Comment threads and annotations for ACCMS and otherwise

The content management system of the Accelerator or otherwise can back multiple websites. To support comment threads or annotations on a web component such as a chart, there is need for a multi-tenant service. This repo defines that service.

Use

Making first public tenant

from tenants.models import Client, Domain

tenant = Client(name="Main", schema_name="public")
Domain.objects.create(
    domain="admin.localhost",
    tenant=tenant,
    is_primary=True
)

Make tenant superuser

from django_tenants.utils import schema_context
from django.contrib.auth import get_user_model

User = get_user_model()

with schema_context('tenant1'):
    User.objects.create_superuser(
        username='admin',
        email='admin@example.com',
        password='admin123'
    )

Codebase overview

Django-based multi-tenant RESTful API designed for managing nested comment threads and annotations, primarily for integration with the Accelerator Content Management System (ACCMS) or similar platforms. It supports multiple websites/tenants via django-tenants, allowing isolated schemas for different clients (e.g., public tenant and custom ones like 'tenant1').

Key components:

• Multi-Tenancy: Uses tenants app with Client and Domain models. Shared apps include tenant management; tenant-specific apps handle comments and auth.
• Comments App: Core functionality for threaded comments. Models include Comment with user, text, parent (for replies), thread_id, likes (ManyToMany), soft-delete flag. Supports nesting, likes/unlikes, email notifications to thread participants on new comments/likes.
• Authentication: Token-based (DRF), dj-rest-auth, allauth for email verification (mandatory), Google OAuth, anonymous login (creates user with random password). Email auth preferred over username.
• API Features: DRF ViewSet for CRUD on comments (list roots only by default, filter by thread prefix, search text). Custom actions: like/unlike, soft-delete. OpenAPI schema via drf-spectacular.
• Other: PostgreSQL backend, production-ready (Whitenoise, Gunicorn, CORS), email via SMTP/console, static files handling. Docker support implied.

The API is exposed for frontend integration (e.g., charts/annotations), with endpoints for threads, replies, interactions. No frontend code; pure backend service.