Per the M language specification, #sections[Section1] should return a record containing one field per query in the project, with that field's value set to the results of the given query.
Power BI Desktop and Excel work this way. However, Power Query in dataflows does not. This limits the ability to write M code that dynamically adapts based on the queries present in a project.
Can PQ in dataflows be fixed so that #sections[Section1] works according to the specification?
Thank you for your consideration!
To Reproduce
In both Power BI Desktop and in dataflows, create the following two queries:
- A—defined as '123'
- B—defend as '#sections[Section1]'
In both Power BI Desktop and dataflows, the output of B will be a record that includes a field named "A". In Desktop, this field's value will be 123 (expected behavior, per the M spec); however, in dataflows, this field's value will be null (incorrect behavior).
yexu
on:
CopyJob Query as Source
frithjof_v
on:
Monitor: Filter by item name or activity name
on:
New Fabric Certification for Data Science