run task function with different parameter

[Jerry-Ma]

Hi,
I am trying to achieving the following work flow in an concise way but failed:

  collate_func (    (a.i.1, a.ii.1) , p1   ) -> a.p1.1   
               (    (a.i.2, a.ii.2) , p1  ) -> a.p1.2
               (    (a.i.1, a.ii.1) , p2  ) -> b.p2.1   
               (    (a.i.2, a.ii.2) , p2  ) -> b.p2.2

Basically, I am trying to run a collation on a.(id).? to get to a.?, using collate, but collate_func takes an additional argument p, being some sort of parameter (python objects). In my example above, I want the collation be done using two different parameters, p1 and p2, so in total four jobs will be created, and the supposedly outputs being a.(param).?, instead of a.?

My current approach is to define two tasks using two variations (partial) of the collate_func. I also looked product but it seems only woks for real files, not python objects. I can imagine that parallel works, but that will make the files being not managed as collate do. I also can see that subdivide being useful in this case, (and maybe it is the best way to go, ) but I am not sure how to run collate and subdivide at the same time.

So is there any suggestions?