PyModel Tester

PyModel is an open-source model-based testing framework in Python.


Usage:

  pmt [options] models

models is a list of one or more module names (with no .py suffix). Each module named in models must contain a model, that is: a model program, an FSM, or a test suite. In addition, a module named in models can contain additional configuration information for a model program. Multiple models (model programs including configuration, FSMs, and test suites) are composed into a product. The tester generates traces by executing the product, at random or guided by the strategy option. To just view the traces, omit the implementation option. To use pmt as an offline test generator, omit the implementation option and use the output option to save the traces in a test suite module. To execute tests, provide the implementation option. To execute tests generated offline, provide that test suite module as the single command argument. To generate and execute tests on-the-fly, provide one or more model programs and/or scenario FSMs as command line arguments. To save on-the-fly tests so they can be reproduced later, use the output option to save them in a test suite module.

Limitations: pmt can only execute traces consisting of controllable actions that the tester can call to make the implementation do something. Future versions will also support observable actions: events that the tester can detect in the implementation.

Options:

  -h, --help            show this help message and exit
  -a ACTION, --action=ACTION
                        Action to execute, as many as needed, if no -a execute
                        all actions
  -c CLEANUPSTEPS, --cleanupSteps=CLEANUPSTEPS
                        Maximum number of steps in the cleanup phase, default
                        0 for no cleanup
  -d, --debug           Run the main function in the debugger
  -e EXCLUDE, --exclude=EXCLUDE
                        Action to exclude from execution, as many as needed
  -g STRATEGY, --strategy=STRATEGY
                        Strategy (module name), omit to use random strategy
  -i IUT, --iut=IUT     Implementation (stepper module name), omit to just run
                        model
  -n NSTEPS, --nsteps=NSTEPS
                        Number of steps in a single test run before beginning
                        cleanup phase, default 0 for no limit
  -o OUTPUT, --output=OUTPUT
                        Output test suite module name (with no .py suffix), no
                        default (no output file if omitted)
  -q, --quiet           Omit arguments and return value from progress messages
                        (to suppress printing very large buffers etc.)
  -r NRUNS, --nruns=NRUNS
                        Number of test runs, default 1, 0 for no limit
  -s SEED, --seed=SEED  Random seed, use any nonzero value to make runs
                        reproducible
  -t SEC --timeout=SEC  Number of seconds to wait for stepper to return from 
                        action before timing out and failing test, 
                        default 0 (no timeout, wait forever).
                        (Might not work on Windows)

Revised April 2011