Organize settings

Author: webstackdev

.gitignore

@@ -6,6 +6,7 @@ __pycache__/
 .venv/
 venv/
 .env
+celerybeat-schedule*
 db.sqlite3
 staticfiles/
 

celerybeat-schedule

@@ -1,27 +1,26 @@
 x-app-common: &app-common
   image: newsletter-maker-app:dev
   pull_policy: never
+  build:
+    context: .
+    dockerfile: docker/web/Dockerfile
   env_file:
     - .env
-  environment:
-    DATABASE_URL: postgresql://newsletter:newsletter@postgres:5432/newsletter_maker
-    REDIS_URL: redis://redis:6379/0
-    QDRANT_URL: http://qdrant:6333
   volumes:
     - .:/app
     - static_data:/app/staticfiles
 
+x-shared-env: &shared-env
+  DATABASE_URL: postgresql://newsletter:newsletter@postgres:5432/newsletter_maker
+  REDIS_URL: redis://redis:6379/0
+  QDRANT_URL: http://qdrant:6333
+
 services:
   django:
     <<: *app-common
-    build:
-      context: .
-      dockerfile: docker/web/Dockerfile
     environment:
       BOOTSTRAP_APP: "true"
-      DATABASE_URL: postgresql://newsletter:newsletter@postgres:5432/newsletter_maker
-      REDIS_URL: redis://redis:6379/0
-      QDRANT_URL: http://qdrant:6333
+      <<: *shared-env
     command: ["gunicorn", "newsletter_maker.wsgi:application", "--bind", "0.0.0.0:8000", "--workers", "2"]
     depends_on:
       postgres:
@@ -40,9 +39,7 @@ services:
     <<: *app-common
     environment:
       BOOTSTRAP_APP: "false"
-      DATABASE_URL: postgresql://newsletter:newsletter@postgres:5432/newsletter_maker
-      REDIS_URL: redis://redis:6379/0
-      QDRANT_URL: http://qdrant:6333
+      <<: *shared-env
     command: ["celery", "-A", "newsletter_maker", "worker", "--loglevel=info"]
     depends_on:
       postgres:
@@ -54,9 +51,7 @@ services:
     <<: *app-common
     environment:
       BOOTSTRAP_APP: "false"
-      DATABASE_URL: postgresql://newsletter:newsletter@postgres:5432/newsletter_maker
-      REDIS_URL: redis://redis:6379/0
-      QDRANT_URL: http://qdrant:6333
+      <<: *shared-env
     command: ["celery", "-A", "newsletter_maker", "beat", "--loglevel=info"]
     depends_on:
       postgres:

celerybeat-schedule-shm

@@ -0,0 +1,28 @@
+# File Organization
+
+In Django, you have one **project** (the container for your settings and main routing) and one or more **apps** (reusable modules that handle specific logic).
+
+## File Breakdown: `newsletter_maker/` (The Project)
+
+This is the "brain." It contains your global settings, main URL configuration, and WSGI/ASGI entry points for the server.
+
+- **`settings.py`**: All configuration (Database URLs, installed apps, Celery broker settings).
+- **`urls.py`**: The "main" URL file that imports the `api_urls.py` and `core.urls`.
+- **`wsgi.py` / `asgi.py`**: The interface Gunicorn uses to run your app.
+- **`celery.py`**: Where Celery is initialized for the project.
+
+## File Breakdown: `core/` (The Application)
+
+This is where the actual "features" live. Django encourages putting logic into apps so you could, in theory, pluck this `core` folder out and drop it into a different project.
+
+- **`models.py`**: The most important file. It defines your database schema using Python classes.
+- **`serializers.py`**: Part of **DRF** (Django REST Framework). It converts your `models.py` data into JSON for the API.
+- **`tasks.py`**: Contains your **Celery** background tasks (e.g., the actual code that sends the newsletter).
+- **`api.py` & `api_urls.py`**: These likely house your DRF logic (ViewSets) and the specific URL paths for your API endpoints.
+- **`admin.py`**: Configures how your models look in the built-in Django `/admin` interface.
+- **`views.py` & `urls.py`**: Handle standard web requests and map them to templates.
+- **`embeddings.py`**: Likely a custom file for your specific app (given the `qdrant` service you have), probably handling Vector Search or AI logic.
+- **`migrations/`**: A history of your database changes.
+- **`templates/` & `static/`**: Your HTML files and CSS/JS/images.
+- **`management/`**: Contains custom terminal commands (e.g., `python manage.py my_custom_command`).
+- **`tests.py`**: Where your automated tests live.

celerybeat-schedule-wal

@@ -114,15 +114,3 @@ If your primary bottleneck is the time spent reading 50 articles to find the 5 t
 - **Make.com:** Visual AI workflow automation platform. You can build a scenario that watches your RSS feeds and social mentions, sends them to **OpenAI/Claude** to generate a 2-sentence summary and a "relevance score," and then dumps the winners into a **Notion** board or **Airtable** for your final review.
 
 - **n8n**: A low-code tool frequently used to build custom newsletter summarizers. You can create a workflow that monitors your RSS feeds, sends new content to an LLM (like GPT-4 or Claude) for a technical summary, and stores the results in a JSON file or Airtable.
-
-## Containers
-
-celery-beat
-celery-worker
-django
-nginx
-postgres
-qdrant
-redis
-
-newsletter-maker

docker-compose.yml

@@ -0,0 +1,4 @@
+from .base import *  # noqa: F403
+from .ai import *  # noqa: F403
+from .celery import *  # noqa: F403
+from .logging import *  # noqa: F403

docs/FILES.md

@@ -0,0 +1,14 @@
+import os
+
+from .base import env_bool
+
+
+QDRANT_URL = os.getenv("QDRANT_URL", "http://qdrant:6333")
+OPENROUTER_API_KEY = os.getenv("OPENROUTER_API_KEY", "")
+OPENROUTER_API_BASE = os.getenv("OPENROUTER_API_BASE", "https://openrouter.ai/api/v1")
+OPENROUTER_APP_URL = os.getenv("OPENROUTER_APP_URL", "")
+OPENROUTER_APP_NAME = os.getenv("OPENROUTER_APP_NAME", "newsletter-maker")
+EMBEDDING_PROVIDER = os.getenv("EMBEDDING_PROVIDER", "sentence-transformers")
+EMBEDDING_MODEL = os.getenv("EMBEDDING_MODEL", "sentence-transformers/all-MiniLM-L6-v2")
+EMBEDDING_TRUST_REMOTE_CODE = env_bool("EMBEDDING_TRUST_REMOTE_CODE", default=False)
+OLLAMA_URL = os.getenv("OLLAMA_URL", "http://localhost:11434")