Merge pull request #154 from ddxv/main

New blog post, some minor UIs

apps/docs/blog/posts/svelte-analytics-dash.md

@@ -0,0 +1,40 @@
+---
+draft: false
+
+
+date: 2024-10-06
+
+authors: 
+ - ddxv
+
+categories:
+ - Updates
+---
+
+# Analytics Dash: Switch from Superset to Custom Svelte
+
+After quite a long break I got back into working on Open Attribution. My point of re-entering was to finally start working on the `admin-db` which would house all the user facing settings and analytics dashboard. 
+
+This very quickly brought up the issue of Superset, a no code drop in fully customizeable analytics dashboard. I was loving it, but trying to get it to work for my usecase always ended up being a time suck, and as I moved towards a more managed approach Superset wouldn't be a good fit. Maybe we can go back to it in the future.
+
+So with that, we changed the frontend dash quite a bit. The analytics dashboard now consists of 3 apps:
+
+ 1. `admin-db`: a postgres database for managing the tables for things like users, authentication, netowrks, apps etc. 
+ 2. `dash-backend`: a python Litestar backend API, currently only exposed to localhost
+ 3. `dash-frontend`: a svelte frontend for managing the analytics dashboard. We also switched from `skeleton` to `shadcn` due to wanting to start building this project with Svelte 5 to be well positioned for the future (and because I wanted to try it out).
+
+## Docker Re-Do
+
+Meanwhile, I've also completely revamped the docker environment for development / deployment. A big part of this was due to removing Superset, which had come with a lof of it's own docker baggage, and more just because I'm still learning Docker.
+
+
+## More to Do
+
+- [ ] Tons more work to finish MVP of frontend: authentication, reporting levels, creatives, spend, revenue, revents
+- [ ] Rewrite Docker documentation
+- [ ] Remove old non-Docker documentation
+- [ ] Create testing section for frontend
+- [ ] Pull data from first ad network
+- [ ] Start iOS and/or Android SDK
+
+Some features are live at [demo.openattribution.com](http://demo.openattribution.com).

apps/postback-api/tests/generate_impressions_and_clicks.py

@@ -1,3 +1,5 @@
+"""Generate impressions and clicks for testing."""
+
 import random
 import time
 import uuid
@@ -15,13 +17,18 @@
 
 ADS = ["Hi!", "NewVideo123"]
 
+INSTALL_RATE = 0.3
+CLICK_THROUGH_RATE = 0.5
+D1_APP_OPEN_RATE = 0.4
+
 
-def main(endpoint:str) -> None:
+def main(endpoint: str) -> None:
+ """Start here."""
  logger.info("Start continuous generate")
  while True:
  for network in NETWORKS:
  for app in APPS:
- if random.random() < 0.3:
+ if random.random() < INSTALL_RATE: # noqa: S311
  # Simulate organic install and return
  ifa = str(uuid.uuid4())
  make_inapp_request(
@@ -44,8 +51,8 @@ def main(endpoint:str) -> None:
  endpoint=endpoint,
  )
  # Decide if a click should be generated
- if random.random() < 0.5: # % chance for a click
- time.sleep(random.uniform(0.1, 1.0)) # Simulate delay
+ if random.random() < CLICK_THROUGH_RATE: # noqa: S311
+ time.sleep(random.uniform(0.1, 1.0)) # noqa: S311
  impression_or_click(
  mytype="clicks",
  myapp=app,
@@ -55,8 +62,8 @@ def main(endpoint:str) -> None:
  myad=ad,
  endpoint=endpoint,
  )
- if random.random() < 0.5:
- time.sleep(random.uniform(0.1, 1.0)) # Simulate delay
+ if random.random() < D1_APP_OPEN_RATE: # noqa: S311
+ time.sleep(random.uniform(0.1, 1.0)) # noqa: S311
  make_inapp_request(
  event_id="app_open",
  myapp=app,