New Functions
-
sempy_labs.semantic_model
-
find_unused_objects An interactive UI which offers 2 methods for identifying used/unused semantic model objects. Method 1: workspace monitoring. Method 2: downstream reports. Choose to either analyze a range of queries from workspace monitoring (and dependencies) or downstream reports. Each method will analyze which semantic models are not used in that context and which semantic model objects were used (and count how many times they were referenced). This can help you identify which tables/columns/measures you may be able to remove from your model due to inactivity.
-
migrate_to_direct_lake Until now, migration from import/DirectQuery to Direct Lake was done via running code snippets in a notebook. This function offers an interactive UI showing the exact process - what is supported and what is not supported (what will be migrated and what will not) and walks you through the process - no code involved. This also supports in-place conversions for import/DirectQuery models (if they contain no features which are unsupported in Direct Lake mode). And, this supports the Direct Lake model sourcing from either a Lakehouse or a Warehouse. By default, the Direct Lake model is set to Direct Lake on OneLake.
-
-
sempy_labs.variable_library
Updated Functions
-
lineage_view Made additional enhancements to the UI (change model/workspace in the UI, change size of model/report panels)
-
vertipaq_analyzer The interactive UI now has a 'delta analyzer' button for Direct Lake models to fetch delta analyzer stats for tables in Direct Lake mode which source from a lakehouse. This is especially useful as Direct Lake models do not always show table/column stats as this Vertipaq info is blocked. In order to use this 'delta analyzer' feature, you must run this in a PySpark notebook.
-
run_model_bpa has a new rule 'Avoid using SUM or AVERAGE on a string column' (#1222)
Bug Fixes
- #1261
- Fixed the convert_direct_lake_to_import and migrate_direct_lake_to_import functions (#1291)