If an opaque proof contained as Skip (i.e. a syntax error) it was not delegated anyway by the scehduler.
I'm rebasing #915 on top of this one since it makes is much cleaner.
@rtetley please have a look at the code for id_of_first/last_task. I think my new code is equivalent, but I don't get in which cases the default would be different than the computed one.
If I remember correctly, the opener and closer referred to the first proof step. This often lead us to not treat the Proof. or Qed. lines when coloring in delegation mode ?
If an opaque proof contained as Skip (i.e. a syntax error) it was not delegated anyway by the scehduler.
I'm rebasing #915 on top of this one since it makes is much cleaner.
@rtetley please have a look at the code for
id_of_first/last_task
. I think my new code is equivalent, but I don't get in which cases the default would be different than the computed one.