2026.04.09

Changed

• Dynamic polling — The coordinator now computes the next poll time based on collection dates, scheduling the next fetch for the day after the earliest upcoming collection. This reduces requests from ~1/day to ~1/week
• Removed user-configurable scan interval — The scan_interval config option has been removed from the setup and options flows since polling is now automatic. Existing config entries with a stored value are unaffected (the value is simply ignored)
• Renamed DEFAULT_SCAN_INTERVAL to FALLBACK_SCAN_INTERVAL — Clarifies that the 24-hour interval is only used until the first successful data fetch
• Replaced verbose_logging toggle with native HA debug logging — Removed the custom verbose_logging config option from the setup and options flows. All diagnostic messages now use standard _LOGGER.debug() calls, which are controlled via Home Assistant's built-in "Enable debug logging" button (Settings → Devices & Services → integration menu) or logger configuration in configuration.yaml

Fixed

• Excessive proxy requests — Switched sensor base class from SensorEntity to CoordinatorEntity, which sets should_poll = False. Previously HA's default ~30s entity polling called async_request_refresh() on every cycle, bypassing the coordinator's update interval and causing ~2-3 HTTP requests per minute
• Redundant startup request — Removed the _validate_property_id() HTTP call from __init__.py that duplicated the coordinator's first refresh, eliminating a second proxy hit on every startup and options-change reload

Temporary Addition

• Cloudflare Worker proxy support — Optional proxy URL and token can be configured to route requests through a Cloudflare Worker, providing a fallback when direct requests are blocked by Auckland Council rate limiting. If you get HTTP 406 errors please contact me for more information.