I have been using the method described in this post for setting up IPython Notebook to play nicely with Django. The gist of the method is to create an IPython extension which sets the DJANGO_SETTINGS_MODULE and runs django.setup() when IPython starts.
The code for the extension is:
def load_ipython_extension(ipython):
# The `ipython` argument is the currently active `InteractiveShell`
# instance, which can be used in any way. This allows you to register
# new magics or aliases, for example.
try:
import os
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "settings")
import django
django.setup()
except ImportError:
pass
With a recent upgrade to Jupyter Notebook this setup is now broken for me. I am able to run Django code in the Jupyter notebook by adding a similar bit of code to the first cell of the notebook. However, I was not able to figure out how to get Jupyter to run the extension automatically so I would not have to do this again for each and every notebook I am creating.
What should I do to get Django and Jupyter to play nicely?
UPDATE:
For #DarkLight - I am using Django 1.8.5 with Jupyter 1.0.0. The code I run in the notebook is:
import os, sys
sys.path.insert(0, '/path/to/project')
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "settingsfile")
import django
django.setup()
Install django-extensions from https://github.com/django-extensions/django-extensions/blob/master/docs/index.rst
pip install django-extensions
Change your settings file to include 'django-extensions'
INSTALLED_APPS += ['django_extensions']
Run your Django server like this:
python manage.py shell_plus --notebook
alter to suit, and run this in your first cell
import os, sys
PWD = os.getenv('PWD')
os.chdir(PWD)
sys.path.insert(0, os.getenv('PWD'))
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "local_settings.py")
import django
django.setup()
Now you should be able to import your django models etc. eg:
from app.models import Foobar
Foobar.objects.all()
Just for completeness (but it's 2018, so maybe things changed since this question was posted): you can actually install a Jupyter Python kernel in your Django environment that will then connect (run under) a different Jupyter server/environment (one where you've installed widgets, extensions, changed the theme, etc.). django_extensions right now still does only part of the required work :-)
This assumes you have a Jupyter virtual environment that's separate from Django's one and whose kernels/extensions are installed with --user. All the Jupyter extensions (and their dependencies) are installed in this venv instead of the Django's one/ones (you'll still need pandas, matplotlib, etc. in the Django environment if you need to use them together with Django code).
In your Django virtual environment (that can run a different version of Python, including a version 2 interpreter) install the ipython kernel:
pip install -U ipykernel
ipython kernel install --user --name='environment_name' --display-name='Your Project'
This will create a kernel configuration directory with the specified -–name in your user’s Jupyter kernel directory (on Linux it's ~/.jupyter/ while on OSX it’s ~/Library/Jupyter/) containing its kernel.json file and images/icons (by default the default Jupyter icon for the kernel we’re installing are used). This kernel will run inside the virtual environment what was active at creation, thus using the exact same version of python and all the installed modules used by our Django project.
Running ./manage.py shell_plus --notebook does something very similar, but in addition to requiring everything (including the Jupyter server and all the extensions) installed in the current venv, it’s also unable to run notebooks in directories different from the project’s root (the one containing ./manage.py). In addition it’ll run the kernel using the first executable called python it finds on the path, not the virtual environment’s one, making it misbehave when not started from the command line inside an active Django virtual environment.
To fix these problems so that we're able to create a Notebook running inside any Django project we have so configured and to be able to run notebooks stored anywhere on the filesystem, we need to:
make sure the first ‘argv’ parameter contains the full path to the python interpreter contained in the virtual environment
add (if not already present) an ‘env’ section that will contain shell environment variables, then use these to tell Python where to find our project and which Django settings it should use. We do this by adding something like the following:
"env": {
"DJANGO_SETTINGS_MODULE": "my_project.settings",
"PYTHONPATH": "$PYTHONPATH:/home/projectuser/projectfolder/my_project"
}
optional: change ‘display_name’ to be human friendly and replace the icons.
editing this environment kernel.json file you'll see something similar:
{
"display_name": "My Project",
"language": "python",
"env": {
"DJANGO_SETTINGS_MODULE": "my_project.settings",
"PYTHONPATH": "$PYTHONPATH:/home/projectuser/projectfolder/my_project"
},
"argv": [
"/home/projectuser/.pyenv/versions/2.7.15/envs/my_project_venv/bin/python",
"-m",
"ipykernel_launcher",
"-f",
"{connection_file}",
"--ext",
"django_extensions.management.notebook_extension"
]
}
Notable lines:
"DJANGO_SETTINGS_MODULE": "my_project.settings": your settings, usually as seen inside your project's manage.py
"PYTHONPATH": "$PYTHONPATH:/home/projectuser/projectfolder/my_project": PYTHONPATH is extended to include your project's main directory (the one containing manage.py) so that settings can be found even if the kernel isn't run in that exact directory (here django_extensions will use a generic python, thus running the wrong virtual environment unless the whole Jupyter server is launched from inside it: adding this to the kernel.json created by django_extensions will enable it to run notebooks anywhere in the Django project directory)
"/home/projectuser/.pyenv/versions/2.7.15/envs/my_project_venv/bin/python": first argument (argv list) of the kernel execution, should be the full path to your project's virtual environment's python interpreter (this is another thing django_extensions gets wrong: fixing this will allow any notebook server to run that specific Django environment's kernel with all its installed modules)
"django_extensions.management.notebook_extension": this is the extension that will load the 'shell_plus' functionality in the notebook (optional but useful :-) )
Here's what just worked for me
install Django Extensions (I used 1.9.6) as per other answers
install jupyterpip install jupyter
some stuff I did to setup jupyter inside my Docker container -- see below if this applies to you †
from your base Django directory, create a directory for notebooks, e.g. mkdir notebooks
Go to that directory cd notebooks
start django extensions shell_plus from inside that directory: ../manage.py shell_plus --notebook
The notebook server should now be running, and may launch a new browser. If it doesn't launch a browser window, follow the instructions to paste a link or a token.
from the browser, open a new "Django Shell Plus" notebook, as per John Mee's answer's screenshot
AND, importantly, what didn't work was changing directories from inside the notebook environment. If I tried to work with any notebook that was not in the directory that manage.py shell_plus --notebook was run in, then the kernal was not configured correctly. For me, having the notebook be configured for just a single directory at a time was good enough. If you need a more robust solution, you should be able set PYTHONPATH prior to starting jupyter. For example add export PYTHONPATH="$PYTHONPATH:/path/to/django/project" to a virtualenv activate script. But I haven't tried this.
† Docker Setup (optional)
add a port mapping for your container for port 8888
For example, in your docker compose file;
ports:
- "8890:8888"
Configure your project settings file to use ip 0.0.0.0
This is what I did:
NOTEBOOK_ARGUMENTS = [
'--ip', '0.0.0.0',
'--allow-root',
'--no-browser',
]
Note: I am using Python 3.7 and Django 2.1, it works for Django 2.2. I don't have to run anything in my first cell, and this works like charm as long as you don't mind having the notebooks in the root of your Django project.
It is assumed that you have a virtual environment for your project, and it is activated. I use pipenv to create virtual environments and track dependencies of my python projects, but it is up to you what tool you use.
It is also assumed that you have created a Django project and your current working directory is the root of this project.
Steps
Install jupyter
Using pip
pip install jupyter
Using pipenv
pipenv install jupyter
Install django-extentions
Using pip
pip install django-extensions
Using pipenv
pipenv install django-extensions
Set up django-extensions by adding it to the INSTALLED_APPS setting of your Django project settings.py file.:
INSTALLED_APPS = (
...
'django_extensions',
)
Run the shell_plus management command that is part of django-extensions. Use the option --notebook to start a notebook:
python manage.py shell_plus --notebook
Jupyter Notebooks will open automatically in your browser.
Start a new Django Shell-Plus notebook
That's it!
Again, you don't have to run anything in the first cell, and you can corroborate by running dir() to see the names in the current local scope.
Edit:
If you want to put your notebooks in a directory called notebooks at the root directory, you can do the following:
$ mkdir notebooks && cd notebooks
$ python ../manage.py shell_plus --notebook
Thanks to Mark Chackerian whose answer provided the idea to make run the notebooks in a directory other than the project's root.
These are the modules that are imported automatically thanks to shell_plus:
# Shell Plus Model Imports
from django.contrib.admin.models import LogEntry
from django.contrib.auth.models import Group, Permission, User
from django.contrib.contenttypes.models import ContentType
from django.contrib.sessions.models import Session
# Shell Plus Django Imports
from django.core.cache import cache
from django.conf import settings
from django.contrib.auth import get_user_model
from django.db import transaction
from django.db.models import Avg, Case, Count, F, Max, Min, Prefetch, Q, Sum, When, Exists, OuterRef, Subquery
from django.utils import timezone
from django.urls import reverse
Actually turns out you (might not) need to do all that crap. Just install django-extensions and run jupyter!
(myprojectvenv)$ cd myproject
(myprojectvenv)$ pip install jupyter
(myprojectvenv)$ pip install django-extensions
(myprojectvenv)$ jupyter notebook
In the browser, start a new "Django Shell-Plus":
And you should be good to go. eg:
from myproject.models import Foobar
Foobar.objects.all()
While the accepted answer from RobM works, it was less clear than it could be and has a few unnecessary steps. Simply put, to run notebooks through Django from a notebook environment outside of the project directory:
Install:
pip install django-extensions
Add 'django-extensions' to your INSTALLED_APPS list in settings.py
INSTALLED_APPS += ['django_extensions']
Run a notebook from within Django, then close it:
python manage.py shell_plus --notebook
This will create your kernel, which we will now edit to point to an absolute path of Python rather than a relative path.
On OSX, the kernel file is at: ~/Library/Jupyter/kernels/django_extensions/kernel.json
On Linux: ~/.jupyter/kernels/django_extensions/kernel.json
We only need to make two changes:
The first is to edit the first value in the "argv" list from "python" to the full address of the python version in your Django virtual environment. E.g.: "/Users/$USERNAME/Documents/PROJECT_FOLDER/venv/bin/python"
Secondly, to the "env" dictionary, add "DJANGO_SETTINGS_MODULE": "mysite.settings", where mysite is the folder that contains your Django settings.
Optionally, change the value of "display_name".
Now when you run a notebook from any directory, choosing the "Django Shell-Plus" kernel will allow your notebooks to interact with Django. Any packages such as pandas will need to be installed in the Django venv.
The following does work for me using win10, Python 3.5, Django 1.10:
Install Python with the Anaconda distribution so Jupyter will be installed as well
Install Django and install django-extensions:
pip install Django
pip install django-extensions
Start a new Django project. You have to do that in that part of your tree of directories which can be accessed by Jupyter later.
django-admin startproject _myDjangoProject_
Start Jypter
navigate Jupyter to the directory myDjangoProject and enter the first/top myDjangoProject-directory
Start within the first/top myDjangoProject-directory a new Jupyter noteboke: new --> Django Shell-Plus
enter and run the following piece of code :
import os
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "myDjangoProject.settings")
import django
django.setup()
Note that this piece of code is the same as in manage.py, and note that "myDjangoProject.settings" points to myDjangoProject/settings.py
Now you can start with examples, e.g.:
from django.template import Template, Context
template = Template('The name of this project is {{ projectName }}')
context = Context({'projectName': 'MyJypyterDjangoSite'})
template.render(context)
Run this command.
PYTHONPATH=/path/to/project/root DJANGO_SETTINGS_MODULE=settings python manage.py shell_plus --notebook
I will add some information to the very complete answer of RobM, for the benefit of the very rare developers that use buildout along with djangorecipe djangorecipe as I do... I refer to jupyter lab as I use that but I think all info can be applied to old jupyter notebooks.
When using buildout you end up with a 'bin/django' handler you'll use instead of 'manage.py'. That's the script that defines the whole path. I added one more part in my buildout.cfg:
[ipython]
recipe = zc.recipe.egg
eggs = ${buildout:eggs}
extra-paths = ${buildout:directory}/apps
initialization = import os
os.environ['DJANGO_SETTINGS_MODULE'] = 'web.settings'
so that another script named ipython will be created in ./bin directory. I point kernelspec to that interpreter. Moreover I use kernel argument rather than "-m", "ipykernel_launcher" so that the kernel definition I use is:
{
"argv": [
"/misc/src/hg/siti/trepalchi/bin/ipython",
"kernel",
"-f",
"{connection_file}",
"--ext",
"django_extensions.management.notebook_extension"
],
"display_name": "Trepalchi",
"language": "python"
}
Due to how the ipython script is created by buildout there's no need to add environmental variables in my case.
As Rob already mentioned, jupiterlab is only installed in one environment where I start it with the command:
jupyter lab
not in the environment of Django project whare I only install ipykernel (that has already a bunch of 20 dependencies).
Since I tend to have quite a lot of projects I find it usefull to have a single point where I start jupyter lab with many links to the projects so that I can reach them easily. Thanks to the extension provided by django_extension I don't need any extra cell to initialize the notebook.
Any single kernel added in this way can be found with the command:
jupyter kernelspec list
And clearly listed in the launcher of jupyter lab
When I try to run my Django app via PyCharm inside of Vagrant, it sends my Windows Path over SSH which then provides an error in the shell:
> ssh://vagrant#127.0.0.1:2200/usr/bin/python -u C:/Projects/dev_project/dev/manage.py runserver 0.0.0.0:8000
bash: line 0: cd:
C:/Projects/dev_project/dev: No such
file or directory /usr/bin/python: can't open file
'C:/Projects/dev_project/dev/manage.py':
[Errno 2] No such file or directory
I created a Python Project in PyCharm (and instantiated a Django Project in a sub-folder) in order to incorporate Vagrant.
dev_project (PyCharm project root)
|--.vagrant
|--dev (Django project root)
|--dev
|--app
|--manage.py
|--Vagrantfile
In Settings
I enabled Django Support (providing the Windows Paths as there is no other option) in Settings > Languages & Frameworks > Django.
The Vagrant Python Interpreter is selected as the Project Interpreter (and Django Console & Python Console)
In Run Configurations
My host is 0.0.0.0 and port is 8000
My Python Interpreter is the Vagrant Environment. I am also adding the Content Roots & Source Roots to the Python Path.
The bizarre problem is it was working fine, and then I exited out and it broke again. Also, I do not have 'Working Directory' explicitly defined anywhere.
This blog article showed me my issue, as I did not have my Path Mappings setup within my Run Configuration.
From the Article (Configure Your Project to Use the Correct Interpreter):
Select Run->Edit Configurations…
Select the configuration on the right PyCharm created for us. In my case it was called “session_tracker”.
Change “Python interpreter” to our newly created one.
A new field will appear. Click the button next to “Path mappings”. A new window will appear to let you create your mappings.
Vagrant shares a local directory with the VM. My Vagrant file is configured so a folder is called “django_shared” in our [VagrantFolder] locally, and “django_shared” in the home directory on the VM are the same. You need to enter the full paths of each of those in the “Edit Path Mappings” window. - Click the + button to create a new mapping, then enter the values for the mapping on each side.
When no other option works I do the following:
Create seperate virtualenv in my local directory
Reconfigure PyCharm to use the interpreter in that venv
Restart PyCharm
Maybe not a very good solution, but works.
I've deployed a Django application on Heroku. The application by itself works fine. I can run commands such as heroku run python project/manage.py syncdband heroku run python project/manage.py shell and this works well.
My Django project makes use of the Python web scraping library called Scrapy. Scrapy comes with a command called scrapy crawl abc which helps me scrape websites I have defined in the scrapy application. When I run a scrapy command such as scrapy crawl spidername on my local machine, the application is able to scrape date and copy it to my database. However when I run the same command on Heroku under a sub-directory of my project directory heroku run scrapy crawl spidername, nothing happens.
I don't see anything in the Heroku logs which can point to where I'm going wrong:
2012-01-26T15:45:38+00:00 heroku[run.1]: State changed from created to starting
2012-01-26T15:45:43+00:00 app[run.1]: Awaiting client
2012-01-26T15:45:43+00:00 app[run.1]: Starting process with command `project/spiderMainDir scrapy crawl spidername`
2012-01-26T15:45:44+00:00 heroku[run.1]: State changed from starting to up
2012-01-26T15:45:46+00:00 heroku[run.1]: State changed from up to complete
2012-01-26T15:45:46+00:00 heroku[run.1]: Process exited
Some additional information:
My scrapy app calls pipelines.py to save the scraped items to the database. In the pipelines.py file, this is what I've written to invoke the Django settings so that I can import my models and save data to the database from the scrapy application.
import os,sys
PROJECT_PATH = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
sys.path.append(PROJECT_PATH)
os.environ['DJANGO_SETTINGS_MODULE'] = 'settings'
Any pointers on where exactly am I going wrong? How do I execute the scrapy command on Heroku such that my application can scrape an external website and save that data to the database. Isn't the way external commands are run in Heroku like - heroku run command?
I'm answering my own question because I discovered what the problem was. Heroku for some reason was not able to find scrapy when I executed the command from a sub-directory and not the top-level directory.
The command heroku run ... is generally run from the top-level directory. For my project which uses scrapy, I was required to go to a sub-directory and run the scrapy command from the sub-directory (this is how scrapy is designed). This wasn't working in Heroku. So I went to the Heroku bash by typing heroku run bash to see what was going on. When I ran the scrapy command from the top-level directory, Heroku recognized the command but when I went to a sub-directory, it failed to recognize the scrapy command. I suppose there is some problem related to the path. From the sub-directory, I had to specify the complete path to scrapy (~/bin/scrapy crawl spidername) to be able to execute it.
To run the scrapy command without going to the Heroku bash manually each time, my work around this problem was that I created a shell script containing the following code and put it under the bin directory of my top-level directory and pushed the changes to Heroku.
bin/scrapy.sh :
#!/usr/bin/env bash
cd ~/project/spiderSubDirectory
~/bin/scrapy $#
After this was done, I could execute $ heroku run scrapy.sh crawl spidername from my local bash. I suppose its not the best solution but this works.
Isn't the way external commands are run in Heroku like - heroku run
appdir command?
It's actually heroku run command. By including your appdir in there, it resulted in an invalid command. Heroku's output doesn't give useful error messages when these commands fail, and instead just tells you that the command finished which is what you're seeing. So for you, just change the command to something like:
heroku run scrapy crawl spidername
On my local machine (Mac OSX 10.6) I wrote a django custom admin command which works great. I can use it both within and outside my project directory just fine. For some reason on my CentOS 5.6 server, it won't work from outside the project directory. This is really annoying since using this custom admin command in a cron job requires it to run from the home directory.
in short:
When I run "python ./manage.py scrape" or "python manage.py scrape", everything is fine.
When I run "python /home/[username]/webapps/myproject/manage.py scrape" or "python myproject/manage.py scrape", I get the following error:
unknown command: 'scrape'
Type 'manage.py help' for usage.
On CentOS, when I run manage.py help inside the project directory, scrape shows up as a command; but if I run it outside the project directory, scrape does not appear as a valid command. On OS-X scrape appears as a valid command regardless of where I run manage.py help from.
Any idea how I can fix this?
I know CentOS ships with Python 2.4, so is your code running on 2.4 or are you using a contained environment, this is usually fixed by adding your PYTHONPATH correctly
import sys
print sys.path
verify such for starters
This should get you up and running: http://djangosnippets.org/snippets/374/
I have a custom admin command named temperature.py which is under /home/user/project/monitor/management/commands. If I change directory to /home/user/ and execute:
user#localhost:~/project$ ./manage.py temperature
It runs ok, and its listed in the available commands. But if I try running it with the absolute path:
user#localhost:/$ /home/user/project/manage.py temperature
It says the command does not exist, it does not show up in the available commands either. I have django 1.2.1 with python 2.6.5 in ubuntu 10.04. Could this be a problem with django? is it the python version? Thanks in advance
Found the reason, it seems that django is looking for the settings under the main directory, if it fails to find one, it will use the defaults. You can change your python path or use this in your manage.py file