When are classes better to use than user-defined functions in python?

user_18349448 0
Viewed 63

I am learning classes in python, and I have two methods of webscraping a website -namely, the functions gets the urls to paginate through. One way is written via a class Method and the other is just a straight up function. I am confused, they are both working and do the same output, but I am confused which way is more pythonic and efficient?

Using Class

class Get_URL:
    def __init__(self,city,price_max,price_min, bedrm_min, bath_min):
        if price_max and price_min != None and price_max <= price_min:
            raise ValueError
        self.url = f'&for_sale=1&quicksearch={city}&listing_price_max={price_max}&listing_price_min={price_min}&bedroom_min={bedrm_min}&full_bath_min={bath_min}&property_class_id=1,2,6,4'

        while price_max == None:
            self.url = self.url.replace(f'&listing_price_max={price_max}', '')
            break
        while price_min == None:
            self.url = self.url.replace(f'&listing_price_min={price_min}', '')
            break
        while bedrm_min == None:
            self.url = self.url.replace(f'&bedroom_min={bedrm_min}', '')
            break
        while bath_min == None:
            self.url = self.url.replace(f'&full_bath_min={bath_min}', '')
            break
    
    def get_urls(self):
        self.url_base = 'https://www.har.com/search/dosearch?page='
        self.url_lst = []
        for number in range(1,21):
            new_url = f'{self.url_base}{number}{self.url}'
            self.url_lst.append(new_url)

Output:

query1 = Get_URL('Houston', 100000,50000,None, None)
query1.get_urls()
query1.url_lst

['https://www.har.com/search/dosearch?page=1&for_sale=1&quicksearch=Houston&listing_price_max=100000&listing_price_min=50000&property_class_id=1,2,6,4',
 'https://www.har.com/search/dosearch?page=2&for_sale=1&quicksearch=Houston&listing_price_max=100000&listing_price_min=50000&property_class_id=1,2,6,4',
 'https://www.har.com/search/dosearch?page=3&for_sale=1&quicksearch=Houston&listing_price_max=100000&listing_price_min=50000&property_class_id=1,2,6,4',
:
:
:
]

Using User Defined Function

def get_houses(city, price_max, price_min,bedrm_min, bath_min):
    # raise error if price max less than price min
    if price_max and price_min != None and price_max <= price_min:
        raise ValueError
    # define url
    page = 1
    url = f'&for_sale=1&quicksearch={city}\
        &listing_price_max={price_max}\
        &listing_price_min={price_min}\
        &bedroom_min={bedrm_min}\
        &full_bath_min={bath_min}\
        &property_class_id=1,2,6,4'
    
    while price_max == None:
        url = url.replace(f'&listing_price_max={price_max}', '')
        break
    while price_min == None:
        url = url.replace(f'&listing_price_min={price_min}', '')
        break
    while bedrm_min == None:
        url = url.replace(f'&bedroom_min={bedrm_min}', '')
        break
    while bath_min == None:
        url = url.replace(f'&full_bath_min={bath_min}', '')
        break
    
    # Get URL List
    
    url_lst = []
    for number in range(1,21):
        url_base = f'https://www.har.com/search/dosearch?page={number}'
        url_lst.append(url_base+url)

    return(url_lst)

Outputs:

get_houses('Houston', 100000,50000,None, None)

['https://www.har.com/search/dosearch?page=1&for_sale=1&quicksearch=Houston&listing_price_max=100000&listing_price_min=50000&property_class_id=1,2,6,4',
 'https://www.har.com/search/dosearch?page=2&for_sale=1&quicksearch=Houston&listing_price_max=100000&listing_price_min=50000&property_class_id=1,2,6,4',
 'https://www.har.com/search/dosearch?page=3&for_sale=1&quicksearch=Houston&listing_price_max=100000&listing_price_min=50000&property_class_id=1,2,6,4',
:
:
:
]
user-defined-functions class python
3 Answers

For your current project, functional programming works well and it is easy to develop. But Object Oriented Programming (using classes) have many advantages, especially for a larger project or a project that will be developed or re-used in the future. Here is some pros [1]:

  1. Modularity for easier troubleshooting
  2. Reuse of code through inheritance
  3. Flexibility

1: 4 Advantages of Object-Oriented Programming

user_12070250

Im not an authority on the subject, but the difference, in my opinion, is in the use case. Use what you need.

If you are creating something small and simple, why bother with OOP and classes?

On the other hand, if you are thinking of starting a project that you know will be big and complex, or maybe you just want to give yourselve this oportunity to easily add and extend the code, then classes might be a way to go.

Having said that, I would look into OOP and its principles and see exactly what that is about. It can be complex for a beginer, however, using right tools for the job is usualy a way to go.

Hope this helps, if you have any more questions just ask.

user_112719270

Actually it's about the main difference between OOP & Structured Programming. We're using OOP languages (like Python) for easier writing and better understanding about our code. Because class & object terms are closer to real life concepts.

For example you have a car, it has 4 wheels & it can move. In Structured Approach you should declare variables as separated things or you can use struct keyword(which is pretty similar to class) but in Object Oriented Approach you can declare a class as car then give it an attribute as wheels then give it a method as move.

See, it is pretty more understandable and easy to code it for a developer. Main reason for prefer OOP is what i explained above. As the project getting bigger it becomes harder to understand the code and find bugs when they happen also if you are in a team project this factor becomes crucial.

To sum up we can say for real life projects, readability is essential so do not underrate it. Use classes for more readable codes

Maybe this video helps too for understanding about this subject: https://www.youtube.com/watch?v=o7h_sYMk_oc&list=PLUl4u3cNGP63VIBQVWguXxZZi0566y7Wf

user_164056360
Related