SELECT 1"))
    print(result.fetchall())

---

Data Type Mappings

Databend	Python	Notes
Numeric Types
BOOLEAN	bool
TINYINT	int
SMALLINT	int
INT	int
BIGINT	int
FLOAT	float
DOUBLE	float
DECIMAL	decimal.Decimal	Precision preserved
Date/Time
DATE	datetime.date
TIMESTAMP	datetime.datetime
INTERVAL	datetime.timedelta
Text/Binary
VARCHAR	str	UTF-8 encoded
BINARY	bytes
Complex Types
ARRAY	list	Nested structures supported
TUPLE	tuple
MAP	dict
VARIANT	str	JSON-encoded
BITMAP	str	Base64-encoded
GEOMETRY	str	WKT format

Resources

• PyPI: databend-driver • databend-sqlalchemy
• GitHub: databend-driver • databend-sqlalchemy