Learning chaining

[Jimmy]

I'm struggling with understanding how to create a chaining syntax like this with matrix in Cerberus-X, is it possible?

landsca = AffineMat3f().Translate(p).Rotate(r).Scale(s,s)

 

[Holzchopf]

If all the functions AffineMat3f(), Translate(p), Rotate(r), Scale(s,s) return an object of the same type, then that's possible. The chained methods act on the returned value from the previous function/method, so whatever they return must have these methods.

That's the reason why often functions that mutate the input object also return it (which might seem like an unnecessary redundancy, but only like this, chaining is possible)

 

[Jimmy]

Thanks! Okay so as long as the input and output is the same you can do this for as long as you want?
Here the data is just a normal transform array [ix,iy,jx,jy,tx,ty] so I guess they each fill in / change whatever they need to in the chain?