generalizing module import in python

Question

I am actually a little confused about making module import generalize. What I got here is a class shape. What I want to do is want to import corresponding file as a module based on some condition. What i'm trying to do is:

In Shape.py

class shape:
    def __init__(self, shape_id):
        if shape_id == '001':
            from shapes import triangle as imported_shape
        else:
            from shapes import square as imported_shape

In main.py:

from Shape import shape

sqaure = shape('002')
...

The project structure is:

Project
    |
      Shape.py
      main.py
      shapes
          |
            triangle.py
            square.py

but that does not seems to work as the import is made void after the __init__ function. Is there any way i can make this type of importing more generalized?

@PRMoureu That seems to work, but what i want is to import the module when initializing an object so that I could pass on the id while initialization. — Ahmed Dhanani
– Ahmed Dhanani, Commented Sep 11, 2017 at 5:36

Yaroslav Surzhikov · Accepted Answer · 2017-09-11 07:12:26Z

1

I can't reproduce your bug.

As test I've included similar method to both square and triangle modules, which prints square or triangle respectively, something like that:

def a():
    print('square')

I called it in __init__ of shape class and recieved expected output.

class shape:
    def __init__(self, shape_id):
        if shape_id == '001':
            from shapes import triangle as imported_shape
        else:
            from shapes import square as imported_shape

        imported_shape.a()

But if you want to use imported module somewhere elsewhere of __init__ - you should assing imported_shape to self:

class shape:
    def __init__(self, shape_id):
        if shape_id == '001':
            from shapes import triangle as imported_shape
        else:
            from shapes import square as imported_shape

        self.imported_shape = imported_shape

And after that - you can access your module in other methods of shape class:

def test(self):
    self.imported_shape.a()

According to your needs and python code standarts - it would be better to import shapes on top of your module and in __init__ do something like:

import shapes

class shape:
    def __init__(self, shape_id):
        if shape_id == '001':
            self.imported_shape = shapes.triangle
        else:
            self.imported_shape = shapes.square

OOP example:

Asuming that square and triangle have same-named classes:

from shapes.square import square
from shapes.triangle import triangle


class shape(square, triangle):
    def __init__(self, shape_id):
        if shape_id == '001':
            super(triangle, self).__init__()
        else:
            super(square, self).__init__()

edited Sep 11, 2017 at 7:12

answered Sep 11, 2017 at 5:48

Yaroslav Surzhikov

1,6081 gold badge13 silver badges16 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Ahmed Dhanani Over a year ago

It works, but my doubt is, is it a good approach to assign an import to a class variable?

Yaroslav Surzhikov Over a year ago

@MrPyCharm, well, thats not ideal, yeah. Better approach would be to inherit classes triangle and square in class shape and on __init__ call super(shapes.triangle, self).__init__() or super(shapes.square, self).__init__() according to id passed as param

Collectives™ on Stack Overflow

generalizing module import in python

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related