BUG: Devices with many active sensors a large number of `max_trips` become unavailable

[deconstructionalism]

When I set up a device to track between 2 stations (on a Trip basis) and max_trips set to 7, the sensors periodically become unavailable.

Config

• Newtonville to South Station
  • arrive: Newtonville
  • depart: South Station
  • max trips: 7
• South Station to Newtonville
  • arrive: South Station
  • depart: Newtonville
  • max trips: 7

I am currently have these sensors active:

• from
• to
• train
• alerts
• departure time
• departure delay
• arrival time
• arrival delay
• status
• alerts

Issue

The sensors start the day receiving data but then eventually, the number of indicies in the Next attribute for various sensors goes down towards 0, or the sensors themselves become unavailable. Eventually, either the main sensor comes on the next day, or late in the evening, and sometimes more of the Next entries also come back.

My hunch is that this is due to the polling rate which would be:

I'm being very rough with this, assuming one request per sensor per train

2 devices * 10 sensors / device * 8 requests / sensor * 3 requests / minute = 480 requests / minute

However, this is well under the 1000/min X-API rate limit so I may be off about this.

Demo

Screen.Recording.2025-06-26.at.4.36.28.PM.mov

[chiabre]

It’s been a while since I last worked on this code, so I’ll need to double-check to be sure. However, I believe the behavior you're seeing may be due to the extension filtering trips based on the current operating day (from now until 2 AM). At the end of the day, the trip list may briefly drop to zero and then repopulate as the new operating day begins.

Regarding rate limiting, I implemented all the caching mechanisms I could think of. If you’re using a token, you should be safe in that regard.

Unfortunately, I don’t currently have the time to reproduce the issue on my end. If you can share the logs here, I’ll be able to get a better understanding of what’s happening.

[deconstructionalism]

I ended up using a different implementation without the line numbers, so if you want, just close this issue unless someone else runs into this issue. Thanks for your work on this project