https://github.com/thefangbear/sat
Boolean expression parser and solver suite
https://github.com/thefangbear/sat
c cpp parser sat solver
Last synced: 4 months ago
JSON representation
Boolean expression parser and solver suite
- Host: GitHub
- URL: https://github.com/thefangbear/sat
- Owner: thefangbear
- License: mit
- Created: 2018-06-17T08:17:00.000Z (almost 8 years ago)
- Default Branch: master
- Last Pushed: 2018-06-22T10:31:30.000Z (almost 8 years ago)
- Last Synced: 2024-12-29T07:24:07.599Z (over 1 year ago)
- Topics: c, cpp, parser, sat, solver
- Language: C++
- Size: 118 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# SAT
This is an educational repository aims at providing a set of very basic routines for parsing and solving simple boolean expressions in C/C++ and for people who want to implement a very very basic SAT solver for fun.
## Parser
For simplicity, we parse using a very crude grammar that works most of the time (excuse the sloppiness...).
The parser is a simple recursive descent parser that parses the input expression to (approximately) a binary expression tree.
Parenthesization is handled as having the binary expression tree link to sub-trees using an extra pointer on each level, if necessary.
To avoid writing a full LL(*k*)-parser we simply define the grammar to recurse at every left-parenthesis (`(`), and return
at every right-parenthesis (`)`).
[Click here to view the very crude parser with no parenthesization support](https://github.com/thefangbear/SAT/blob/master/sat_parser.cpp);
[Click here to view the parser with parenthesization + eval support](https://github.com/thefangbear/SAT/blob/master/parser.cpp);
[Click here to view the parser template file for SAT solver implementation](https://github.com/thefangbear/SAT/blob/master/constraint_parser.cpp)
## Current solvers
We currently have three kinds of solvers, all implementing the basic binary backtrack method.
**Binary-recursive Backtrack Solver**: ([basic_sat.cpp](https://github.com/thefangbear/SAT/blob/master/basic_sat.cpp)) The solver uses the basic binary-recursive backtracking algorithm. Running time is O(N 2^k).
**Binary-iterative Backtrack Solver**: ([sat.cpp](https://github.com/thefangbear/SAT/blob/master/sat.cpp)) This solver uses an iterative approach to generate all 2^k solutions down the tree.
**Binary Stack-recursive Backtrack Solver**: ([iterative_basic_sat.cpp](https://github.com/thefangbear/SAT/blob/master/iterative_basic_sat.cpp)) This solver uses a stack to explicitly handle the recursion.
## Development status
Currently testing and revising the parser for boolean expressions.
- Step 1: Write a parser that supports parsing non-parenthesized boolean expressions (done, `sat_parser.cpp`)
- Step 2: Write a parser that supports parenthesized boolean expressions (done, `parser.cpp`)
- Step 3: Write a parser that evaluates boolean expressions (done, `parser.cpp`)
- Step 4: Write a parser that handles free-formed expressions (done, `constraint_parser.cpp`)
- Step 5: Implement a basic backtracking-based SAT solving algorithm (done, `basic_sat.cpp`)
- Step 5.5: Implement an iterative version of `_backtrack()` in `basic_sat.cpp` to make it prettier (done, `sat.cpp`)
- Step 6: Implement a more advanced SAT solving algorithm (DPLL; this requires us to write a CNF converter first; see `cnf.txt`)
## Example Usage
Enter the expected (maximum) length of the input expression on the first line, and the expression on the second line.
```
50
a&(b|c^d)
input: a&(b|c^d)
S(): check
S(): parsing B()
B(): @a
S(): parsing OP(), &
OP(): @&
S(): parsing S()
S(): check
S(): parsing B()
B(): @(
S(): check
S(): parsing B()
B(): @b
S(): parsing OP(), |
OP(): @|
S(): parsing S()
S(): check
S(): parsing B()
B(): @c
S(): parsing OP(), ^
OP(): @^
S(): parsing S()
S(): check
S(): parsing B()
B(): @d
T(): return
B(): S() returned, **prog=='\0'? 1 [**prog=0]
T(): return
(& a ((| b (^ c d))))
=== solving ===
solve: got 4 variables, solving [16 possibilities]...
:::::: Viable solution :::::
Variable [a] := T
Variable [b] := T
Variable [c] := F
Variable [d] := F
............................
:::::: Viable solution :::::
Variable [a] := T
Variable [b] := F
Variable [c] := T
Variable [d] := F
............................
:::::: Viable solution :::::
Variable [a] := T
Variable [b] := T
Variable [c] := T
Variable [d] := F
............................
:::::: Viable solution :::::
Variable [a] := T
Variable [b] := F
Variable [c] := F
Variable [d] := T
............................
```
## Author
Rui-Jie Fang
## License
The MIT License.