lpsbinary takes a linear process and replace parameters with
a finite data domain by booleans. The advantage of this is that some tools
are much better in dealing with Booleans than with arbitrary data types. This
for instance the case when binary decision diagrams are used (as used by jittyp).
As an example consider the following linear process:
sort D = struct d1 | d2 | d3; act a: Nat # D # Bool; proc P(n_P: Nat, d_P: D, b_P: Bool) = sum e_P: D. a(n_P, d_P, b_P) . P(n_P = n_P + 1, d_P = e_P, b_P = !b_P) + delta; init P(0, d1, true);
lpsbinary the sort
D is replaced by two booleans. The sort Bool
as well as sorts of which it cannot be determined that they are finite are not changed.
The result is given by:
sort D = struct d1 | d2 | d3; act a: Nat # D # Bool; proc P(n_P: Nat, d_P1,d_P2,b_P: Bool) = sum e_P: D. a(n_P, if(d_P2, if(d_P1, d1, d1), if(d_P1, d2, d3)), b_P) . P(n_P = n_P + 1, d_P1 = e_P == d2, d_P2 = e_P == d1, b_P = !b_P) + delta; init P(0, d1 == d2, d1 == d1, true);
lpsbinary [OPTION]... [INFILE [OUTFILE]]
Replace finite sort variables by vectors of boolean variables in the LPS in INFILE and write the result to OUTFILE. If INFILE is not present, stdin is used. If OUTFILE is not present, stdout is used
Command line options
limit enumeration of quantifiers to NUM iterations. (Default NUM=1000, NUM=0 for unlimited).
use rewrite strategy NAME:
compiled jitty rewriting
jitty rewriting with prover
select parameters that need to be replaced. All selected parameters must have a finite sort other than Bool. Examples: –select=a:,b:Data or –select=:Enum4.
append timing measurements to FILE. Measurements are written to standard error if no FILE is provided
do not display warning messages
display short intermediate messages
display detailed intermediate messages
display intermediate messages up to and including level
display help information
display version information
display help information, including hidden and experimental options