Clear miflora sensor state on exception

[ferbar]

Breaking Change:

Values for a unresponsive device aren't reported any more as if they were valid, so it's possible to send an alert.

Description:

Report avaiable() = False if querying the device fails. The state is then set to unavailable, so it can be tracked if a miflora device isn't responding any more.

Pull request with documentation for home-assistant.io (if applicable): home-assistant/home-assistant.io#<home-assistant.io PR number goes here>

Example entry for automations.yaml:

- id: plant_unreachable
  alias: plant unreachable
  trigger:
  - entity_id: sensor.plant_moisture
    for: 01:00:00
    platform: state
    to: unavailable
  action:
...

Checklist:

• The code change is tested and works locally.
• Local tests pass with tox. Your PR cannot be merged unless tests pass
• There is no commented out code in this PR.
• I have followed the development checklist

If user exposed functionality or configuration variables are added/changed:

• Documentation added/updated in home-assistant.io

If the code communicates with devices, web services, or third-party tools:

• The manifest file has all fields filled out correctly. Update and include derived files by running python3 -m script.hassfest.
• New or updated dependencies have been added to requirements_all.txt by running python3 -m script.gen_requirements_all.
• Untested files have been added to .coveragerc.

If the code does not interact with devices:

• Tests have been added to verify that the new code works.

---

Updates:
changed unknown to unavailable in automations.yaml

[homeassistant]

Hi @ferbar,

It seems you haven't yet signed a CLA. Please do so here.

Once you do that we will be able to review and accept this pull request.

Thanks!

[probot-home-assistant]

Hey there @Danielhiversen, @ChristianKuehnel, mind taking a look at this pull request as its been labeled with a integration (miflora) you are listed as a codeowner for? Thanks!

[balloob]

Instead of self._state = None, we should return False from the available property.

[balloob]

You can combine the except clause.

Suggested change

	except OSError as ioerr:
	except (OSError, BluetoothBackendException) as ioerr:

[balloob]

Use available property.

[ferbar]

I think "available()" and "except merge" should be changed in mitemp_bt, too. (?)

• tested & works

[MartinHjelmare]

Good! Please update the PR description according to latest changes. Then we can merge.

[ChristianKuehnel]

@ferbar: Great idea!

We have quite a few users who's sensors are at the edge of what's feasible with regards to the range of the Bluetooth devices. If you fail on the first exception, this will cause a lot of errors, as the Bluetooth communication is quite unreliable. So they would get tons of errors this way.

I would recommend something like: "fail if no new data for x hours", where x can be configured.

Rationale:
I do not care about every missed measurement and the data changes quite slowly. It's good enough if the moisture/fertilizer values are "new enough". For me, once a day is good enough. But that might differ from user to user.

[ferbar]

@ChristianKuehnel: I'm with you. I was playing around with last_updated and last_changed before. However last_updated is changed even if update() was unsuccessful. I think that a timeout for values to get unavailable should be implemented in Entity or a derived class since this would affect many types of sensors and would produce a lot of copy&paste otherwise.

• I didn't find out why last_updated is changed (I was searching for a long time)
• I tried to override last_update with device_state_attributes. Worked, but I think this is really ugly 😄 IMHO the current solution is better.
• I don't like to track an additional last_update timestamp.

Do you have any idea how to prevent last_updated get updated in an error case?

[ChristianKuehnel]

No I don't. What about creating your own stopwatch: remember the time of the last successful update. If it's older than your threshold: null the values. Am So., 1. Dez. 2019 um 18:43 Uhr schrieb Christian Ferbar < notifications@github.com>:

…

@ChristianKuehnel <https://github.com/ChristianKuehnel>: I'm with you. I was playing around with last_updated and last_changed before. However last_updated is changed even if update() was unsuccessful. I think that a timeout for values to get unavailable should be implemented in Entity or a derived class since this would affect many types of sensors and would produce a lot of copy&paste otherwise. - I didn't find out why last_updated is changed (I was searching for a long time) - I tried to override last_update with device_state_attributes. Worked, but I think this is really ugly :-D IMHO the current solution is better. - I don't like to track an additional last_update timestamp. Do you have any idea how to prevent last_updated get updated in an error case? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#29276>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEYJMCCPBH2P6JH3MSTWDBDQWPZS3ANCNFSM4JTIV64A> .