Test Anything Protocol handling for cats
|Date:||Feb-Apr 2014, Jul 2018|
Test Anything Protocol handling for cats *rawwr*
Table of contents
taptaptap provides parsers, writers and APIs to handle the Test Anything Protocol (TAP). The implementation focuses on the most-current TAP version 13. TAP originates from the Perl community, but is a general format to document runs of testsuites. The reference to cats is just a pun for the noise of cats sneaking on floors.
taptaptap is only supposed to be working with python 2.7 (due to with statements and argparse). It fully supports unicode. It has been tested with Xubuntu 18.04 (Linux 4.15 x86_64) on 2.7.15rc1. A version for python 3.x is available as package taptaptap3.
|1.2.1:||declare encoding in setup.py, README updates|
|1.2.0:||Bugfix: do not drop memo in TapBailout, shebangs: “python” → “python2”, Bugfix: initialize next_number with 1, Bugfix: issue #1|
|1.1.3:||tapmerge: support merging more than 2 TAP files, support “-” to denote stdin|
|1.1.2:||TapWriter: do not reuse TapWriter instance in TapCreator|
|1.1.1:||TapWriter: support data handling in TapBailout|
|1.1.0:||procedural API writes to stderr not stdout, bugfix TapCreator: Fix number of testcase determination|
|1.0.5:||bugfix procedural API: write version even if default version|
|1.0.4:||more tests, fix output/source bug in testsuite|
|1.0.3:||install_requires in setup.py|
|1.0.0:||First stable release, packaging improvements, full testsuite|
|0.8.0:||Unstable release, minimal testsuite|
A basic introduction is given by Wikipedia. The format was specified by the Perl community.
taptaptap comes with a testsuite, which covers many special cases of the TAP format and tests the provided APIs. Please don’t hesitate to report any issues.
You can run the taptaptap testcases yourself using:
in the tests directory. The testsuite also shows some API usage examples, but I want to provide some here. The procedural API is well-suited if you are in the python REPL:
from taptaptap.proc import plan, ok, not_ok, out plan(tests=10) ok('Starting the robot') not_ok('Starting the engine') not_ok('Find the object', skip='Setup required') not_ok('Terminate', skip='Setup required') out()
The output looks like this:
1..10 ok - Starting the robot not ok - Starting the engine not ok - Find the object # SKIP Setup required not ok - Terminate # SKIP Setup required
Be aware that the state is stored within the module. This is not what you want if you are outside the REPL. The TapWriter class is more convenient in this case:
import taptaptap writer = taptaptap.TapWriter() writer.plan(1, 3) writer.ok('This testcase went fine') writer.ok('And another one') writer.ok('And also the last one')
If you like python’s generators, you want to use SimpleTapCreator:
@taptaptap.SimpleTapCreator def runTests(): yield True yield True yield False print runTests()
1..3 ok ok not ok
Or take a look at the more sophisticated TapCreator. If you are a real expert, you can use TapDocument directly, which covers all possibilities of TAP.
You can also invoke taptaptap directly from the command line:
python -m taptaptap.__main__ some_tap_file_to_validate.tap
This command will parse the file and write the file in a way how it was understood by the module. The exit code indicates its validity:
- Everything fine.
- The TAP file is missing some testcases or contains failed testcases.
- A bailout was raised. So the testing environment crashed during the run.
All objects are pickable.
Does taptaptap suite your needs? It does, if you are looking for a parser and validator for your TAP documents and you don’t want to care about details and just need a gentle API.
best regards, meisterluk