What is __init__.py
for in a Python source directory?
12 Answers
It used to be a required part of a package (old, pre-3.3 "regular package", not newer 3.3+ "namespace package").
Python defines two types of packages, regular packages and namespace packages. Regular packages are traditional packages as they existed in Python 3.2 and earlier. A regular package is typically implemented as a directory containing an
__init__.py
file. When a regular package is imported, this__init__.py
file is implicitly executed, and the objects it defines are bound to names in the package’s namespace. The__init__.py
file can contain the same Python code that any other module can contain, and Python will add some additional attributes to the module when it is imported.
But just click the link, it contains an example, more information, and an explanation of namespace packages, the kind of packages without __init__.py
.
Files named __init__.py
are used to mark directories on disk as Python package directories.
If you have the files
mydir/spam/__init__.py
mydir/spam/module.py
and mydir
is on your path, you can import the code in module.py
as
import spam.module
or
from spam import module
If you remove the __init__.py
file, Python will no longer look for submodules inside that directory, so attempts to import the module will fail.
The __init__.py
file is usually empty, but can be used to export selected portions of the package under more convenient name, hold convenience functions, etc.
Given the example above, the contents of the init module can be accessed as
import spam
based on this
In addition to labeling a directory as a Python package and defining __all__
, __init__.py
allows you to define any variable at the package level. Doing so is often convenient if a package defines something that will be imported frequently, in an API-like fashion. This pattern promotes adherence to the Pythonic "flat is better than nested" philosophy.
An example
Here is an example from one of my projects, in which I frequently import a sessionmaker
called Session
to interact with my database. I wrote a "database" package with a few modules:
database/
__init__.py
schema.py
insertions.py
queries.py
My __init__.py
contains the following code:
import os
from sqlalchemy.orm import sessionmaker
from sqlalchemy import create_engine
engine = create_engine(os.environ['DATABASE_URL'])
Session = sessionmaker(bind=engine)
Since I define Session
here, I can start a new session using the syntax below. This code would be the same executed from inside or outside of the "database" package directory.
from database import Session
session = Session()
Of course, this is a small convenience -- the alternative would be to define Session
in a new file like "create_session.py" in my database package, and start new sessions using:
from database.create_session import Session
session = Session()
Further reading
There is a pretty interesting reddit thread covering appropriate uses of __init__.py
here:
http://www.reddit.com/r/Python/comments/1bbbwk/whats_your_opinion_on_what_to_include_in_init_py/
The majority opinion seems to be that __init__.py
files should be very thin to avoid violating the "explicit is better than implicit" philosophy.
There are 2 main reasons for __init__.py
For convenience: the other users will not need to know your functions' exact location in your package hierarchy.
your_package/ __init__.py file1.py file2.py ... fileN.py
# in __init__.py from file1 import * from file2 import * ... from fileN import *
# in file1.py def add(): pass
then others can call add() by
from your_package import add
without knowing file1, like
from your_package.file1 import add
If you want something to be initialized; for example, logging (which should be put in the top level):
import logging.config logging.config.dictConfig(Your_logging_config)
Since Python 3.3, __init__.py
is no longer required to define directories as importable Python packages.
Check PEP 420: Implicit Namespace Packages:
Native support for package directories that don’t require
__init__.py
marker files and can automatically span multiple path segments (inspired by various third party approaches to namespace packages, as described in PEP 420)
Here's the test:
$ mkdir -p /tmp/test_init
$ touch /tmp/test_init/module.py /tmp/test_init/__init__.py
$ tree -at /tmp/test_init
/tmp/test_init
├── module.py
└── __init__.py
$ python3
>>> import sys
>>> sys.path.insert(0, '/tmp')
>>> from test_init import module
>>> import test_init.module
$ rm -f /tmp/test_init/__init__.py
$ tree -at /tmp/test_init
/tmp/test_init
└── module.py
$ python3
>>> import sys
>>> sys.path.insert(0, '/tmp')
>>> from test_init import module
>>> import test_init.module
references:
https://docs.python.org/3/whatsnew/3.3.html#pep-420-implicit-namespace-packages
https://www.python.org/dev/peps/pep-0420/
Is __init__.py not required for packages in Python 3?
Although Python works without an __init__.py
file you should still include one.
It specifies that the directory should be treated as a package, so therefore include it (even if it is empty).
There is also a case where you may actually use an __init__.py
file:
Imagine you had the following file structure:
main_methods
|- methods.py
And methods.py
contained this:
def foo():
return 'foo'
To use foo()
you would need one of the following:
from main_methods.methods import foo # Call with foo()
from main_methods import methods # Call with methods.foo()
import main_methods.methods # Call with main_methods.methods.foo()
Maybe there you need (or want) to keep methods.py
inside main_methods
(runtimes/dependencies for example) but you only want to import main_methods
.
If you changed the name of methods.py
to __init__.py
then you could use foo()
by just importing main_methods
:
import main_methods
print(main_methods.foo()) # Prints 'foo'
This works because __init__.py
is treated as part of the package.
Some Python packages actually do this. An example is with JSON, where running import json
is actually importing __init__.py
from the json
package (see the package file structure here):
Source code:
Lib/json/__init__.py
In Python the definition of package is very simple. Like Java the hierarchical structure and the directory structure are the same. But you have to have __init__.py
in a package. I will explain the __init__.py
file with the example below:
package_x/
|-- __init__.py
|-- subPackage_a/
|------ __init__.py
|------ module_m1.py
|-- subPackage_b/
|------ __init__.py
|------ module_n1.py
|------ module_n2.py
|------ module_n3.py
__init__.py
can be empty, as long as it exists. It indicates that the directory should be regarded as a package. Of course, __init__.py
can also set the appropriate content.
If we add a function in module_n1:
def function_X():
print "function_X in module_n1"
return
After running:
>>>from package_x.subPackage_b.module_n1 import function_X
>>>function_X()
function_X in module_n1
Then we followed the hierarchy package and called module_n1 the function. We can use __init__.py
in subPackage_b like this:
__all__ = ['module_n2', 'module_n3']
After running:
>>>from package_x.subPackage_b import *
>>>module_n1.function_X()
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: No module named module_n1
Hence using * importing, module package is subject to __init__.py
content.
__init__.py
will treat the directory it is in as a loadable module.
For people who prefer reading code, I put Two-Bit Alchemist's comment here.
$ find /tmp/mydir/
/tmp/mydir/
/tmp/mydir//spam
/tmp/mydir//spam/__init__.py
/tmp/mydir//spam/module.py
$ cd ~
$ python
>>> import sys
>>> sys.path.insert(0, '/tmp/mydir')
>>> from spam import module
>>> module.myfun(3)
9
>>> exit()
$
$ rm /tmp/mydir/spam/__init__.py*
$
$ python
>>> import sys
>>> sys.path.insert(0, '/tmp/mydir')
>>> from spam import module
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: No module named spam
>>>
It facilitates importing other python files. When you placed this file in a directory (say stuff)containing other py files, then you can do something like import stuff.other.
root\
stuff\
other.py
morestuff\
another.py
Without this __init__.py
inside the directory stuff, you couldn't import other.py, because Python doesn't know where the source code for stuff is and unable to recognize it as a package.
An __init__.py
file makes imports easy. When an __init__.py
is present within a package, function a()
can be imported from file b.py
like so:
from b import a
Without it, however, you can't import directly. You have to amend the system path:
import sys
sys.path.insert(0, 'path/to/b.py')
from b import a
One thing __init__.py allows is converting a module to a package without breaking the API or creating extraneous nested namespaces or private modules*. This helps when I want to extend a namespace.
If I have a file util.py containing
def foo():
...
then users will access foo
with
from util import foo
If I then want to add utility functions for database interaction, and I want them to have their own namespace under util
, I'll need a new directory**, and to keep API compatibility (so that from util import foo
still works), I'll call it util/. I could move util.py into util/ like so,
util/
__init__.py
util.py
db.py
and in util/__init__.py do
from util import *
but this is redundant. Instead of having a util/util.py file, we can just put the util.py contents in __init__.py and the user can now
from util import foo
from util.db import check_schema
I think this nicely highlights how a util
package's __init__.py acts in a similar way to a util
module
* this is hinted at in the other answers, but I want to highlight it here
** short of employing import gymnastics. Note it won't work to create a new package with the same name as the file, see this
__init__
is namespace package, not a regular package. It's not the same thing as @methane pointed out with an example here. – Catbuilts