Tapir Member & Subscription Management System

Tapir is a member and subscription management system to be used by WirGarten e.V.

Tapir has a trunk, but not quite such a beautiful one as Mme. l'élephan. Tapir is badass, but not quite as badass as the other animals. Let's teach Tapir some tricks!

Getting started

docker-compose up

This starts a container with an LDAP server and automatically loads the test data into the LDAP.

Next, set up the test database and load test data

# Create tables
docker-compose exec web poetry run python manage.py migrate

# Import master data
docker-compose exec web ./import-masterdata.sh

# Import configuration parameter definitions
docker-compose exec web poetry run python manage.py parameter_definitions

# Load admin (password: admin) account
docker-compose exec web poetry run python manage.py loaddata admin_account

# Load lots of test users
docker-compose exec web poetry run python manage.py populate --reset_all

Django Shell

docker-compose exec web poetry run python manage.py shell_plus

LDAP

For reading or modifying the LDAP, Apache Directory Studio is pretty handy.

Running tests

For running the test should have a clean openldap container with the test data.

docker-compose up -d openldap

Then, run the tests.

docker-compose run web poetry run pytest

Selenium Tests

You can connect to the selenium container via VNC for debugging purpose. The address is localhost:5900, password : secret

Model

Generate Class Diagram

Run the following command to generate a class diagram:
docker-compose exec web sh -c "apt install -y graphviz graphviz-dev && poetry run pip install pygraphviz && poetry run python manage.py graph_models -a -g -o models.png"

Vocabulary

A few definitions to help newcomers understand the model classes.

Class	Definition
ShareOwnership	Represents a person that is owning at least one share.
TapirUser	Represents a person with a user account. Accounts are linked between Tapir and the Wiki for example. Gets created when the member becomes active (part of the shift system etc.), but can become inactive.

Translations

To generate the translation files, first use "makemessages" and specify the language you want to generate:

docker-compose exec -w /app/tapir web poetry run python ../manage.py makemessages -l de

Update tapir/translations/locale/de/LC_MESSAGES/django.po with your translations.

For the changes to take effect, restart the Docker container. This will run manage.py compilemessages automatically.

Configuration Parameters

We have introduced a generic way to configure the application with key-value pairs.

Currently the datatypes string, float, integer and boolean are supported.

You can retrieve parameter values in Python in the following way:

from tapir.configuration.parameter import get_parameter_value
from tapir.wirgarten.parameters import Parameter

get_parameter_value(Parameter.SITE_NAME) # --> WirGarten Lüneburg eG

There is also a template tag available for reading parameter values:

{% load configuration %}

{{ 'wirgarten.coop.site_name' | parameter }}

How to define parameters

Best practice for those parameters is to create a <app>/parameters.py file per app.

It contains constants for parameter names and categories (so it is easy to find usages in the IDE), and the actual parameter definitions (with a description, initial value, datatype).

When running the manage.py parameter_definitions command, all instances of TapirParameterDefinitionImporter will be found and imported through the import_definitions() function.

The key should have the following format: <app>.<category>.<name>

Example (see wirgarten/parameters.py for the full example):

# Constants for parameter categories
class ParameterCategory:
    SITE = "Standort"
    # [...]

# Constants for parameter names
class Parameter:
    SITE_NAME = "wirgarten.site.name"
    SITE_EMAIL = "wirgarten.site.email"
    HARVEST_SHARES_SUBSCRIBABLE = "wirgarten.harvest.harvest_shares_subscribable"

    # [...]

# Parameter definitions (for automatic import)
class ParameterDefinitions(TapirParameterDefinitionImporter):
    def import_definitions(self):
        
        parameter_definition(
            key=Parameter.SITE_NAME,
            datatype=TapirParameterDatatype.STRING,
            initial_value="WirGarten Lüneburg eG",
            description="Der Name des WirGarten Standorts. Beispiel: 'WirGarten Lüneburg eG'",
            category=ParameterCategory.SITE,
            order_priority=1000,
        )

        parameter_definition(
            key=Parameter.SITE_EMAIL,
            datatype=TapirParameterDatatype.STRING,
            initial_value="[email protected]",
            description="Die Kontakt Email-Adresse des WirGarten Standorts. Beispiel: '[email protected]'",
            category=ParameterCategory.SITE,
            order_priority=900,
            meta=ParameterMeta(
                validators=[EmailValidator()]
            )
        )
        
        parameter_definition(
            key=Parameter.HARVEST_SHARES_SUBSCRIBABLE,
            label="Ernteanteile zeichenbar",
            datatype=TapirParameterDatatype.INTEGER,
            initial_value=1,
            description="Wenn aktiv, dann sind Enteateile von Mitgliedern zeichenbar.",
            category=ParameterCategory.HARVEST,
            meta=ParameterMeta(
                options=[
                    ( 2, "automatisch"),
                    ( 1, "zeichenbar"),
                    ( 0, "nicht zeichenbar"),
                ]
            )
        )
        
        # [...]

if order_priority (type int) is provided, the item will be placed
at the beginning of the category, with high order_prority values first. Items which do not provide order_priority will be ordered alphabetically after those which provide the attribute.

ParameterMeta

The meta attribute can be set for advanced configuration.

options is defined as array of 2-tuple
which elements are defined as (value, label) pairs,
with value of type TapirParameterDatatype
and label of type string.
The programmer has to make sure, the value of value is never None

validators is an array of callables, this means you can pass any function or Django Validator. Useful Django valdators are EmailValidator, MinValueValidator, MaxValueValidator, RegexValidator, URLValidator.

UI

A simple UI for editing parameter values can be found under "Administration -> Configuration" (permission: coop.admin) . The parameters are grouped by the category.

Modal Forms

We have the possibility to use Django Forms in Bootstrap modals.

Example:

1) Define the view

views.py

# make sure your function signature is the same -> your_function(request, **kwargs)
def get_product_type_edit_form(request, **kwargs):
    return get_form_modal(
        request=request,
        
        # your form class (not instance!!)
        form=ProductTypeForm, 
        
        # the handler is called if the form is submitted and valid
        handler=lambda form: save_product_type(form.cleaned_data), # the handler will be called on success
        
        # This callback function creates the redirect url.
        # The result of the handler function is passed, so you could include the ID for newly generated entities
        redirect_url_resolver=lambda data: f"""{reverse_lazy(PAGE_ROOT)}?{request.environ["QUERY_STRING"]}""",
        **kwargs,

2) Add URL routing for the view

urls.py

    # [...]
    
    path(
        "product/<str:periodId>/typeadd",
        get_product_type_edit_form, # this is the function from above
        name="product_type_edit",
    ),
    
    # [...]

3) Show the modal on button click

  <button onclick="FormModal.load(view_url, title)">
    Open Modal        
  </button>