Commit Graph

5 Commits

Author SHA1 Message Date
John Mueller 21820a7c84
Include note about Daylight Savings Time duplication/skipping (#3771)
It's probably clear to most people that setting up a trigger for the 31st of every month will fail in February, but apparently triggers between 1am and 3am are also fragile once a year. For non-disruptive or regular triggers, this might not be an issue -- if you want to turn off a device at 2am using on_time, you should know that it will fail once a week. Similarly, if you want to do a daily trigger at 1am, you should know that it will trigger twice a day once a year. 

This does not seem super-obvious until you've run into it, so I propose adding a note to the documentation. It also doesn't feel like something that should be changed in the functionality to match cron's behavior, given that it would change the functionality of existing triggers. 

I'm not invested in the exact wording / location; I suspect pulling it out a bit into a more visible place makes it less likely for people to miss.

(Brought to you by a trigger that turns off an expensive device at 2am, only for it to run a full day instead :-)).
2024-06-04 15:18:47 +02:00
sebPomme a3f6465de3
[time] Fix cron typo to real every 5min (#3017)
Co-authored-by: H. Árkosi Róbert <robreg@zsurob.hu>
2023-06-27 18:59:27 +02:00
TheGroundZero dcd8c1d058
Add TZ timezone string examples (#2806) 2023-04-11 10:25:23 +02:00
H. Árkosi Róbert 4131c3f656
Update index.rst (#2738) 2023-03-11 12:25:36 +01:00
Jesse Hills 150d149d60
Restructure time docs into its own folder and split (#2566) 2022-12-23 11:57:01 +13:00