GitLab

This leads to slightly fewer questions in the main analysis (-10% on c.mly)
and many fewer nodes considered during backward search (4K instead of 60K on
c.mly), yet it is slightly slower (+10% on c.mly).

Generalized [Dijkstra] to return all distances.

Added [Dijkstra], homemade.
Modified [Coverage] to use [Dijkstra].
Also found a heavy bug where [causes_an_error] used [s] instead of [s']?

This actually speeds things up (almost 2x).

New performance counter [bs].

entry states.

triggering an error in this state.
This is about 10x faster for alphaCaml.mly. Still, alphaprolog.mly
does not go through.

Try to reach a pair (s', z) only if this pair causes an error.