Spatial and Action Based Resolver
Pull request Compare This branch is 109 commits behind dbunker:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


SABR (Spatial and Action Based Resolver) 
------------ GENERAL ------------

SABR (Spatial and Action Based Resolver) is a constraint programming 
language designed with an emphasis on spatial and temporal logic. 

Programs compile to CNF (Conjunctive Normal Form), which is then solved
by a CNF solver, whose results are processed into human readable form. 

The goal of the language is to make representing puzzles and operations 
research problems not just possible, but also intuitive and the solving 
of these problems efficient. 

From a theoretical perspective, the language can also be used to prove 
a problem is NP-Easy and can be used to verify properties about automata 
and systems that can be modeled as automata such as circuits and programs.


------------ INSTALL ------------

SABR is installed using

python depend
	This command will create a dependency diagram using a dot file 
	indicating which files primarilly use which other files and how 
	the program progresses and place it in doc/depend.

python minisat
	This will install the default cnf solver used SABR. The default 
	cnf solver is minisat. More information can be found at		

python parser
	This will use flex and bison to create the c files from the lex 
	and yacc files in the parser directory.
python build
	This will build SABR from core utils and parser.

python clean
	This will clear all unnesary files in the SABR directory.

python cleanall
	This will also clear minisat installation.

------------ RUN ------------

To run SABR, you must provide the number of stages as the first argument. 
An optional type of compilation flag. Then input the SABR program to run.
The output file will appear as result.out.
For more information on the language view doc/LANGUAGE.

./sabr 20 -all < test/Simple/simple.tb


	No flag will produce the for the cnf and execute and 
	process it to human readable form.

	This will only produce the After this, a more advanced 
	sat solver could be used.

	This will be like initial, but will output debug information.

	This will assume dimout.out will come from another cnf solver.

	This will run like with no flag, but also output debug results.

------------ TESTING ------------

Testing is done with
-full can be added where indicated to fill in debug results

python debug
	Tests compiler post-process syntax checking.

python simple
python simple-full
	Does simple tests for compiler functionality. 

python adv
python adv-full
	Does advanced tests for compiler functionality. 

python hard
python hard-full
	Does difficult tests for compiler functionality. 

python all
python all-full
	Does all tests for compiler functionality. 

python clear-debug
	Clears the debug results for all tests (reduces project size).