Idea: Piping

[FeignDeath]

In regard to the request for preprocessing, I wanted to ask whether it is possible to have multiple encodings solve the problem in a piped manner.

For example:
1.lp has #show cat/3.
then 2.lp would get all atoms passed via show statements and continue computing.

Use case:
I might be interested in researching partitioning of the flatland environment. To that end I would run an encoding partitioning.lp and afterwards the correct encoding which takes the partitioning into account. Since partitioning would be an optimization problem, it would need to have its results piped so following encodings don't ground for all possible partitionings.

Example

1. env.lp + partitioning.lp (for example with 2 partitions)
2. 1_result.lp + partition1.lp
3. 2_result.lp + partition2.lp

I just wanted to ask whether you think that would be possible, before I think about it too much.