[Python] How to use a hidden key for cross-filtering across multiple charts in Observable Plot?

I'm working on a dashboard using Observable Plot and I want to implement cross-filtering across multiple bar charts. The issue I'm facing is that the key I want to use to filter across charts (e.g., KEY) should not appear on the x or y axes — I want it to be used internally just for linking/filtering data between plots. here is the code I have tried:

params:
brush_main: { select: crossfilter }

- plot:
- mark: rectY
data: { from: Data, filterBy: $brush_main }
x: Yearmonth
y: { sum: Purchase_order_value_local }
...
- select: intervalX
x: Yearmonth
key: D08_ZF_KY_TEST


My problem is:

I don’t want to use Yearmonth or Supplier as the key to connect the plots.

Instead, I want to use a hidden column (KEY) that exists in all datasets as the internal crossfilter key.

When I try this, I get errors like:

Error: Unrecognized param type: intervalX
TypeError: Cannot destructure property 'type' of 'scaleTransform(...)' as it is null.



My question is:

How can I define a selection/filter (like intervalX or crossfilter) that operates on a hidden key (not rendered in x or y axes) to enable cross-filtering across multiple charts?

Any ideas or workarounds would be greatly appreciated!

Thanks in advance

Continue reading...