自动导入给定文件夹中的所有Python文件？

Question

I am relatively quite new to Python and I try to learn the "Pythonic" way of doing things to build a solid foundation in terms of Python development. Perhaps what I want to achieve is not Python at all, but I am nonetheless seeking to find out the "right" way to solve this issue.

I am building an application, for which I am creating modules. I just noticed that a module of mine has 7 different .py Python files, all importing 3 different same things. So all these files share these imports.

I tried removing them, and inserting these import to the empty init .py in the folder, but it did not do the trick.

If possible, since these imports are needed by all these module files, I would not like them to be imported in each file one by one.

What can I do to perform the common import?

Thank you very much, I really appreciate your kind help.

Answer 1

As the Zen of Python states, "Explicit is better than implicit", and this is a good example.

It's very useful to have the dependencies of a module listed explicitly in the imports and it means that every symbol in a file can be traced to its origin with a simple text search. Eg if you search for some_identifier in your file, you'll either find a definition in the file, or from some_module import some_identifier . It's even more obvious with direct references to some_module.some_identifier . (This is also one reason why you should not do from module import * .)

One thing you could do, without losing the above property, is to import your three shared modules into a fourth module:

#fourth.py
import first
import second
import third

then...

#another.py
import fourth

fourth.first.some_function()
#etc.

If you can't stomach that (it does make calls more verbose, after all) then the duplication of three imports is fine, really.

Answer 2

I agree with DrewV, it is perfectly pythonic to do

File1:

import xyz

import abc

...

File2:

import xyz

An almost identical question has also been addressed in the following link:

python multiple imports for a common module

As it explains, Python does the job of optimising the module load, so you can write multiple import statements and not worry about performance losses, because the module is only loaded once. In fact, listing out all the imports in each file makes it explicitly clear what each file depends on.

And for a discussion of how imports interact with namespaces, see:

Python imports across modules and global variables