Disable QCheck2 shrinker with a parameter about qcheck HOT 3 CLOSED

I think I can extract one example pretty easily. One co-worker wrote a Merkle Tree or something, it shouldn't have too much dependency with our project.

from qcheck.

On the one hand I'm supportive of this feature.
I've used set_shrink Shrink.nil some_gen in similar situations for "QCheck1" in the past.
One could go for a similar design for QCheck2 where the operation changes/removes the Tree children.

On the other hand, to me this indicates short-comings of the QCheck2 shrinkers.
I observed these and commented on them in #157 #163 and #153 (comment) while I was trying to test QCheck2 a bit.
We need example cases to reproduce and understand what is going on when shrinking takes too long in order to improve the situation. I understand that carving out those from a bigger project can require quite some effort though.

from qcheck.

I think I can extract one example pretty easily. One co-worker wrote a Merkle Tree or something, it shouldn't have too much dependency with our project.

Well, you shouldn't trust my words.

Anyway, I feel that we need it even if the generator is not that long. It's very situational but sometimes, no shrinker is the best solution.
My dirty solution currently is:

  make_primitive
    ~gen:(fun rand -> generate1 ~rand gen)
    ~shrink:(fun _ -> Seq.empty)

I'll try to tackle this.

from qcheck.