As you start to develop more intricate Django applications, you might find it more convienent to install Django on your local computer and then use github to move your tested code into your PythonAnywhere server so you it is available on the Internet for grading or otherwise sharing.
There are many sources of tutorial material on how to install Python3 and Django on your computer. Since the rest of this material uses the Mozilla Developer Network tutorials, you might as well use it:
You can use a virtual environment or if you have a suitable version of Python 3.x on your
computer and can use
pip3 to install a suitable version of Django for your whole computer,
there is no need to put things in a virtual environment.
Also you should install git on your computer using any of the good tutorials out there.
You know Python and Django are correctly installed when these commands show reasonable version numbers:
$ python3 --version Python 3.6.0 $ python3 -m django --version 2.0.5
If the above does not work on Windows, try:
> py --version Python 3.6.0 > py -m django --version 2.0.5
You also need to have
git installed and available in the shell / command line.
There are a number of advantages to doing development work locally:
You never have to
Reload your application. The Django
runserver process monitors
changes to your files and completely restarts itself as soon as any file changes in your
project. This makes for much quicker edit-test cycles.
You can use a fancy text editor like VScode, Atom, or Sublime.
No more need to change the WGSI configuration file when you want to switch between your project and some sample code - you can even run more than one application at the same time on different ports.
You can put debug
print() statements and they come right out without having to look
at the error or server logs. Error tracebacks and error logs come right out.
These instructions assume you that you are already set up using Github and PythonAnywhere and want to edit your applications on your computer and present them on PythonAnywhere. But you can always use the ngrok application to submit your assignments to the autograder if you like.
We suggest you make a folder somewhere to store all of your Django projects. This folder should be easy to find so you can use your cool VSCode, Atom, or Sublime text editor.
Linux / MacOS / Windows bash shell:
cd ~ cd Desktop mkdir django
Windows Command Line:
cd cd Desktop mkdir django
Then lets checkout the dj4e-samples repo and get things started:
cd Desktop # If not already there cd django git clone https://github.com/csev/dj4e-samples cd dj4e-samples python3 manage.py migrate # Makemigrations is already in git python3 manage.py createsuperuser python3 manage.py runscript gview_load python3 manage.py runscript many_load
To run the server get into the folder with
manage.py and then:
python3 manage.py runserver
Then navigate to http://localhost:8000 to see the page.
Then just for fun, open a second terminal / shell / command line and:
cd Desktop cd django cd dj4e-samples python3 manage.py runserver 8001
Then navigate to http://localhost:8001 to see the page.
You can abort the
runserver applications in the command line, switch to
a new folder and start runserver again.
This section shows how to get a Django application working - in the next section we show you to work on your code on two computers and move the changes back and forth.
If you want, you can keep a local copy your
django_projects synchronized with your
copy on PythonAnywhere using
Before you start doing this, make sure that your code in the PythonAnywhere shell is fully checked in to GitHub:
$ cd ~/django_projects $ git status On branch master Your branch is up-to-date with 'origin/master'. nothing to commit, working directory clean
If you have any outstanding git modifications on PYAW - clean it up and push it to the repo.
If you have not yet uploaded your
django_projects folder to github, first follow the
these instructions to get your application uploaded to github.
Once your application is in github, you can simply check it out to your laptop computer using commands like:
cd ~ # or simple 'cd' for Windows cd Desktop cd django git clone https://github.com/drchuck/django_projects.git
drchuck with your github account. This should bring a copy of your
application from github down to your computer and store it in the folder
Note that you should keep the
django_projects next to each
other since they are both in github.
On your laptop:
So you are on your local laptop / computer and are making changes
cd ~ # or simple 'cd' for Windows cd Desktop cd django cd django_projects cd locallibrary # Or whatever project you want to work on # edit some files :) python3 manage.py makemigrations # If you changed your models python3 manage.py migrate # If you changed your models.py git status git add .... git commit -a git push
Then you can go into PythonAnywhere in a bash shell and type:
cd ~/django_projects/locallibrary git pull python3 manage.py migrate # If the pull included new migrations
Then in the "Web" tab reload your application and visit it.
As long as you follow the pattern of doing
git push from your laptop/desktop and
from PythonAnywhere, things will go very smoothly.
If you edit two places and push from one of the places, the push will work - but the push won't work from the second place and pull won't work either becausee you have local changes. If this is what you did, there is a simple workaround. On the system where you have un-pushed changes and want to do a pull before pushing, do this:
git stash git pull git stash apply
This takes your un-pushed changes and hides them in the "stash", allowing the
git pull to
work and then the
stash apply re-modifies the files.
Most of the time this works if all you did is edited two places and tried to push from both.