Add syntactic sugar in rules to check if item maintains state for specified duration

[boc-tothefuture]

Add syntactic sugar in rules to check if item maintains state for specified duration

[5iver]

Thank for for submitting this feature request, but the core helper libraries are only a temporary crutch until the rule engine has been further built out with features. It would be possible to implement something like this in Jython, but by implementing it in the rule engine, it could be used in rules built through the UI rule editor and any scripting language. There have been discussions about implementing Actions or Profiles as a replacement for the Expire binding functionality in OH 3.0.

That said, Area Triggers and Actions implements similar functionality by just adding some Item metadata.

[mjcumming]

@boc-tothefuture what is the use case for this scenario?

[boc-tothefuture]

@mjcumming -
In this particular case I want to send a notification if any of my ESPHome devices go offline for more than 30 minutes. I have done this in the past with virtual items w/ the expire binding and creating and maintaining my own timers. However, I wanted an easier solution with little to no boilerplate.

[5iver]

I use some ESP12Es and check that they are still online with the network binding. Here is the metadata I use to get a notification when either are offline for more than an hour...

set_metadata("Status_Network_ESP12E_01", "area_triggers_and_actions", {
    "actions": {"notification_action": {"OFF": {"delay": 3600, "types": {"audio": True, "mobile": True}, "Priority": 0, "recurring": True}}}}, overwrite=True)
set_metadata("Status_Network_ESP12E_02", "area_triggers_and_actions", {
    "actions": {"notification_action": {"OFF": {"delay": 3600, "types": {"audio": True, "mobile": True}, "Priority": 0, "recurring": True}}}}, overwrite=True)

[rkoshak]

#233 provides a library that does all the Timer management book keeping for you. With one function call (check()) you can create a new timer or reschedule it if it already exits. You pass it up to two functions, one that optionally gets called if the Timer already existed when check was called and the other that gets called when the Timer expires.

So for the simple case where you want a Timer to check when foo is ON for five minutes

from core.rules import rule
from core.triggers import when
from community.timer_mgr import TimerMgr # probably should go in personal.timer_mgr until/if it gets merged

timers = TimerMgr()

def on_too_long(itemName):
    # what you do when the timer expires

@rule("timers for being on too long")
@when("Item foo changed to ON")
def changed_on(event):
    timers.check(event.itemName, 300000, lambda: on_too_long(event.itemName))

With the above code, if no Timer exists when check is called with the name event.itemName, one is created. If check is called before the timer triggers it gets cancelled. If check isn't called before the five minutes is up on_too_long gets called.

The library also has the option to reschedule the timer instead of cancelling it and you can pass an optional function to it to that gets called when check is called and the Timer already exists, useful for sending alerts that an Item is flapping for example.

[boc-tothefuture]

@rkoshak -
This does not cancel the timer if the item get changed to "OFF", Correct? That logic would still need to be implemented (potentially in the on_too_long method).

[5iver]

This issue is resolved and the requested feature will not be implemented. If you wish to discuss alternatives, please do so in the forum.