My team is enjoying using python to solve problems for our business. We write many small independent scripty applications.
However, we have to have a central windows box that runs these along with legacy applications.
Our challenge is going through a build and deploy process.
We want to have Bamboo check the script out of git, install requirements and run tests, then if all is green, just deploy to our production box.
We'd like libraries to be isolated from script to script so we don't have dependency issues.
We've tried to get virtualenvs to be portable but that seems a no go.
Pex looked promising, but it doesn't work on windows.
Ideally you'd see a folder like so:
AppOne
/Script.py
/Libs
/bar.egg
/foo.egg
AppTwo
/Script2.py
/Libs
/fnord.egg
/fleebly.py
Are we thinking about this wrong? What's the pythonic way to distribute scripts within an enterprise?
You may be able to do that with a neat if relatively unknown feature that was sneaked into Python 2.6 without much ado: executing zip files as Python applications. It got a bit (just a bit) more of publicity after PEP 441 (which is the one PEX is inspired in), although I think most people is still unaware of it. The idea is that you create a zip file (the recommeded extension is .pyz or .pyzw for windowed applications, but that's obviously not important) with all the code and modules that you want and then you simply run it with Python. The interpreter will add the contents of the zip file to sys.path and look for a top level module named __main__ and run it. Python 3.5 even introduced the convenience module zipapp to create such packaged applications, but there is really no magic in it and you may as well create it by hand or script.
In your case, I guess Bamboo could do the check out, dependency install and tests in virtualenvs and then package the application along with the environment libraries. It's not a one-click solution but it may do the trick without additional tools.
TL:DR:
Use Docker
A short story long:
You can use docker to create an independent image for every script that you want to deploy.
You can install a python image (slim is the lightest) as a base environment for each script or a group of scripts/applications and use it like a "virtualenv" in which you can install all your dependencies for that script.
There is also an integration for Bamboo and Docker which you may find useful.
Here is the Docker documentation for reference.
You can test each script individually in a separated container and if it passes then you can use the same container to deploy it in your main server.
It is not exactly what you are asking, but you can use this solution in every platform (Windows, Linux, etc.), you can deploy all your scripts to the enterprise server (or anywhere for that matter) and use them across your company.
Disclaimer: This is not THE solution, it is a solution that I am aware of which applies to the time of this answer (2017)
Another possibility is pyinstaller. It creates an executable that can be deployed. Python is not even required to be installed on the deployed production box. It is harder to debug problems that occur only on the deployed box. You also can't modify the scripts on the deployed box which depending on your trust of the owners of the machine is either a positive or negative. See http://www.pyinstaller.org/
As I understand it, you want to create self-contained application directories on a build server, then copy them over to a production server and run scripts directly from them. In particular, you want all dependencies (your own and external packages) installed within a Libs subdirectory in each application directory. Here's a fairly robust way to do that:
Create the top-level application directory (AppOne) and the Libs subdirectory inside it.
Use pip install --ignore-installed --target=Libs package_name to install dependencies into the Libs subdirectory.
Copy your own packages and modules into the Libs subdirectory (or install them there with pip).
Copy Script.py into the top-level directory.
Include code at the top of Script.py to add the Libs directory to sys.path:
import os, sys
app_path = os.path.dirname(__file__)
lib_path = os.path.abspath(os.path.join(app_path, 'Libs'))
sys.path.insert(0, lib_path)
This will make packages like Libs\bar.egg and modules like Libs\fleebly.py available to your script via import bar or import fleebly. Without code like this, there is no way for your script to find those packages and modules.
If you want to streamline this part of your script, there are a couple of options: (1) Put these lines in a separate fix_path.py module in the top-level directory and just call import fix_path at the start of your script. (2) Create a Libs\__init__.py file with the line sys.path.insert(0, os.path.dirname(__file__)), and then call import Libs from your script. After that, Libs\x can be imported via import x. This is neat, but it's a nonstandard use of the package and path mechanisms (it uses Libs as both a library directory and a package), so it could create some confusion about how importing works.
Once these directories and files are in place, you can copy this whole structure over to any Windows system with Python installed, and then run it using cd AppOne; python Script.py or python AppOne\Script.py. If you name your top-level script __main__.py instead of Script.py, then you can run your app just by executing python AppOne.
Further, as #jdehesa pointed out, if your script is named __main__.py, you can compress the contents of the AppOne directory (but not the AppOne directory itself) into a file called AppOne.zip, and then copy that to your production server and run it by calling python AppOne.zip. (On Python 3.5 or later, you can also create the zip file via python -m zipapp AppOne if your script is called __main__.py. You may also be able to use python -m zipapp AppOne -m Script if your script is called Script.py. See https://docs.python.org/3/library/zipapp.html.)
This kind of thing can be easily dealt with python setup.py
Sample setup.py
from setuptools import setup
setup(
name=name_for_distribution,
version=version_number,
py_modules=[pythonfiles],
install_requires=[
python packages that need to be installed
]
)
Create a virtual environment , activate it and run :
python setup.py install
I feel this is the most pythonic way to distribute and package your project.
Reading links:
https://pythonhosted.org/an_example_pypi_project/setuptools.html
https://docs.python.org/2/distutils/setupscript.html
So you can use distutils to create a file, such as
PIL-1.1.6.win32-py2.5.exe
which you can run and use to easily install something. However, the installation requires user input to proceed (you have to click 'OK' three times). I want to create an easily installable windows version that you can just run as a cmd line program, that doesn't require input from the user. Is this possible? Do these .exe files do it already, but you need to pass them a magic cmd line argument to work?
See this post which describes an idea to modify the stub installer like this:
It also mentions another alternative: use setup.py bdist_msi instead, which will produce an msi package, that can be installed unattended
You get the executable by running "setup.py bdist_wininst". You can have something simpler by running "setup.py bdist_dumb". This will produce a .zip file which, unzipped at the root of the drive where Python is installed, provided that it's installed in the same directory as the machine you've build it, will install the library.
Now I don't know if there is an unzip command-line utility under Windows that can be used to do that; I usually have Cygwin installed on all my Windows boxes, but it may prove quite simple to just ship it with the .zip.
I have done this before using a simple batch file to call setuptools' install script passing the egg file path as an argument to it. The only trouble is that you need to ensure that script is in the PATH, which it might not be.
Assuming Python itself is in the PATH you can try something like this in a python script you would distribute with your egg (call it install.py or somesuch).
import sys
from pkg_resources import load_entry_point
def install_egg(egg_file_path):
sys.argv[1] = egg_file_path
easy_install = load_entry_point(
'setuptools==0.6c9',
'console_scripts',
'easy_install'
)
easy_install()
if __name__ == "__main__":
install_egg("PIL-1.1.6.win32-py2.5.egg")
Essentially this does the same as the "easy_install.py" script. You're locating the entrypoint for that script, and setting up sys.argv so that the first argument points at your egg file. It should do the rest for you.
HTH
Python is installed in a local directory.
My directory tree looks like this:
(local directory)/site-packages/toolkit/interface.py
My code is in here:
(local directory)/site-packages/toolkit/examples/mountain.py
To run the example, I write python mountain.py, and in the code I have:
from toolkit.interface import interface
And I get the error:
Traceback (most recent call last):
File "mountain.py", line 28, in ?
from toolkit.interface import interface
ImportError: No module named toolkit.interface
I have already checked sys.path and there I have the directory /site-packages. Also, I have the file __init__.py.bin in the toolkit folder to indicate to Python that this is a package. I also have a __init__.py.bin in the examples directory.
I do not know why Python cannot find the file when it is in sys.path. Any ideas? Can it be a permissions problem? Do I need some execution permission?
Based on your comments to orip's post, I guess this is what happened:
You edited __init__.py on windows.
The windows editor added something non-printing, perhaps a carriage-return (end-of-line in Windows is CR/LF; in unix it is LF only), or perhaps a CTRL-Z (windows end-of-file).
You used WinSCP to copy the file to your unix box.
WinSCP thought: "This has something that's not basic text; I'll put a .bin extension to indicate binary data."
The missing __init__.py (now called __init__.py.bin) means python doesn't understand toolkit as a package.
You create __init__.py in the appropriate directory and everything works... ?
Does
(local directory)/site-packages/toolkit
have a __init__.py?
To make import walk through your directories every directory must have a __init__.py file.
I ran into something very similar when I did this exercise in LPTHW; I could never get Python to recognise that I had files in the directory I was calling from. But I was able to get it to work in the end. What I did, and what I recommend, is to try this:
(NOTE: From your initial post, I am assuming you are using an *NIX-based machine and are running things from the command line, so this advice is tailored to that. Since I run Ubuntu, this is what I did)
Change directory (cd) to the directory above the directory where your files are. In this case, you're trying to run the mountain.py file, and trying to call the toolkit.interface.py module, which are in separate directories. In this case, you would go to the directory that contains paths to both those files (or in other words, the closest directory that the paths of both those files share). Which in this case is the toolkit directory.
When you are in the toolkit directory, enter this line of code on your command line:
export PYTHONPATH=.
This sets your PYTHONPATH to ".", which basically means that your PYTHONPATH will now look for any called files within the directory you are currently in, (and more to the point, in the sub-directory branches of the directory you are in. So it doesn't just look in your current directory, but in all the directories that are in your current directory).
After you've set your PYTHONPATH in the step above, run your module from your current directory (the toolkit directory). Python should now find and load the modules you specified.
On *nix, also make sure that PYTHONPATH is configured correctly, especially that it has this format:
.:/usr/local/lib/python
(Mind the .: at the beginning, so that it can search on the current directory, too.)
It may also be in other locations, depending on the version:
.:/usr/lib/python
.:/usr/lib/python2.6
.:/usr/lib/python2.7 and etc.
You are reading this answer says that your __init__.py is in the right place, you have installed all the dependencies and you are still getting the ImportError.
I was facing a similar issue except that my program would run fine when ran using PyCharm but the above error when I would run it from the terminal. After digging further, I found out that PYTHONPATH didn't have the entry for the project directory. So, I set PYTHONPATH per Import statement works on PyCharm but not from terminal:
export PYTHONPATH=$PYTHONPATH:`pwd` (OR your project root directory)
There's another way to do this using sys.path as:
import sys
sys.path.insert(0,'<project directory>') OR
sys.path.append('<project directory>')
You can use insert/append based on the order in which you want your project to be searched.
Using PyCharm (part of the JetBrains suite) you need to define your script directory as Source:
Right Click > Mark Directory as > Sources Root
For me, it was something really stupid. I installed the library using pip3 install but was running my program as python program.py as opposed to python3 program.py.
I solved my own problem, and I will write a summary of the things that were wrong and the solution:
The file needs to be called exactly __init__.py. If the extension is different such as in my case .py.bin then Python cannot move through the directories and then it cannot find the modules. To edit the files you need to use a Linux editor, such as vi or nano. If you use a Windows editor this will write some hidden characters.
Another problem that was affecting it was that I had another Python version installed by the root, so if someone is working with a local installation of python, be sure that the Python installation that is running the programs is the local Python. To check this, just do which python, and see if the executable is the one that is in your local directory. If not, change the path, but be sure that the local Python directory is before than the other Python.
To mark a directory as a package you need a file named __init__.py, does this help?
an easy solution is to install the module using python -m pip install <library-name> instead of pip install <library-name>
you may use sudo in case of admin restrictions
To all those who still have this issue. I believe Pycharm gets confused with imports. For me, when i write 'from namespace import something', the previous line gets underlined in red, signaling that there is an error, but works. However ''from .namespace import something' doesn't get underlined, but also doesn't work.
Try
try:
from namespace import something
except NameError:
from .namespace import something
Yup. You need the directory to contain the __init__.py file, which is the file that initializes the package. Here, have a look at this.
The __init__.py files are required to make Python treat the directories as containing packages; this is done to prevent directories with a common name, such as string, from unintentionally hiding valid modules that occur later on the module search path. In the simplest case, __init__.py can just be an empty file, but it can also execute initialization code for the package or set the __all__ variable, described later.
If you have tried all methods provided above but failed, maybe your module has the same name as a built-in module. Or, a module with the same name existing in a folder that has a high priority in sys.path than your module's.
To debug, say your from foo.bar import baz complaints ImportError: No module named bar. Changing to import foo; print foo, which will show the path of foo. Is it what you expect?
If not, Either rename foo or use absolute imports.
You must have the file __ init__.py in the same directory where it's the file that you are importing.
You can not try to import a file that has the same name and be a file from 2 folders configured on the PYTHONPATH.
eg:
/etc/environment
PYTHONPATH=$PYTHONPATH:/opt/folder1:/opt/folder2
/opt/folder1/foo
/opt/folder2/foo
And, if you are trying to import foo file, python will not know which one you want.
from foo import ... >>> importerror: no module named foo
My two cents:
Spit:
Traceback (most recent call last):
File "bash\bash.py", line 454, in main
import bosh
File "Wrye Bash Launcher.pyw", line 63, in load_module
mod = imp.load_source(fullname,filename+ext,fp)
File "bash\bosh.py", line 69, in <module>
from game.oblivion.RecordGroups import MobWorlds, MobDials, MobICells, \
ImportError: No module named RecordGroups
This confused the hell out of me - went through posts and posts suggesting ugly syspath hacks (as you see my __init__.py were all there). Well turns out that game/oblivion.py and game/oblivion was confusing python
which spit out the rather unhelpful "No module named RecordGroups". I'd be interested in a workaround and/or links documenting this (same name) behavior -> EDIT (2017.01.24) - have a look at What If I Have a Module and a Package With The Same Name? Interestingly normally packages take precedence but apparently our launcher violates this.
EDIT (2015.01.17): I did not mention we use a custom launcher dissected here.
Fixed my issue by writing print (sys.path) and found out that python was using out of date packages despite a clean install. Deleting these made python automatically use the correct packages.
In my case, because I'm using PyCharm and PyCharm create a 'venv' for every project in project folder, but it is only a mini env of python. Although you have installed the libraries you need in Python, but in your custom project 'venv', it is not available. This is the real reason of 'ImportError: No module named xxxxxx' occurred in PyCharm.
To resolve this issue, you must add libraries to your project custom env by these steps:
In PyCharm, from menu 'File'->Settings
In Settings dialog, Project: XXXProject->Project Interpreter
Click "Add" button, it will show you 'Available Packages' dialog
Search your library, click 'Install Package'
Then, all you needed package will be installed in you project custom 'venv' folder.
Enjoy.
Linux: Imported modules are located in /usr/local/lib/python2.7/dist-packages
If you're using a module compiled in C, don't forget to chmod the .so file after sudo setup.py install.
sudo chmod 755 /usr/local/lib/python2.7/dist-packages/*.so
In my case, the problem was I was linking to debug python & boost::Python, which requires that the extension be FooLib_d.pyd, not just FooLib.pyd; renaming the file or updating CMakeLists.txt properties fixed the error.
My problem was that I added the directory with the __init__.py file to PYTHONPATH, when actually I needed to add its parent directory.
For me, running the file as a module helped.
Instead of
python myapp/app.py
using
python -m myapp.app
It's not exactly the same but it might be a better approach in some cases.
If you are using a setup script/utility (e.g. setuptools) to deploy your package, don't forget to add the respective files/modules to the installer.
When supported, use find_packages() or similar to automatically add new packages to the setup script. This will absolutely save you from a headache, especially if you put your project aside for some time and then add something later on.
import setuptools
setuptools.setup(
name="example-pkg",
version="0.0.1",
author="Example Author",
author_email="author#example.com",
description="A small example package",
packages=setuptools.find_packages(),
classifiers=[
"Programming Language :: Python :: 3",
"Operating System :: OS Independent",
],
python_requires='>=3.6',
)
(Example taken from setuptools documentation)
I had the same problem (Python 2.7 Linux), I have found the solution and i would like to share it. In my case i had the structure below:
Booklet
-> __init__.py
-> Booklet.py
-> Question.py
default
-> __init_.py
-> main.py
In 'main.py' I had tried unsuccessfully all the combinations bellow:
from Booklet import Question
from Question import Question
from Booklet.Question import Question
from Booklet.Question import *
import Booklet.Question
# and many othet various combinations ...
The solution was much more simple than I thought. I renamed the folder "Booklet" into "booklet" and that's it. Now Python can import the class Question normally by using in 'main.py' the code:
from booklet.Booklet import Booklet
from booklet.Question import Question
from booklet.Question import AnotherClass
From this I can conclude that Package-Names (folders) like 'booklet' must start from lower-case, else Python confuses it with Class names and Filenames.
Apparently, this was not your problem, but John Fouhy's answer is very good and this thread has almost anything that can cause this issue. So, this is one more thing and I hope that maybe this could help others.
In linux server try dos2unix script_name
(remove all (if there is any) pyc files with command find . -name '*.pyc' -delete)
and re run in the case if you worked on script on windows
In my case, I was using sys.path.insert() to import a local module and was getting module not found from a different library. I had to put sys.path.insert() below the imports that reported module not found. I guess the best practice is to put sys.path.insert() at the bottom of your imports.
I've found that changing the name (via GUI) of aliased folders (Mac) can cause issues with loading modules. If the original folder name is changed, remake the symbolic link. I'm unsure how prevalent this behavior may be, but it was frustrating to debug.
another cause makes this issue
file.py
#!/bin/python
from bs4 import BeautifulSoup
if your default python is pyyhon2
$ file $(which python)
/sbin/python: symbolic link to python2
file.py need python3, for this case(bs4)
you can not execute this module with python2 like this:
$ python file.py
# or
$ file.py
# or
$ file.py # if locate in $PATH
Tow way to fix this error,
# should be to make python3 as default by symlink
$ rm $(which python) && ln -s $(which python3) /usr/bin/python
# or use alias
alias python='/usr/bin.../python3'
or change shebang in file.py to
#!/usr/bin/...python3
After just suffering the same issue I found my resolution was to delete all pyc files from my project, it seems like these cached files were somehow causing this error.
Easiest way I found to do this was to navigate to my project folder in Windows explorer and searching for *.pyc, then selecting all (Ctrl+A) and deleting them (Ctrl+X).
Its possible I could have resolved my issues by just deleting the specific pyc file but I never tried this
I faced the same problem: Import error. In addition the library've been installed 100% correctly. The source of the problem was that on my PC 3 version of python (anaconda packet) have been installed). This is why the library was installed no to the right place. After that I just changed to the proper version of python in the my IDE PyCharm.
I had the same error. It was caused by somebody creating a folder in the same folder as my script, the name of which conflicted with a module I was importing from elsewhere. Instead of importing the external module, it looked inside this folder which obviously didn't contain the expected modules.