Files
maru-hleda-byt/project/todo.md
Jan Novak c2bc3f452f
All checks were successful
Build and Push / build (push) Successful in 13s
Unify server, persist ratings via API, refresh scraper data
- Replace split setup (ratings_server.py on :8081 + http.server on :8080)
  with a single combined Flask server (server.py) on :8080 that serves
  static files and the /api/ratings GET/POST endpoints
- Ratings are now persisted server-side: mapa_bytu.html loads ratings
  from GET /api/ratings on startup (API as source of truth) and POSTs
  on every change — enables cross-browser and cross-device state sharing
  while keeping localStorage as a synchronous read cache
- Dockerfile: install flask, copy server.py instead of ratings_server.py,
  expose only port 8080
- entrypoint.sh: start single server process instead of two
- Makefile: add serve / serve-debug targets for local development
- scrape_psn.py: fix log label, add --max-pages stub arg for CLI parity
- Refresh all scraped property data

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-25 13:51:02 +01:00

669 B

Features to add

  • testing: run very limited scrape: 1 page, 10 properties so that we can do some validations on it
  • makefile: add target to run locally (with the webserver)
  • feature: store date of the last scrape somewhere, so that we know how fresh data are
  • feature: ?? mark property with scrape when it first appeared - we might be able to look at recent diffs only

code organization

  • prepare reasonable code structure from the bunch of "random" files

documentation

  • precisely document original intent of the app (Maru has to provide this)

  • prepare production run
    • probably in home kubernetes
    • maru-hleda-byt.lab.home.hrajfrisbee.cz