Databricks getArgument value cannot be used within Spark SQL IN clause - apache-spark-sql

Question

I am getting a list of comma separated strings from a databricks notebook widget but I cannot use that value within Spark SQL's IN clause
I am within a %sql cell within the notebook and would like to NOT have to jump into %py or %scala cell type.
Also, I cannot use a select statement within the IN clause as the IN clause is NOT within a select filter. One gets an error: "IN/EXISTS predicate sub-queries can only be used in a Filter" if you try that syntax.
Here ITEM is a widget created like this in a %sql cell
CREATE WIDGET TEXT ITEM DEFAULT "'0111','0112'"
select *
from items_table
where ItemSKU in (string(getArgument("ITEM")));
or
select *
from items_table
where ItemSKU in (getArgument("ITEM"));
it does not return any rows
But if I directly substitute the value of the comma separated string in code, i.e.
select *
from items_table
where ItemSKU in ('0111','0112');
I get the expected result.

Answer 1

This is possible, try:
...
WHERE ItemSKU IN (SELECT explode(split(getArgument("ITEM"),',')))
This will also work for multi-select widgets.
The text widget option will be susceptible to user error if your users are allowed to enter text directly.

Answer 2

I don't think you can have multiple items as a default value of a multiselect widget, and definetely not a text widget.
If you look at the signature of the multiselect function
multiselect(name: String, defaultValue: String, choices: Seq, label: String)
You can see that the defaultValue only accepts a String while the choices is a Sequence. I know this is the scala function and you are lookign for the SQL version, but this implies that there is not meant to be a multiselect default option yet that actually has multiple selections.
Link - https://docs.databricks.com/user-guide/dev-tools/dbutils.html