For creating unions, it's often really important to make slight modifications to columns. This is especially true for types, but also for basic data transformations. Many recipes have the option to create computed columns, but it's missing from the stack recipe. It'd be great to add this in. Optimally, it could optionally only execute for those datasets where a used column is present. This will speed up data blending a lot.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.