PerlActor 0.02

PerlActor is a simple automated executable acceptance test framework for Perl.

PerlActor is a simple automated executable acceptance test framework for Perl. It can be used by Extreme Programming (XP) teams to rapidly develop executable acceptance tests for their Perl code.

XP mandates that the Customer writes acceptance tests for the application under development. The acceptance tests, also known as customer tests, demonstrate that application features are complete and that they work as expected. Ideally, the tests should be directly executable and automated so that they can be run continuously, without manual effort.

PerlActor enables an XP Customer to write *executable* acceptance tests as plain text files, using a very simple syntax. Each test consists of a number of parameterized commands (one per line), which PerlActor uses to exercise the application. PerlActor parses the test, invokes the commands with any parameters, and reports the result. The developers provide glue code to implement the commands required by the tests.

PerlActor allows tests to be grouped into suites so that related tests can be run as a group. The Customer can also just place test scripts in a directory structure and have PerlActor find and execute them all.

The PerlActor approach to acceptance testing has a number of advantages:

1) As the application grows the team will gradually develop a comprehensive set of Commands for testing the application.
2) The Customer, independent of the development team, can write and execute new tests for the application at any time, using any of the existing Commands.
3) Existing tests may be changed by the Customer at any time, again without developer help.

EXAMPLE SCRIPT

The following sample test script is adapted from one of the example scripts contained in the examples/calculator directory of this distribution. It tests a toy calculator "application":

# Script to check addition
# 10 + 32 = 42

# Create a new calculator application
NewCalculator

# Ensure that the display reads 0, initially
CheckDisplayReads 0

# Press key 1, then key 0
PressKeys 1 0
CheckDisplayReads 10

PressKeys +
CheckDisplayReads 10

PressKeys 3 2
CheckDisplayReads 32

PressKeys =
CheckDisplayReads 42

The following is sample output from the included test runner:

Running Acceptance Tests at Sun Apr 10 17:59:11 GMT 2005
==========================================================
.....
0.545077 wallclock secs ( 0.42 usr + 0.11 sys = 0.53 CPU)
Run: 5, Passed: 5, Failed: 0, Aborted: 0.
and with test failure:
Running Acceptance Tests at Sun Apr 10 18:04:36 GMT 2005
==========================================================
.F...
1.24811 wallclock secs ( 0.47 usr + 0.07 sys = 0.54 CPU)
Run: 5, Passed: 4, Failed: 1, Aborted: 0.

!!!FAILED!!!

1) FAILED: Display is wrong: expected 41, got 42 in scripts/test_addition.pact
at CheckDisplayReads 41, line 37