Introduction
this proposal is to add a MDX compiler to analyze Cube MDX expression and build TM1py Native View Object.
mdx = """
SELECT
{[DIM A].[HIE A].[ELE A]}
* {[DIM B].[ELE B]}
ON ROWS,
NONEMPTY
{[DIM C].[HIE C].MEMBERS}
ON COLUMNS
FROM [CUBE NAME]
WHERE (
[DIM D].[ELE D]
)
"""
native_view = MdxBuilder.to_tm1py_native_view(mdx)
tm1.views.create_or_update(cueb_name, view_name, native_view)Motivation
Comparing to MDX View, Native View has a significant performance advantage on retrieving large data from Cube. But currently we don't have an direct approach to create a Native View with MDX.
I think we may have a good enough compiler to convert simple MDX to Native View to low computing cost of query.
Dependencies
this feature is requiring a new technology, which may increase a cost of maintenance.
Lark is a parser library to create a Concrete Syntax Tree with simple grammar setup.
Introduction
this proposal is to add a MDX compiler to analyze Cube MDX expression and build TM1py Native View Object.
Motivation
Comparing to MDX View, Native View has a significant performance advantage on retrieving large data from Cube. But currently we don't have an direct approach to create a Native View with MDX.
I think we may have a good enough compiler to convert simple MDX to Native View to low computing cost of query.
Dependencies
this feature is requiring a new technology, which may increase a cost of maintenance.
Lark is a parser library to create a Concrete Syntax Tree with simple grammar setup.