How to transfer multiple cells to write instead of one cell at a time?

[ishmeets920]

Hello All,

Hope all are doing good.

Usecase : Need to write the data from one cube to another

Current code : 
cellset = {(Dimensionlist[0], Dimensionlist[1], Dimensionlist[2], Dimensionlist[3]): Value for
            Dimensionlist[0], Dimensionlist[1], Dimensionlist[2], Dimensionlist[3], Value in df.values}
 
 tm2.cubes.cells.write_values(source_cube, cellset,skip_non_updateable=True, dimensions=Target_Dimensions)

Note: It works but takes one cell to copy at one time. How can I fasten the write process?

[MariusWirtz]

Check out this help article. It explains how to write to TM1 most efficiently with TM1py:

https://cubewise-code.github.io/tm1py-tales/2021/write-with-tm1py.html

[ishmeets920]

Hi @MariusWirtz,
Thanks for replying.

Needed to know how can I create this type of dataframe dynamically for the sourceview i fetched.

data = pd.DataFrame({
'd1': ['e1', 'e2', 'e4'],
'd2': ['e2', 'e3', 'e1'],
'value': [11.5, 12.3, 7.5]
})

Let say if i have more number of dimensions.

[MariusWirtz]

Here is a sample with more dimensions. Does that help you?

data = pd.DataFrame({
'd1': ['e1', 'e2', 'e4'],
'd2': ['e2', 'e3', 'e1'],
'd3': ['e1', 'e4', 'e2'],
'measure': ['value', 'value', 'value'],
'value': [11.5, 12.3, 7.5]
})

[ishmeets920]

I want to create this thing dynamically based on the length of elements in dimensions.
Like, if i have total of 4 dimensions like

Client = [E1, E2, E3, E4, E5]
Server = [F1, F2, F3, F4]
Lines = [A1........A200]
Data = [F1....F100000]

I have to create a dataframe based on what I fetch from cube dynamically, like the length of
Lines and values in the Data dimension for lines.

[MariusWirtz]

You don't need to use data frame at all if you don't want to. You can use the write function which takes a dictionary as input. You can read about it here: https://cubewise-code.github.io/tm1py-tales/2021/write-with-tm1py.html

Whatever structure you use, it must map to the cube!
So if you have a cube with 5 dimensions, your data frame must have 6 columns (5 dimensions + value column).
If you use a dictionary, your keys will be tuples of 5 elements.

[ishmeets920]

I have tried with write function with this code

cellset = {(Dimensionlist[0], Dimensionlist[1], Dimensionlist[2], Dimensionlist[3]): Value for
Dimensionlist[0], Dimensionlist[1], Dimensionlist[2], Dimensionlist[3], Value in df.values}

tm2.cubes.cells.write_values(source_cube, cellset,skip_non_updateable=True, dimensions=Target_Dimensions)

It is taking one cell and then copying it and so on.
I want to copy a complete line let say 10 cells in a row at single time. Is that possible ?

[MariusWirtz]

Please read the help article carefully and you will understand how to work with the different write functions. It is all explained and samples are provided for each function.
The write functions do NOT write "one cell at a time". If they do, you are doing something wrong in your code.