New Upload weather data from web services into TM1/Planning Analytics — Read the post →

Clearing cube data with TM1py

Often when interacting with TM1 through TM1py, it is necessary to clear out a slice of a cube. Since release 1.5 there are four options. Let's look at the pros and cons of each.

1. Calling a clear process from TM1py

Execute }bedrock.cube.data.clear using tm1.processes.execute_with_return:

Intuitive, but less flexible than MDX approaches, can't clear over alternate hierarchies, and can break if element names contain separation characters.

2. C spread on a consolidated cell

Use cells.clear_spread by referencing one element or consolidation for each dimension:

Intuitive but slower than all other options.

3. Clear with MDX

Define a slice with MDX and call cells.clear_with_mdx:

Maximum flexibility through MDX. Requires admin permissions, verbose, error prone (all dimensions must be referenced), and requires TM1 11.7 minimum.

4. Use the new clear method (TM1py 1.5+)

cells.clear enables clearing without MDX or referencing all dimensions. If a dimension is not referenced, all leaves will be cleared:

Maximum flexibility, intuitive, concise, and easy to use. Requires admin permissions and TM1 11.7 minimum.