Library to convert python code to Control Flow Graph and back to code.
The conversion is done with an intermediary step of converting to an Abstract Syntax Tree:
Python -> AST -> CFG.
CFG is then being interpreted.
We need CFG because we want to do pattern matching on the code while it's being executed.
from agci import Interpreter
CODE = """
def main():
print(factorial(7))
def factorial(x):
if x == 1:
return x
return x * factorial(x - 1)
"""
inr = Interpreter(global_vars={
'print': print,
})
inr.load_code(CODE)
list(inr.run_main())import ast
from agci.sst import ast_to_sst
from agci.sst import sst_to_ast
CODE = """
def main():
print(factorial(7))
def factorial(x):
if x == 1:
return x
return x * factorial(x - 1)
"""
ast_graph = ast.parse(CODE)
cfg_graph = ast_to_sst.convert(ast_graph.body[0])
# And convert back
ast_graph = sst_to_ast.convert(cfg_graph)
code = ast.unparse(ast_graph)