The motivation comes from usage scenarios placed somewhere between shallow and deep parsing.
In some cases when we would like to bulid higher (more comprehensive) syntax tree for given sentence, it would be preferable to "disjoin" some groups to and make a step back to apply set of rules in altered order.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It seems to me it doesn't fit into Spejd domain (fast shallow parsing without any backtracking). Maybe you should do some postprocessing after the Spejd grammar and then use Spejd again with that different rules?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We haven't. Why do you need such an operation? Could you give a good, motivating example?
The motivation comes from usage scenarios placed somewhere between shallow and deep parsing.
In some cases when we would like to bulid higher (more comprehensive) syntax tree for given sentence, it would be preferable to "disjoin" some groups to and make a step back to apply set of rules in altered order.
It seems to me it doesn't fit into Spejd domain (fast shallow parsing without any backtracking). Maybe you should do some postprocessing after the Spejd grammar and then use Spejd again with that different rules?