Importing a dotted-name package locally for unit t

I am trying to develop and test a package foo.bar. I have the following directory layout:

myproject
  setup.py (for package foo.bar)
  foo
    bar
      __init__.py
      ...
  tests
    main.py
    test_something.py

In test_something.py, I want to import the local copy of foo.bar ideally just using 'import foo.bar'.

In main.py I have:

import os
import sys

sys.path.append(
    os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))

import test_something

However, I get an error:

File "./tests/main.py", line 22, in <module>
  import test_something
File "/myproject/tests/test_something.py", line 28, in <module>
  import foo.bar
ImportError: No module named bar

Is it possible to do this?

If you want foo to be considered as a package, you need to have a __init__.py directly under foo directory, it can be empty, but the file needs to be there. Otherwise Python would not consider foo as a package, and you cannot do -

import foo.bar

As said in the comments -

In any event I don't want foo to be a package, I want foo.bar to be the package.

This is not directly possible, as in, you cannot do import foo.bar unless foo is a Python package (or a namespace package explained below).

The only other alternative here would be to add foo folder directly into sys.path using -

sys.path.append(
    os.path.abspath(os.path.join(os.path.dirname(__file__), '..', 'foo')))

And then you would have to import it as -

import bar

If you are using Python 3.3 + , then you can use namespace packages , and you can import foo.bar without having to define a __init__.py inside foo.

This is explained well in - PEP 0420 , According to the specification, the new laoding procedure for python packages/modules (from 3.3 +) is -

During import processing, the import machinery will continue to iterate over each directory in the parent path as it does in Python 3.2. While looking for a module or package named "foo", for each directory in the parent path:

If <directory>/foo/__init__.py is found, a regular package is imported and returned.

If not, but <directory>/foo.{py,pyc,so,pyd} is found, a module is imported and returned. The exact list of extension varies by platform and whether the -O flag is specified. The list here is representative.

If not, but <directory>/foo is found and is a directory, it is recorded and the scan continues with the next directory in the parent path.

Otherwise the scan continues with the next directory in the parent path.

If the scan completes without returning a module or package, and at least one directory was recorded, then a namespace package is created. The new namespace package:

Has a __path__ attribute set to an iterable of the path strings that were found and recorded during the scan.

Does not have a __file__ attribute.