python-style-qa.md

Line Lengths

Question: I think the only thing that really confused me was the line length limit. I know python to be a language where long one liners are fairly common, so I'm shocked to see a line limit enforced, and I still don't really understand why.

Answer: Line length limits are programming language agnostic.

Consider this line in C++:

if (process_vector(sum_vectors(create_new_vector(0, 1, 0), create_new_vector(1, 1, 0))) != 0 && is_valid_input(user_string) && as_int(user_string) > 5 && as_int(user_string) < 20) {

This line is simply too long to be legible! It wraps unpleasantly in IDEs and text editors, or requires scrolling to the right. It should be broken up simply to make it more legible:

vector3 up_vector = create_new_vector(0, 1, 0);
vector3 up_right_vector = create_new_vector(1, 1, 0);
vector3 summed_vector = sum_vectors(up_vector, up_right_vector);
int process_result = process_vector(summed_vector);
if (process_result != 0
	&& is_valid_input(user_string)
	&& as_int(user_string) > 5
	&& as_int(user_string) < 20)
{

The same applies to Python and to all programming languages and text files.

Calling Main

Question: Something I am still confused about is why we do if name == "main": main() to call main.

Answer: Python scripts simply run top-to-bottom when executed or imported. Say you write a python file named lists.py. It contains useful fuctions like build_list() and sort_list(). When run, it uses build_list() to get input from the user and sort_list() to send the input to a server for sorting. It looks something like this:

def build_list():
	... code here ...

def sort_list(l):
	... code here ...

print(sort_list(build_list()))

Now you want to write another Python file that also wants to deal with lists. It would be handy to import the build_list() and sort_list() functions so you don't have to write them again. So, you write this import statement at the top of your new file:

import lists

When running it, it asks the user for list input and sends it to a server for sorting. Oops! That's because the import statement runs the entire target file.

To get the best of both worlds, where Python files can be imported or run and behave appropriately in both situations, we change lists.py slightly:

def build_list():
	... same as before ...

def sort_list(l):
	... same as before ...

if __name__ == '__main__':
	# this code will not run when imported!
	print(sort_list(build_list()))

Now executing lists.py still does we wanted, but importing lists.py only imports the functions without also executing the main code of the file.

Docstrings

Question: I am a bit confused about why docstrings are the first statement inside a function/class. In pretty much every other language I've seen, it seems like the convention is to have it before the object/function/statement it describes. It seems like a multi-line docstring would make a function less readable since it splits up the function name and arguments from the rest of the code.

Answer: Docstrings are actually a part of the Python interpreter! You can use the built-in help() function to read the docstring of any function or module that has one. Putting the docstring after the function name makes it unambiguous to parse, whereas a docstring before a function could just be an errant multi-line comment.

Problems Caused by Defying Style

Question: While I know we need to adhere to the style guide for the labs and assignments, how much of this would cause problems when trying to run code if we didn't follow the style guide? Are certain conventions used to prevent problems that only appear sometimes but not all the time?

Answer: This will not cause any compilation errors, but since all python projects use this style guide, collaboration (which is how almost all code is written) will become difficult.

The style is meant to reduce bugs, for example by mismatched indentation causing unexpected behavior.

Tab vs. Spaces

"Question": What I find most "interesting" is the whole tab vs 4 spaces thing, I just find it really convenient to just press tab so I'm very curious as why they decided to opt for 4 spaces instead.

Answer: This isn't really a question but I wanted to address it anyway -- you don't have to press space 4 times to input 4 spaces! You can still just press tab! It's a matter of character encoding -- there's a difference between the tab character '\t' and the space character ' '. Tabs can be configured to render at different indentation levels, whereas spaces cannot. The use of spaces in Python is to prevent a tabstop of 8 characters instead of 4 characters from causing ambiguous rendering. Take this code, written using tabs and not spaaces:

if x == y:
    if y == z:
        print('x = y = z')

With an 8 character tabstop, this code would render like this:

if x == y:
        if y == z:
                print('x = y = z')

Since whitespace in Python matters, the shift in indentation on different screens can cause wrapping to look very different to two different people working on the same code, which can cause communication problems.