Geoshop

Requirements

• PostgreSQL >= 11 + PostGIS
• Python >= 3.9

Getting started

Fork and clone this repository. Make a copy of the .env file and adapt it to your environment settings:

cd back
cp .env.sample .env.local
cd ..

Database

The best is to backup and restore a production db. Otherwise, to start from scratch follow this.

Create a geoshop user if not existing yet, set your password according to your env.local:

CREATE ROLE geoshop WITH LOGIN PASSWORD <password>;

Then, set up a database in psql:

CREATE DATABASE geoshop OWNER geoshop;
REVOKE ALL ON DATABASE geoshop FROM PUBLIC;

Then connect to the geoshop database \c geoshop and create extensions:

CREATE EXTENSION postgis;
CREATE EXTENSION unaccent;
CREATE EXTENSION "uuid-ossp";
CREATE SCHEMA geoshop AUTHORIZATION geoshop;
CREATE TEXT SEARCH CONFIGURATION fr (COPY = simple);
ALTER TEXT SEARCH CONFIGURATION fr ALTER MAPPING FOR hword, hword_part, word
WITH unaccent, simple;

Now that the database is ready, you can start backend either with Docker or not.

Backend with docker

To ease debug, you can add DEBUG=True to your env.local file. Never add this to a container exposed on internet.

Build backend image:

cd back
docker build -t sitn/geoshop-dev-api --build-arg ENV_FILE=.env.local .

If you started with an empty database, run this once in order to create admin account:

docker run --rm --name geoshop --env-file=.env.local sitn/geoshop-dev-api python manage.py fixturize

Now you can run it with:

docker run -d --rm --name geoshop --env-file=.env.local -p 8000:8000 -v ${PWD}:/app/geoshop_back sitn/geoshop-dev-api gunicorn --reload wsgi -b :8000

You should be able to visit the API at http://localhost:8000.

Run tests:

docker run --rm --env-file=.env.local -v ${PWD}:/app/geoshop_back sitn/geoshop-dev-api python manage.py test api

Make messages for translation:

docker run --rm --env-file=.env.local -v ${PWD}:/app/geoshop_back:rw sitn/geoshop-dev-api python manage.py makemessages -l fr

Stop the server:

docker stop geoshop

Backend without docker on Windows

⚠️ No longer maintained: Installing GDAL in windows is really painfull, use docker for backend.

If not using docker, additionnal packages are required:

• pipenv (pip install pipenv)
• GDAL 2.4 (see instructions below to install it in your venv)

Install the app. If you want your venv to be inside your project directory, you need to set PIPENV_VENV_IN_PROJECT environment variable, otherwise it'll go to your profile, if you want DEBUG to be enabled, change it in settings.py file but never commit it with debug enabled:

cd back
$env:PIPENV_VENV_IN_PROJECT="True"
pipenv install --dev           # installs everything needed
pipenv shell                   # activates venv and reads .env file

Installing GDAL on Windows, only once per machine

Download the GDAL 2.4 wheel (3.X isn't supported yet by Django) on this page: https://www.lfd.uci.edu/~gohlke/pythonlibs/#gdal. For example, if you have Python 3.6 64bit, choose GDAL‑2.4.1‑cp36‑cp36m‑win_amd64.whl. Then install it weather system wide or in your venv (the example below will show the venv variant and expects you have your venv activated):

pip install path\to\your\GDAL-2.4XXXX.whl

You'll then need to add GDAL dll to your PATH if you installed it system wide. You can get the dll path with:

python

from pathlib import Path, PureWindowsPath
from osgeo import gdal

print(PureWindowsPath(gdal.__file__).parent)

Otherwise, if you installed it in your venv, configure .env properly.

Migrate and run

You should now be able to run migrations:

python manage.py migrate

If you're starting with a fresh new database you'll need to create an user or restore a dump:

python manage.py createsuperuser --email admin@example.com --username admin

Your database should be ready, now you can run the backend:

python manage.py runserver

Translations can be generated, static files collected with:

python manage.py compilemessages --locale=fr
python .\manage.py collectstatic

Then, set DEBUG to True in back/settings.py and you can finally run the server:

python manage.py runserver

⚠️ DO NOT COMMIT settings.py with DEBUG set to True !

Run tests

python manage.py test api

Frontend

Install the current LTS version of Nodejs.

Install @angular/cli and typescript globally

npm install -g @angular/cli typescript

Install the dependances of the frontend

cd front
npm install

Edit the settings, the file is /front/copy-config.ps1

Typicall values for dev purposes are:

  "apiUrl": "https://sitn.ne.ch/geoshop2_prepub_api",
  "mediaUrl": "https://sitn.ne.ch/geoshop2_prepub_media/images",

To start the debug of the frontend

npm start

1. Will run the batch /front/copy-config.ps1
2. Will automatically run chrome with insecure flags to allow CORS request
3. Chrome browser will be waiting with Geoshop2

Deploy

Create an env.prod file base on env.sample.

Application deployment

.\scripts\4_deploy_prod.ps1

Create a scheduled task that runs scripts/geoshop_clean_orders.ps1 every month.

More info on bookstack

Upgrading front-end packages

Go to https://update.angular.io/?v=16.0-17.0 and check there are not manual steps to be done.

Then follow instructions. After upgrading, some packages will need to be upgraded according to angular version you will be using. For instance, if you'll upgrade to version 17, you need to target version 17 for these packages:

ng update @ngrx/store@17 angular-split@17

After, you can update other packages:

ng update lodash-es rxjs tslib zone.js

Finally, update those related to OpenLayers:

ng update ol ol-ext ol-geocoder proj4

Then all the front-end tests should be done.