LNLDB

Intro

LNLDB now runs under Python3.x and Django 2.2 or later.

To Install (Testing)

Install required system packages

You are going to need some basic Python/Git tools to run the code. The most important are Python3 and virtualenv (which allows you to install python libs without root). The rest are to compile the python binary libraries, namely MySQL and image manipulation.

Linux:

sudo apt-get install python3 python3-dev python-pip python-virtualenv git libtiff5-dev libjpeg8-dev zlib1g-dev libfreetype6-dev liblcms2-dev libwebp-dev libmysqlclient-dev

macOS:

curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
python get-pip.py
pip install virtualenv

It is certainly possible to run via Windows practically without changes, but figuring out that setup is left as an excercise for the user (hint: use pycharm).

Get the sources

If you're reading this outside of Github, you can skip this.

git clone https://github.com/WPI-LNL/lnldb.git
cd lnldb

Install Python packages

This uses a lot of library functionalities in the form of plugins. We will make a virtualenv to keep all of this away from the system Python installation (ie. don't need root), and install our packages directly to the env folder.

virtualenv env
source env/bin/activate
pip install -r requirements_debug.txt

Initialize the database

The first line makes/loads the actual database schema, by walking through all of the previous schemas and making necessary changes to the database one-by-one so that no data is ever lost regardless of versions. On your machine, the database will by default be an Sqlite file in the runtime folder, on the server, it's WPI's hosted MySQL server.

The second line will populate any data with a useful initial value. Examples include groups, permissions, locations, etc. If you have updated these or have existing data, don't run this. You can also use this method to load a backup from the production database (dbbak on server).

The third will create an account to let you log in once you've started the server.

python manage.py migrate
python manage.py loaddata fixtures/*.json
python manage.py createsuperuser

Run the tests

This app includes a number of self-checks to sanity test new code. All new patches are expected to have tests included, and will be checked automatically when pushed to Github. Also try using the '-n' flag to speed it up.

python manage.py test

Run it!

python manage.py runserver or python manage.py shell_plus Note that in future sessions, you must first call source env/bin/activate to set up the local path

Notes

All server-specific keys, directories, etc. haven't and won't be included. The app checks for a file at lnldb/local_settings.py, where those and any platform-specific options can be included.

Installing on heroku

You can now deploy a test server using one button! Well.... almost. You still need to figure out s3, and make a login.

Fill in the allowed_hosts field with eg. 'yourappname.herokuapp.com,yourcustomdomain.com'
(Not required unless storing uploads) Make an S3 bucket. Just leave everything disabled. Copy the bucket name and region id (eg. us-east-1) into the form.
(Not required unless storing uploads) Make an IAM identity to log in under. Copy the id and secret key into the form.
(Not required unless storing uploads) Add a IAM Inline Policy to grant permission to read/write on the bucket. A good base is below.
Hit deploy and wait.
heroku run python manage.py createsuperuser

Minimal IAM Policy

{
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "s3:GetObject",
                "s3:PutObject",
                "s3:PutObjectAcl",
                "s3:ListBucket"
            ],
            "Resource": [
                "arn:aws:s3:::yourbucket",
                "arn:aws:s3:::yourbucket/*"
            ]
        }
    ]
}

WPI-LNL / lnldb

readme