Skip to content

J-A-I-L/TDD

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

TDD

Test Driven Designing in Python 3

Why did you end up making this

I made this tdd app during a Coursera bioinformatics course. You can find most (if not all) the course's problens in Rosalind. Problem's input and output always had the same format: a string. You can ckeck this if you open a Rosalind problem and look its specification, with the sample dataset, extra dataset and debug datasets.

Sample input and output.png

So, as I already had all the datasets before starting programming, I could see as a Test Driven Development. I decided to fully automate the process, so I made a database with a table. Each entry has 3 fields: method name, input data and output data. I made an app that opens the database, and for each entry reads the name of the method, creates a call to that method using reflection, passes the data input as a parameter and gets its response. Then, compares the actual response with the expected output in the database.

sample io in sqlite database.png

Well, actually it's a bit more complicated: the method is not actually called by my app. Instead, it creates a test suite using PyUnit (Python's equivalent to JUnit), and creates one test case for every method in the database. But you get the idea.

That's enough drag. I wanna see it working!

So... Have you already tried to run it!?

no module named foo.png

Ok, if you do so, you'll get an error saying "No module named 'foo'". That's because the app run in a module called test_foo.py expects to find the methods to test in foo.py. More generically, test_<module>.py looks for methods in <module>.py.

No database found.png

Now an SQLite database has been created. It has the name foo.sqlite3, or generically <module>.sqlite3. It's a database with a single table. There you can add the name of the method to test, and its input and expected output. Don't worry if you haven't already written it in <module>.py: you can set a 0 in the perform_test column, or even if it is set to 1 and the method doesn't exist, a PyUnit pragma is used specifiying the method is not implemented yet.

So, now you can rename test_foo.py to your test_<module>.py and run it once to create the database. Then you can start adding tests into the database, and writing your methods. Run test_<module>.py whenever you want to pass the tests.

Footnote

I did this development using Eclipse's plugin for Python and PyUnit. I'm currently trying to make this piece of code independent from Eclipse. Meanwile, it'll show error when running from plain console.

About

Test Driven Designing in Python 3

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages