Description
The purpose of this project is to help you to become familiar with Python 2.7.2, and with Bison, a parser
generator. All input must be read by the flex scanner that has been provided for you, and all cyclomatic
complexity computation must be computed in the provided Python 2.7.2 parser, hereafter referred to as
MyPy, with possible modifications to the corresponding flex scanner. Thus, an added benefit of this project
is that it will provide a gentle introduction to MyPy, which you will be using for the rest of the semester.
The project entails computing cyclomatic complexity for each function in a Python program; include
Python constructs that are back ported from 3x into 2.7.2. Print the complexity measure as a three tuple
where the items in the tuple are: (1) the line number of the beginning of the function; (2) the name of the
function, quoted to indicate that it’s a string; and (3) the cyclomatic complexity of the function.
Your oracle for the project is the mccabe flake module in Python, and your McCabe numbers should
exactly match those provided by the flake module.
You can install the flake mccabe module using one of the following commands:
$ pip install mccabe
$ pip install –user mccabe
And you can execute the module on a Python program called main.py as follows:
$ python -m mccabe main.py
$ python -m mccabe –min 5 main.py
1
Not All Equal:
Cyclomatic complexity tools will typically report different results and sometimes these results are drastically different. To compare complexity results you might consider installing radon and executing it on
some python test cases:
pip install radon
radon cc -s main.py
Methodology:
Your approach to complexity computation must entail using the scanner and parser in MyPy, which has
been placed in the project 3 folder of the course repository. you cannot use any other tool, such as grep or
Python itself to compute complexity. To accomplish the computation, you should study the productions in
parse.y of MyPy and try to find the ones that will be involved in McCabe complexity computation.
Testing: An advantage in using MyPy is that it will accept all valid Python programs, which will make
it easier for you to write test cases: you can validate them against any Python 2.7.2 parser and compare
the results with your parser. Also, by matching your output to the output of the flake module and using the
provided test script, test.py, it will also be easy for you to write test cases.
Two test scripts have been provided in the parser directory of the project repository:
• alltest.py – if you run this test script it will execute the test cases one directory away, copied from the
Python development web page, with 1537 test cases Passed, and 3 test cases Failed.
• test.py – this test script will compile your program if necessary, run the test cases in the directory
cases, and compare the results from your program with the results in the file with prefix the same
as the test case and suffix of .out. For example, a test case, x.py is provided in the cases directory,
with corresponding expected output in x.out. You must augment the test case in cases with at least 3
additional test cases that adequately test your complexity tool.
List of Tasks:
1. Modify the scanner and parser in MyPy in the project 3 directory to build a tool to compute McCabe’s
complexity. The metrics result should exactly match those provided by the flake8 mccabe module.
2. Add test cases to the directory cases to adequately test your tool.
Web Pages:
The mccabe module for python from flake8:
https://pypi.python.org/pypi/mccabe
The radon module:
https://radon.readthedocs.io/en/latest/intro.html#
https://radon.readthedocs.io/en/latest/commandline.html
2