clasp - 3.3.5-4.2 main

clasp is an answer set solver for (extended) normal logic
programs. It combines the high-level modeling capacities of answer
set programming (ASP) with state-of-the-art techniques from the area
of Boolean constraint solving. The primary clasp algorithm relies on
conflict-driven nogood learning, a technique that proved very
successful for satisfiability checking (SAT). Unlike other learning
ASP solvers, clasp does not rely on legacy software, such as a SAT
solver or any other existing ASP solver. Rather, clasp has been
genuinely developed for answer set solving based on conflict-driven
nogood learning. clasp can be applied as an ASP solver (on LPARSE
output format), as a SAT solver (on simplified DIMACS/CNF format), or
as a PB solver (on OPB format).