Before the Lab

Dive into Python 3, Chapter 1

Discussion

Time: 5 min =Activity= Group Discussion

Questions about the quiz?
Questions about A4

A first example

Time: 10 min
Activity: Demo

Background for this section is DiP3 1.1
Download humansize.py and save it as ~fcshome/cs2613/labs/L14/humansize.py
Run it from the command line
```
  $ python humansize.py
```
Run the debugger
```
  $ pudb3 humansize.py
```

Use Ctrl-X to switch to the bottom debugger window (Command line) and run

  >>> from humansize import approximate_size
  >>> approximate_size(10000)
  >>> approximate_size(10000,False)

Repeat the previous step by running in a terminal
```
  $ python
```
and then typing in the same lines.

Pytest

Time: 20 min
Activity: individual

In this part of the course we will be using pytest to write unit tests.

download an initial test file and save it as ~fcshome/cs2613/labs/L14/test_humansize.py
open a terminal
```
  $ pytest-3 test_humansize.py
```
Convert each working example in DiP3 1.2.1 into a test.
Test coverage reports can be gotten with
```
  pytest-3 --cov=mymodule --cov-report=term-missing
```
where mymodule is the name of the module in question. Most likely you will need to add more tests for complete coverage. You can do this after the lab.

Modules

Time: 15 min
Activity: individual

Create a new file ~fcshome/cs2613/labs/L14/client.py
Import the module humansize in client.py
Define a new function approximate_size that calls the function approximate_size from the humansize module, but has a default of False for the parameter a kilobyte_is_1024_bytes
Observe that the code in humansize.py guarded by if __name__ =='__main__' does not run when imported into client.py (you might see echos of Racket submodules from Lab 5). Create a similar block in client.py, and run it from the command line.

More testing, docstrings

Time: 15 min
Activity: individual

create test_client.py by copying and modifying (if necessary) test_humansize.py
from the command line, run
```
  $ pytest-3
```
add a test to test_client.py to ensure that your new function has a docstring
make sure your new function has a docstring (i.e. that the test you just added passes) and that it makes sense.

Indentation

Time: 15 min
Activity: individual

One initially surprising aspect of Python is it's use of indentation to define blocks
Start a new file ~/fcshome/cs2613/labs/L14/fizzbuzz.by. Add the following code, and fix the indentation so that it runs

    for i in range(1,101):
    if (i%3 == 0 and i%5 == 0):
    print("FizzBuzz")
    elif (i%5==0):
    print("Buzz")
    else:
    print(i)

Exceptions

Time: 20 min
Activity: individual

Python throws exceptions when dividing by zero. Suppose we want to return NaN, JavaScript style.
Start with the following code as ~/fcshome/cs2613/labs/L14/divisive.py.

    def fraction(a,b):
        return a/b;

Add a try/except block to fraction so that the following test-suite (saved as test_divisive.py) passes

from divisive import fraction
import math

def test_fraction_int():
    assert fraction(4,2) == 2;

def test_fraction_NaN():
    assert math.isnan(fraction(4,0))

Hint: you can use float('nan') or math.nan to generate a NaN

On your own

Lookup up the definition of the FizzBuzz problem and add the missing case for "Fizz" to the program discussed above.
Add enough tests to get complete test coverage for humansize.py