Good software practices say you should do automated testing, but what do you do when you work in an environment that can’t get good test tools? Use a simple testing protocol like TAP.
TAP is the Test Anything Protocol. The protocol is so simple that you could probably write code to make TAP output in minutes. It is a line based protocol which means that any language that has some sort of print method can generate TAP. That’s basically all software languages in common use.
I could provide the rules, but an example would be easier.
# This is a comment because it starts with a hash symbol. ok 1 Something passed here. not ok 2 Whoops, there was a failure. When there is a line that TAP does not understand (like this one), it will skip it. # TAP files should have a plan in the format of "1..<some number of tests>". 1..2
The key to the protocol is
not ok. When lines start with one of
these two phrases, TAP readers will interpret it as a pass or failure. Besides
the test number, everything else on the line is extra. Toss in how many tests
were run and you have fully functional TAP output. We can prove it with
Perl’s aptly name
matt@eden:~$ prove sample.tap sample.tap .. Failed 1/2 subtests Test Summary Report ------------------- sample.tap (Wstat: 0 Tests: 2 Failed: 1) Failed test: 2 Files=1, Tests=2, 0 wallclock secs ( 0.01 usr + 0.00 sys = 0.01 CPU) Result: FAIL
There are some additional features that I did not discuss here (like TODO tracking). See the TAP website for more details. More adventurous individuals can check out the specifics by looking at the de facto parser, TAP::Parser, on Perl’s CPAN.
Learn more about Python and other software topics
You can join my newsletter and get access to other useful topics to help you learn more about software development. I hope you'll join up!