Weekly update¶
This week we:
- Started work on an experimental metadata model toolkit and a new model for multi-dimensional datasets (cubes)
- Added the ability to use
get_by_guid
in the Python SDK, without requiring anasset_type
to be provided - Added some minor model and connector updates and fixed various issues
Custom types¶
We've created an experimental toolkit for defining custom metadata types, which we're calling the "typedef toolkit". This is a Pkl-based approach for defining extensions to our metadata model.
Atlanian use only
For now, this is for Atlanian use only, but we hope to open it up more broadly in the future.
Python¶
We've added:
- (Experimental) Support for Sigma and SQL Server crawlers.
- The ability to use
get_by_guid
andretrieve_minimal
methods on assets without needing to specify anasset_type
. - Adds AWS IoT connector types:
AWS_SITE_WISE
andAWS_GREENGRASS
. - Adds credential widget to package toolkit.
We've fixed:
- Fixes an error related to
Range
queries ondatetime
attributes. - Fixes create method in the user client that was causing
500
(internal server) errors.
Kotlin (and Java)¶
We've added:
- (Experimental) Model for multi-dimensional datasets (cubes).
- Adds latest model for Spark jobs.
- Adds AWS Green Grass and Site Wise connector types.
We've fixed:
- Reduces default verbosity of caching-related errors (still included at debug-level).
Custom packages¶
We've added:
- (Experimental) New cube asset builder, for loading and managing multi-dimensional datasets (cubes).
We've fixed:
- Issues with sending emails from custom packages: admin export, asset export (basic), adoption export, metadata impact report.
ClassNotFoundException
andIndexOutOfBoundsException
errors in the OpenAPI spec loader package.
We've improved:
- Enrichment migrator package now allows copying enrichment to one or many target set of assets.