`get_elements_dataframe` on v12 fails with `use_blob=True`

[MariusWirtz]

The below script works on v11 but fails on v12.

The MDX View used as the source view for the unbound TI produces a different data source in v12 compared to v11.

This needs to be raised with IBM or adjusted for in TM1py.

with TM1Service(**v12_params) as tm1:
    print(tm1.server.get_product_version())

    df = tm1.elements.get_elements_dataframe(
        dimension_name="Boardgame",
        hierarchy_name="Boardgame",
        skip_consolidations=True,
        skip_parents=False,
        skip_weights=False,
        use_blob=True
    )

    print(df.to_markdown())

WITH 
MEMBER [}ElementAttributes_Boardgame].[L000] 
AS [Boardgame].[Boardgame].CurrentMember.Parent.PROPERTIES('NAME')

MEMBER [}ElementAttributes_Boardgame].[level000_Weight]
AS IIF(
[Boardgame].[Boardgame].CurrentMember.Properties('MEMBER_WEIGHT') = '',
0,
[Boardgame].[Boardgame].CurrentMember.Properties('MEMBER_WEIGHT'))

SELECT
{ { Tm1FilterByLevel({ [Boardgame].[Boardgame].Members }, 0) } } ON ROWS,
{ {Tm1SubsetAll([}ElementAttributes_Boardgame])} + {[}ElementAttributes_Boardgame].[level000_Weight],[}ElementAttributes_Boardgame].[L000]} } ON COLUMNS
FROM [}ElementAttributes_Boardgame]  

[rclapp]

What is the difference in shape?

[MariusWirtz]

use_blob=True

with TM1Service(**v12_params) as tm1:
    print(tm1.server.get_product_version())

    df = tm1.elements.get_elements_dataframe(
        dimension_name="Boardgame",
        hierarchy_name="Boardgame",
        skip_consolidations=True,
        skip_parents=False,
        skip_weights=False,
        use_blob=True
    )

    print(df.to_markdown())

	Boardgame	Type	ProductCode	Age	Players	level000_Weight	level000
0	Monopoly	Numeric	P0001	7+	>4	0	0
1	Scrabble	Numeric	1	All Boardgames	P0002	0	0
2	Connect4	Numeric	10+	>4	1	0	0
3	Battleship	Numeric	All Boardgames	P0003	2+	0	0
4	Trivial Pursuit	Numeric	2	1	All Boardgames	0	0
5	Othello	Numeric	P0004	5+	2	0	0
6	Risk	Numeric	1	All Boardgames	P0005	0	0
7	Catan	Numeric	16+	>4	1	0	0
8	Cluedo	Numeric	All Boardgames	P0006	7+	0	0
9	Ludo	Numeric	2	1	All Boardgames	0	0
10	Other	Numeric	P0007	16+	>4	0	0

The intermediary blob file looks like this. It seems the calculated members aren't calculated correctly when the MDX is a TI datasource.

"Boardgame"~"}ElementAttributes_Boardgame"~"Value"
"Monopoly"~"ProductCode"~"P0001"
"Monopoly"~"Age"~"7+"
"Monopoly"~"Players"~">4"
"Monopoly"~"level000_Weight"~"0"
"Monopoly"~"L000"~"0"
"Scrabble"~"ProductCode"~"1"
"Scrabble"~"Age"~"All Boardgames"
"Scrabble"~"Players"~"P0002"
"Scrabble"~"level000_Weight"~"0"
"Scrabble"~"L000"~"0"
"Connect4"~"ProductCode"~"10+"
"Connect4"~"Age"~">4"
"Connect4"~"Players"~"1"
"Connect4"~"level000_Weight"~"0"
"Connect4"~"L000"~"0"
"Battleship"~"ProductCode"~"All Boardgames"
"Battleship"~"Age"~"P0003"
"Battleship"~"Players"~"2+"
"Battleship"~"level000_Weight"~"0"
"Battleship"~"L000"~"0"
"Trivial Pursuit"~"ProductCode"~"2"
"Trivial Pursuit"~"Age"~"1"
"Trivial Pursuit"~"Players"~"All Boardgames"
"Trivial Pursuit"~"level000_Weight"~"0"
"Trivial Pursuit"~"L000"~"0"
"Othello"~"ProductCode"~"P0004"
"Othello"~"Age"~"5+"
"Othello"~"Players"~"2"
"Othello"~"level000_Weight"~"0"
"Othello"~"L000"~"0"
"Risk"~"ProductCode"~"1"
"Risk"~"Age"~"All Boardgames"
"Risk"~"Players"~"P0005"
"Risk"~"level000_Weight"~"0"
"Risk"~"L000"~"0"
"Catan"~"ProductCode"~"16+"
"Catan"~"Age"~">4"
"Catan"~"Players"~"1"
"Catan"~"level000_Weight"~"0"
"Catan"~"L000"~"0"
"Cluedo"~"ProductCode"~"All Boardgames"
"Cluedo"~"Age"~"P0006"
"Cluedo"~"Players"~"7+"
"Cluedo"~"level000_Weight"~"0"
"Cluedo"~"L000"~"0"
"Ludo"~"ProductCode"~"2"
"Ludo"~"Age"~"1"
"Ludo"~"Players"~"All Boardgames"
"Ludo"~"level000_Weight"~"0"
"Ludo"~"L000"~"0"
"Other"~"ProductCode"~"P0007"
"Other"~"Age"~"16+"
"Other"~"Players"~">4"
"Other"~"level000_Weight"~"0"
"Other"~"L000"~"0"

use_blob=False

with TM1Service(**v12_params) as tm1:
    print(tm1.server.get_product_version())

    df = tm1.elements.get_elements_dataframe(
        dimension_name="Boardgame",
        hierarchy_name="Boardgame",
        skip_consolidations=True,
        skip_parents=False,
        skip_weights=False,
        use_blob=False
    )

    print(df.to_markdown())

	Boardgame	Type	ProductCode	Age	Players	level000_Weight	level000
0	Monopoly	Numeric	P0001	7+	>4	1	All Boardgames
1	Scrabble	Numeric	P0002	10+	>4	1	All Boardgames
2	Connect4	Numeric	P0003	2+	2	1	All Boardgames
3	Battleship	Numeric	P0004	5+	2	1	All Boardgames
4	Trivial Pursuit	Numeric	P0005	16+	>4	1	All Boardgames
5	Othello	Numeric	P0006	7+	2	1	All Boardgames
6	Risk	Numeric	P0007	16+	>4	1	All Boardgames
7	Catan	Numeric	P0008	10+	>4	1	All Boardgames
8	Cluedo	Numeric	P0009	7+	>4	1	All Boardgames
9	Ludo	Numeric	P0010	2+	>4	1	All Boardgames
10	Other	Numeric				1	All Boardgames