[Specification]  [Database]  [Examples]  [Testing]  [Submitting]

This document contains details of how you/we test your scripts.

We provide a testing script check and a set of test files to run tests on the same set of examples in the Examples page. To obtain the tests, first ensure that you are in the directory containing your dirby, releases, etc. scripts. Make sure that this directory does not contain files or directories called check. Then run the command:

unzip ~cs3311/web/22T3/assignments/ass2/testing/testing.zip

This adds a shell script called check into your directory along with a new subdirectory called tests, which contains (surprise!) testing files (e.g. expected outputs). The tests directory contains a file called tests that has a list of all the test being conducted. Note that some of the movie tests require us to push a number into the script via a pipe.

To run the tests, use the following command on a machine where a PostgreSQL server is running:

sh check

which will hopefully produce

Test 01 PASSED
Test 02 PASSED
Test 03 PASSED
Test 04 PASSED
Test 05 PASSED
... etc etc ...

If it fails all the tests make sure that your PostgreSQL server is running. Also, make sure the helpers.py file is present, if you're using it.

It works out whether a test passed by comparing an expected output file against the output from your script. Both the observed and expected output for each test are located in the tests directory.

The check script uses several different variants of the diff command to try to work out why your output is different to the expected. Generally, you will have to examine the differences between the expected and observed outputs (which you can find in the tests directory) to work out any problems.