Unit Tests

[ndanielsen]

Create a file called tests.py

Using your test data in for 'dr jenkyll' and frankenstein, create several tests that show that you are accurately extracting data from the html files.

Test 1: That you can accurately extract the title from an html file assertequals(title, "The Project Gutenberg E-text of Frankenstein, by Mary Wollstonecraft (Godwin) Shelley")

Other tests: Using a shorter amount of text that you can manually verify, write tests for all your functions in analyze/gender.

Extra credit:

Can you figure out how to create a BaseClass for unittests with a setup function so you don't have to repeat yourself and violate DRY.

Read up on creating unit tests, and install 'nose' and read up on it's use. http://docs.python-guide.org/en/latest/writing/tests/

Watch this video for a great rationale: https://www.youtube.com/watch?v=FxSsnHeWQBY

[ndanielsen]

For the json_engine:

Bonus, can you use the setup and teardown to create and delete test json files for testing the engine functionalities?

[ndanielsen]

examples of classes for a recommendation engine and flask here:

https://github.com/ndanielsen/canimakeit-app/tree/deploy

[skasim]

i created the unit tests for the analyze/gender tests. This was super helpful because I found that the code I had written earlier was not actually removing punctuation, so it wasn't correctly counting all the gendered words. I wasn't able to do a test for the title because the Gutenberg site updated their security and I can't scarpe it anymore, but I did run separate tests just to make sure I am able to correctly pass book numbers to the json engine to get the json dicts.

I haven't written a class yet and that is the next thing I want to tackle.

[ndanielsen]

During our session, we can briefly review how to write unit tests. Here's the python docs page for it : https://docs.python.org/2/library/unittest.html