Keri sisuni

digna Python SDK Reference 2026.06

This section documents the Python SDK for digna. It is organized as a multi-page reference: use this overview to understand the client, then continue with the dedicated pages for quickstart, resources, models, errors, and generated API documentation.

The SDK is published as the digna-sdk package and exposes a stable, versioned client for the digna REST API.

SDK Basics

Overview

The SDK follows a resource-oriented client design. Each API area is exposed as a first-class client on the top-level DignaClient, with typed request and response models and consistent error handling.

from digna_sdk import DignaClient

client = DignaClient(
    base_url="https://your-digna-instance",
    token="...",
)

Core Features

  • Typed models — every request and response is validated with pydantic, so your editor and type checker catch mistakes before you hit the network.
  • Resource-oriented — client.projects, client.data_sources, client.data_sets, client.attributes, client.check_definitions, client.db_connections, client.inspection_requests, and client.inspection_statuses each expose simple list / get / create / update / delete methods.
  • Clear errors — API errors raise DignaAPIError (or a more specific subclass like DignaAuthenticationError, DignaAuthorizationError, or DignaNotFoundError) instead of silently returning None.

Installation

pip install digna-sdk

Reference Pages

This release is organized across the following pages:

  • Quickstart — connect and make your first calls.
  • Resources — the full list of available resource clients.
  • Models — the pydantic models used for input/output.
  • Errors — the exception hierarchy.
  • API Reference — auto-generated reference docs.