Tables — Python SDK Quick Reference
Condensed from azure-data-tables-py. Full patterns (batch operations, async client, typed entities, query parameters) in the azure-data-tables-py plugin skill if installed.
Install
pip install azure-data-tables azure-identity
Quick Start
Auth:
DefaultAzureCredentialis for local development. See auth-best-practices.md for production patterns.
from azure.data.tables import TableClient
from azure.identity import DefaultAzureCredential
table_client = TableClient("https://<account>.table.core.windows.net", "mytable", DefaultAzureCredential())Best Practices
- Design partition keys for query patterns and even distribution
- Query within partitions whenever possible (cross-partition is expensive)
- Use batch operations for multiple entities in same partition
- Use
upsert_entityfor idempotent writes - Use parameterized queries to prevent injection
- Keep entities small — max 1MB per entity
- Use async client for high-throughput scenarios