v2.4.5

🎉 New features

• Added username property to the ApiToken model.
• Added functions for validating custom package files to pyatlan.test_utils.
• Added an optional parameter asset_selection to DataProduct.updater() method, allowing users to update assets within the data product.
• Added the DataProductsAssetsDSL.get_asset_selection() method, which returns the asset selection DSL string for a data product based on the specified IndexSearchRequest.

🐞 Bug fixes

• Fixed pagination issues in AuditSearchResults.
• Fixed multipart form handling for AtlanTag image uploads.

Full Changelog: 2.4.4...2.4.5

v2.4.4

🎉 New features

• Added a new function (pkg.utils.set_package_headers) to configure the AtlanClient with the required custom package headers using environment variables.

🐞 Bug fixes

• Fixed pkg.utils.validate_connection() to include other types (Connection, Dict).

🥗 QOL improvements

• Moved common test functions to a separate test_utils package.
• Moved nanoid from the requirements-dev.txt to requirements.txt.

Full Changelog: 2.4.3...2.4.4

v2.4.3

🎉 New features

• Added a new connector type DM for data modeling assets.

Full Changelog: 2.4.2...2.4.3

v2.4.2

🎉 New features

• Added support for GCS presigned URL file uploads.
• Added the find_run_by_id method to the WorkflowClient.

🐞 Bug fixes

• Fixed the interpretation of text-only indexed fields for assets.

🥗 QOL improvements

• Generated the latest typedef models.
• Added pyatlan to the Anaconda packages.

Full Changelog: 2.4.1...2.4.2

v2.4.1

🎉 New features

• Added support for Azure Blob presigned URL file uploads.

🥗 QOL improvements

• Replaced pyatlan.utils.HTTPStatus with the standard library http.HTTPStatus (available since Python 3.4).

Full Changelog: 2.4.0...2.4.1

v2.4.0

🥗 QOL improvements

• Implemented lazy imports for pyatlan.model.assets: This change reduces the import time for assets by deferring the import of modules until they are actually needed. As a result, users will experience faster startup times and reduced memory usage when working with assets.

🐞 Bug fixes

• Fixed missing check for asset.guid in the Readme.creator() method, which caused issues with updating Readme assets when the asset was passed using Asset.ref_by_qualified_name() instead of Asset.ref_by_guid(). The SDK now throws a ValueError if asset.guid is missing in the Readme.creator() method to prevent these issues.

Full Changelog: 2.3.3...2.4.0

v2.3.3

🎉 New features

• Added support for nested aggregations.
• Added support for trident suggestions.
• Added Superset assets creator() methods.

Full Changelog: 2.3.2...2.3.3

v2.3.2

🎉 New features

• Added a new ELT connector type: PREFECT.

🐞 Bug fixes

• Fixed the ErrorCode.CM_ATTR_NOT_FOUND_BY_NAME error message in CustomMetadataCache.get_attr_id_for_name(). Previously, it was receiving the wrong parameters.

Full Changelog: 2.3.1...2.3.2

v2.3.1

🎉 New features

• Added WorkflowClient.find_by_id() method for workflow search/retrieval.

🥗 QOL improvements

• Made AtlanError messages more descriptive.

Full Changelog: 2.3.0...2.3.1

v2.3.0

⛑️ Breaking changes

• Introduced a new pagination approach in AssetClient.search() and FluentSearch.execute() called bulk search (disabled by default). It minimizes system impact when handling large result sets. The SDK switches to this search operation automatically if results exceed a predefined threshold (i.e: 100,000 results). Alternatively, users can enable bulk search explicitly by setting bulk=True in AssetClient.search() or FluentSearch.execute(). The breaking change is in regards to searches that return more than 100,000 results — either the results will now be sorted differently or an error will be thrown.

• The AssetClient.search() and FluentSearch.execute() methods will now raise an exception (InvalidRequestError) in the following scenarios:

  • when bulk search is enabled (bulk=True) and any user-specified sorting options are found in the search request.

  • when bulk search is disabled (bulk=False), the number of results exceeds the predefined threshold (i.e: 100,000 assets), and any user-specified sorting options are found in the search request.

  This is because the bulk search approach ignores user-specified sorting and instead reorders the results based on the creation timestamps of assets to handle large numbers of assets efficiently.

🥗 QOL improvements

• Pinned urllib3>=1.26.0,<3 and moved networkx to the dev requirements to avoid potential version mismatches.