The latest version of PyChecker is 0.8.19

PyChecker is a tool for finding bugs in python source code. It finds problems that are typically caught by a compiler for less dynamic languages, like C and C++. It is similar to lint. Because of the dynamic nature of python, some warnings may be incorrect; however, spurious warnings should be fairly infrequent.

PyChecker works in a combination of ways. First, it imports each module. If there is an import error, the module cannot be processed. The import provides some basic information about the module. The code for each function, class, and method is checked for possible problems.

Types of problems that can be found include:

  • No global found (e.g., using a module without importing it)
  • Passing the wrong number of parameters to functions/methods/constructors
  • Passing the wrong number of parameters to builtin functions & methods
  • Using format strings that don't match arguments
  • Using class methods and attributes that don't exist
  • Changing signature when overriding a method
  • Redefining a function/class/method in the same scope
  • Using a variable before setting it
  • self is not the first parameter defined for a method
  • Unused globals and locals (module or variable)
  • Unused function/method arguments (can ignore self)
  • No doc strings in modules, classes, functions, and methods

Here's an article about PyChecker in Unix Review by Cameron Laird and Kathryn Soraiz.

Using PyChecker

To use PyChecker, pass options and the python source files (or packages) you want to check on the command line:

	pychecker [options] file1.py file2.py ...

Some of the most commonly used options are:
OptionsDescriptionDefault value
--only only warn about files passed on the command line no
-#, --limit the maximum number of warnings to be displayed 10
--no-shadowbuiltin check if a variable shadows a builtin off
-q, --stdlib ignore warnings from files under standard library off
-T, --argsused unused method/function arguments on

Note: On Windows, use pychecker.bat. You may also need to add python/scripts to your PATH.

pychecker and pychecker.bat will only exist if pychecker has been installed. To install, do: python setup.py install

Note: If you haven't installed pychecker, it can be run by doing: python pychecker/checker.py

An alternate way to use PyChecker is to import it in your code. See Importing PyChecker below for more details.

If there are import dependencies in your source files, you should import those files first on the command line in order to get as many files checked as possible.

PyChecker works with Python 2.0 through 2.7. Some features don't work on earlier versions of Python. PyChecker is tested with Python 2.2 through 2.7 using buildbot.

You can use the test files as examples:

	pychecker test_input/*.py

If you want to change the default behaviour, you can pass command line options or define a .pycheckrc file. For an example, look at pycheckrc.

	pychecker -h
will show the available options.

There is a simple GUI which is not maintained much. It is good for showing all the options and also allows you to run pychecker. To run options, you will need to start it manually:

	python pychecker/options.py

If you want to suppress warnings on a module/function/class/method, you can define a suppressions dictionary in .pycheckrc. Examples of keys are: 'module', 'module.function', 'module.class', 'module.class.method', etc.

You can also define suppressions in your code by doing:

        __pychecker__ = 'no-namedargs maxreturns=0 unusednames=foo,bar'

The format for __pychecker__ values and values in the suppressions dictionary are the same. Dashes (--) are optional when preceding long option names.

Importing PyChecker

You can import PyChecker in your code's main module, by doing:

        import pychecker.checker

This will allow each module imported after PyChecker to be checked (other than the main module). NOTE: Modules imported before PyChecker will not be checked. Warnings will be displayed on stdout (ie, PyChecker uses print).

Since you can't pass command line parameters, you can do:

        os.environ['PYCHECKER'] = 'command line options here'

This is equivalent of setting PYCHECKER in the shell environment:

        PYCHECKER='no-namedargs maxreturns=0' /path/to/your/program

If you want to disable the warnings (and processing done by PyChecker), prior to importing PyChecker, do:

        os.environ['PYCHECKER_DISABLED'] = 1

This is equivalent of setting PYCHECKER_DISABLED in the shell environment:

        PYCHECKER_DISABLED=1 /path/to/your/program

Internal Errors

If you find a bug in PyChecker, meaning you see something like:

	pychecker myfile.py

	myfile.py:13 INTERNAL ERROR -- STOPPED PROCESSING FUNCTION --
        Traceback (most recent call last):
          File "./pychecker/warn.py", line 364, in _checkFunction
            stack, oparg, lastLineNum)
          File "./pychecker/warn.py", line 195, in _handleFunctionCall
            kwArgs.append(stack[i].data)
        IndexError: list index out of range
Please post a bug in the SourceForge Tracker or send mail indicating the version of PyChecker, your source file which broke PyChecker (myfile.py in the example above), and the traceback. It is very helpful to provide a simple test case to demonstrate the problem. It helps to have the entire file and all the dependencies if you cannot produce a simple test case. But if you can't provide a test case nor the file(s), I may be able to figure out the problem with just the line which broke PyChecker (myfile.py:13 in the example above).

Buildbot

Pychecker is tested on each commit using Buildbot. See http://build.fluendo.com:8200/

IRC

Feel free to ask questions on #pychecker on irc.freenode.org Our friendly buildbot is there too.

Projects using PyChecker

Pychecker is regularly run on the following projects: Before each release these projects get checked to test that PyChecker works. If your project uses PyChecker too, let us know so we can add it here and to our release checklist.

Here are some quick links for the latest released version of PyChecker:

The development version is usually quite stable. Here are links to track the development status:

The project page for PyChecker can be found on SourceForge.

Good Luck! As always, feedback is greatly appreciated.

Neal

pychecker-list@lists.sourceforge.net