Submit it as a Feature request?
If you updated the code and you verified it is working submit a Pull Request in Github so the project can get the updated code and everyone can use it instead of users needing to edit the files the way it was described. Anyone can submit a PR.
My comment was actually directed to @Quatuor who provided the code changes.
Got it.
Until it gets merged is their any way to do this with a value_template
in the configuration.yaml or entity_registry.yaml for now?
1 Like
you can do a custom component with the folder and changes mentioned above.
1 Like
Did you get dimming to work for the Osram Lightify switch in your code above?
Nope. After a closer look I found some issues with the original PR. So I was working on a few fixes and some improvements (at least I think so 
added state restore upon hass restart). Can you give it a try and let me know? hereās a new homeassistant/components/binary_sensor/zha.py which I was working on. You still are going to need the __init__.py and const.py from my previous post.
"""
Binary sensors on Zigbee Home Automation networks.
For more details on this platform, please refer to the documentation
at https://home-assistant.io/components/binary_sensor.zha/
"""
import logging
import time
from enum import IntEnum
from homeassistant.components.binary_sensor import DOMAIN, BinarySensorDevice
from homeassistant.components import zha
from homeassistant.const import STATE_OFF
from homeassistant.helpers.event import async_track_time_change
from homeassistant.helpers.restore_state import async_get_last_state
_LOGGER = logging.getLogger(__name__)
DEPENDENCIES = ['zha']
# ZigBee Cluster Library Zone Type to Home Assistant device class
CLASS_MAPPING = {
0x000d: 'motion',
0x0015: 'opening',
0x0028: 'smoke',
0x002a: 'moisture',
0x002b: 'gas',
0x002d: 'vibration',
}
MOVING_STEP = 15
class OnOffCommands(IntEnum):
OFF = 0x00
ON = 0x01
TOGGLE = 0x02
OFF_WITH_EFFECT = 0x40
ON_WITH_RECALL_GLOBAL_SCENE = 0x41
ON_WITH_TIMED_OFF = 0x42
class LevelCommands(IntEnum):
MOVE_TO_LEVEL = 0x00
MOVE = 0x01
STEP = 0x02
STOP = 0x03
MOVE_TO_LEVEL_WITH_ON_OFF = 0x04
MOVE_WITH_ON_OFF = 0x05
STEP_WITH_ON_OFF = 0x06
STOP_WITH_ON_OFF = 0x07
async def async_setup_platform(hass, config, async_add_devices,
discovery_info=None):
"""Set up the Zigbee Home Automation binary sensors."""
discovery_info = zha.get_discovery_info(hass, discovery_info)
if discovery_info is None:
return
from zigpy.zcl.clusters.general import OnOff
from zigpy.zcl.clusters.security import IasZone
if IasZone.cluster_id in discovery_info['in_clusters']:
await _async_setup_iaszone(hass, config, async_add_devices,
discovery_info)
elif OnOff.cluster_id in discovery_info['out_clusters']:
await _async_setup_remote(hass, config, async_add_devices,
discovery_info)
async def _async_setup_iaszone(hass, config, async_add_devices,
discovery_info=None):
device_class = None
from zigpy.zcl.clusters.security import IasZone
cluster = discovery_info['in_clusters'][IasZone.cluster_id]
if discovery_info['new_join']:
await cluster.bind()
ieee = cluster.endpoint.device.application.ieee
await cluster.write_attributes({'cie_addr': ieee})
try:
zone_type = await cluster['zone_type']
device_class = CLASS_MAPPING.get(zone_type, None)
except Exception: # pylint: disable=broad-except
# If we fail to read from the device, use a non-specific class
pass
sensor = BinarySensor(device_class, **discovery_info)
async_add_devices([sensor], update_before_add=True)
async def _async_setup_remote(hass, config, async_add_devices, discovery_info):
async def safe(coro):
"""Run coro, catching ZigBee delivery errors, and ignoring them."""
import zigpy.exceptions
try:
await coro
except zigpy.exceptions.DeliveryError as exc:
_LOGGER.info("Ignoring error during setup: %s", exc)
if discovery_info['new_join']:
from zigpy.zcl.clusters.general import OnOff, LevelControl
out_clusters = discovery_info['out_clusters']
if OnOff.cluster_id in out_clusters:
cluster = out_clusters[OnOff.cluster_id]
await safe(cluster.bind())
await safe(cluster.configure_reporting(0, 0, 600, 1))
if LevelControl.cluster_id in out_clusters:
cluster = out_clusters[LevelControl.cluster_id]
await safe(cluster.bind())
await safe(cluster.configure_reporting(0, 1, 600, 1))
sensor = Switch(**discovery_info)
async_add_devices([sensor])
class BinarySensor(zha.Entity, BinarySensorDevice):
"""The ZHA Binary Sensor."""
_domain = DOMAIN
def __init__(self, device_class, **kwargs):
"""Initialize the ZHA binary sensor."""
super().__init__(**kwargs)
self._device_class = device_class
from zigpy.zcl.clusters.security import IasZone
self._ias_zone_cluster = self._in_clusters[IasZone.cluster_id]
@property
def should_poll(self) -> bool:
"""Let zha handle polling."""
return False
@property
def is_on(self) -> bool:
"""Return True if entity is on."""
if self._state == 'unknown':
return False
return bool(self._state)
@property
def device_class(self):
"""Return the class of this device, from component DEVICE_CLASSES."""
return self._device_class
def cluster_command(self, tsn, command_id, args):
"""Handle commands received to this cluster."""
if command_id == 0:
self._state = args[0] & 3
_LOGGER.debug("Updated alarm state: %s", self._state)
self.async_schedule_update_ha_state()
elif command_id == 1:
_LOGGER.debug("Enroll requested")
res = self._ias_zone_cluster.enroll_response(0, 0)
self.hass.async_add_job(res)
async def async_update(self):
"""Retrieve latest state."""
from bellows.types.basic import uint16_t
result = await zha.safe_read(self._endpoint.ias_zone,
['zone_status'])
state = result.get('zone_status', self._state)
if isinstance(state, (int, uint16_t)):
self._state = result.get('zone_status', self._state) & 3
class Switch(zha.Entity, BinarySensorDevice):
"""ZHA switch/remote controller/button."""
_domain = DOMAIN
class OnOffListener:
"""Listener for the OnOff ZigBee cluster."""
def __init__(self, entity):
"""Initialize OnOffListener."""
self._entity = entity
self._state = True
def cluster_command(self, tsn, command_id, args):
"""Handle commands received to this cluster."""
cmds = OnOffCommands
_LOGGER.debug('{}: rx {} command with {} args on OnOff cluster'.
format(self._entity.entity_id, command_id, args))
if command_id in (cmds.OFF, cmds.OFF_WITH_EFFECT):
self.set_state(False)
elif command_id in (cmds.ON, cmds.ON_WITH_RECALL_GLOBAL_SCENE,
cmds.ON_WITH_TIMED_OFF):
self.set_state(True)
elif command_id == cmds.TOGGLE:
self.toggle()
def attribute_updated(self, attrid, value):
"""Handle attribute updates on this cluster."""
_LOGGER.debug('{}: reports {}/{} attr_id/value on OnOff cluster'.
format(self._entity.entity_id, attrid, value))
if attrid == 0:
self._state = value
self._entity.schedule_update_ha_state()
@property
def is_on(self) -> bool:
"""state of the switch"""
return self._state
def set_state(self, state):
"""Set the state."""
self._state = state
if self._state and not bool(self._entity.level.level):
self._entity.level.set_level(1, True)
self._entity.schedule_update_ha_state()
def toggle(self):
"""Toggle the state"""
self._state = not self._state
if self._state and not bool(self._entity.level.level):
self._entity.level.set_level(1, True)
self._entity.schedule_update_ha_state()
class LevelListener:
"""Listener for the LevelControl ZigBee cluster."""
class LvlChange:
"""ammount of Level change"""
def __init__(self, direction=0, rate=0, delta=MOVING_STEP,
start_moving=True):
self._direction = direction
if direction == 0x00:
self._delta = delta
else:
self._delta = -delta
self.still_moving = start_moving
if start_moving:
self.when_started = time.time()
self._rate = rate
@property
def delta(self):
"""amount of level change"""
return self._delta
@property
def delta_since_start(self):
"""return amount of change since start of level move"""
dt = time.time() - self.when_started
return int(dt * self.delta)
def stop(self):
"""Stop moving and returns delta since the start"""
self.still_moving = False
return self.delta_since_start
def __radd__(self, other):
return other + self._delta
def __rmul__(self, other):
return other * self._delta
def __init__(self, entity):
"""Initialize LevelListener."""
self._entity = entity
self._level = 255
self._orig_level = self._level
self.delta = self.LvlChange(start_moving=False)
def cluster_command(self, tsn, command_id, args):
"""Handle commands received to this cluster."""
_LOGGER.debug('{}: rx {} command with {} args on Level cluster'.
format(self._entity.entity_id, command_id, args))
if command_id == LevelCommands.MOVE_TO_LEVEL: # move_to_level
self.set_level(args[0])
elif command_id == LevelCommands.MOVE: # move
# We should dim slowly -- for now, just step once
self.delta = self.LvlChange(*args)
self._orig_level = self._level
self.move_level(self.delta)
elif command_id == LevelCommands.STEP: # step
# Step (technically shouldn't change on/off)
delta = self.LvlChange(*args)
self.move_level(delta)
elif command_id in (LevelCommands.STOP,
LevelCommands.STOP_WITH_ON_OFF):
stop_delta = self.delta.stop()
self.set_level(self._orig_level + stop_delta)
elif command_id == LevelCommands.MOVE_TO_LEVEL_WITH_ON_OFF:
# move_to_level_with_on_off
self.set_level(args[0], True)
elif command_id == LevelCommands.MOVE_WITH_ON_OFF: # move_with_on_off
# We should dim slowly -- for now, just step once
self.delta = self.LvlChange(*args)
self._orig_level = self._level
self.move_level(self.delta, True)
elif command_id == LevelCommands.STEP_WITH_ON_OFF: # step_with_on_off
delta = self.LvlChange(*args)
self.move_level(delta, True)
def attribute_update(self, attrid, value):
"""Handle attribute updates on this cluster."""
_LOGGER.debug('{}: reports {}/{} attr_id/value on level cluster'.
format(self._entity.entity_id, attrid, value))
if attrid == 0:
self.set_level(value)
@property
def level(self):
"""current level"""
return self._level
def move_level(self, change, onoff=False):
"""Increment the level, setting state if appropriate."""
self._level = min(255, max(0, self._level + change))
if onoff:
self._entity.onoff.set_state(bool(self._level))
self._entity.schedule_update_ha_state()
def set_level(self, level, onoff=False):
"""Set the level, setting state if appropriate."""
self._level = min(255, max(0, level))
if onoff:
self._entity.onoff.set_state(bool(self._level))
self._entity.schedule_update_ha_state()
async def async_time_interval(self, tracked_time):
"""if button is held, estimate level change"""
if not self.delta.still_moving:
return
delta = self.delta.delta_since_start
self.set_level(self._level + delta)
async def async_added_to_hass(self):
"""Run when about to be added to Hass"""
# Check if we have an old state
old_state = await async_get_last_state(self.hass, self.entity_id)
if old_state is not None:
_LOGGER.debug('{}: restoring old state {}'.format(self.entity_id,
old_state))
if old_state.attributes.get('level') is not None:
self.level.set_level(old_state.attributes['level'])
if old_state.state == STATE_OFF:
self.onoff.set_state(False)
async_track_time_change(self.hass, self.level.async_time_interval)
def __init__(self, **kwargs):
"""Initialize Switch."""
self.onoff = self.OnOffListener(self)
self.level = self.LevelListener(self)
from zigpy.zcl.clusters import general
self._out_listeners = {
general.OnOff.cluster_id: self.onoff,
general.LevelControl.cluster_id: self.level,
}
super().__init__(**kwargs)
@property
def is_on(self) -> bool:
"""Return true if the binary sensor is on."""
return self.onoff.is_on and bool(self.level.level)
@property
def device_state_attributes(self):
"""Return the device state attributes."""
attrs = {}
if self.is_on:
attrs['level'] = self.level.level
return attrs
with the above binary_sensor Iām using the following automation to control the lights. The idea is: triggered any time the state changes (since we want to catch level change) and if the state of the binary_switch is on then we turn on the lights to the level corresponding to binary_switch attributes[ālevelā]
- id: floor_light_switch_on
alias: Turn on floor lamp when switched on
trigger:
- platform: state
entity_id: binary_sensor.dimming_switch
condition:
- condition: state
entity_id: binary_sensor.dimming_switch
state: 'on'
action:
- service: light.turn_on
data_template:
entity_id: light.floor_lamp
brightness: "{{ trigger.to_state.attributes.level }}"
- id: light_switch_off
alias: Turn off floor lamp when switched off
trigger:
- platform: state
entity_id: binary_sensor.dimming_switch
to: 'off'
action:
- service: light.turn_off
entity_id:
- light.floor_lamp
** edit: fixed set level to stay within 0ā¦255
*** edit: added automation to link binary_switch level change to light control
**** edit: update to turn switch off when level is 0
PS: Iām just starting with python so I understand some things iām doing could be very stupid. Any constructive dialog is welcome
It works well, however dimming is a little slow. One suggestion would be to toggle the state of the switch to off when the level gets to 0.
I am running Hassio on a raspberry pi and ssh into edit config files
can you tell me where to find the components/sensor/zha.py file
you are suggesting to edit.
Started having weird behavior. I have 3 iris open close sensors and 3 iris motion sensors. Iāve been moving over my devices this week, and finally decided to move everything else over. So before I started 2 motion and one open close were added. I have not been able to get the other ones to add at all. Itās like zha permit is not even working.
I even went and deleted the zigbee.db, I was able to add one of the previous devices, but still not the other devices. I then removed the db one more time, and now nothing will add.
Iāve spent all week on this stuff, and not the last few devices I expected to be easy are causing my greif. Any suggestions, or places to look for what may be going on? Thanks!
Yep, you are right. Doesnāt turn off when reaches 0. This is broken.
The dimming is linear and it does feel slow. You could change the āMOVING_STEPā which is currently at 15, but perhaps we should implement āexponentialā progression instead of linear one. Stay tunedā¦
I donāt have experience with docker based installs. If I understand it correctly you really canāt āhackā those setups, which I was doing with my venv based setups.
I need to rewrite these as a custom_component. I started to explore that option, but I wish i had more time to spare I think as a short term this could be an option, but it really makes sense to get this into Yoda-xās zha_new custom component.
I just added the husbzb-1 to my home-assistant setup. The first device attempt is the old smartthings motion (original). It is able to pair/join, but I donāt see it as an available entity in hass. I havenāt had a chance to my other devices as my venture from smartthings to just purely hass.
Blockquote
2018-04-29 09:01:01 WARNING (MainThread) [bellows.zigbee.application] Unexpected response TSN=91 command=256 args=bā\xc5ā
2018-04-29 09:01:01 WARNING (MainThread) [zigpy.endpoint] [0xa291:2] Message on unknown cluster 0xfc04
2018-04-29 09:01:01 INFO (MainThread) [homeassistant.core] Bus:Handling <Event system_log_event[L]: timestamp=1525017661.8279727, level=WARNING, message=Unexpected response TSN=91 command=256 args=bā\xc5ā, exception=, source=core.py>
2018-04-29 09:01:01 INFO (MainThread) [homeassistant.core] Bus:Handling <Event system_log_event[L]: timestamp=1525017661.8951173, level=WARNING, message=[0xa291:2] Message on unknown cluster 0xfc04, exception=, source=core.py>
Blockquote
Device:
NWK: 0xa291
IEEE: d0:52:a8:00:2e:75:00:04
Endpoints:
1: profile=0x104, device_type=314
Input Clusters:
Basic (0)
Output Clusters:
Ota (25)
2: profile=0xfc01, device_type=314
Still having issues adding new zigbee devices. Turned off zwave and was able to add back a previously added iris open close sensor but unable to add anything new. Any recommendations, Iāve delete the db a couple times, but that did not help.
I do have a lot of errors like this in my log
Message on unknown device
Trying to pair a door sensor I get this:
2018-05-04 12:59:32 WARNING (MainThread) [bellows.zigbee.application] Unexpected response TSN=19 command=1 args=[[]] 2018-05-04 12:59:32 WARNING (MainThread) [zigpy.endpoint] [0x7f1c:1] Message on unknown cluster 0x0500 2018-05-04 12:59:42 WARNING (MainThread) [bellows.zigbee.application] Unexpected response TSN=20 command=1 args=[[]]
2018-05-04 12:59:42 WARNING (MainThread) [zigpy.endpoint] [0x0aba:1] Message on unknown cluster 0x0500 2018-05-04 12:59:53 WARNING (MainThread) [bellows.zigbee.application] Unexpected response TSN=21 command=1 args=[[]]
2018-05-04 12:59:53 WARNING (MainThread) [zigpy.endpoint] [0x3140:1] Message on unknown cluster 0x0500 2018-05-04 13:00:03 WARNING (MainThread) [bellows.zigbee.application] Unexpected response TSN=22 command=1 args=[[]]
2018-05-04 13:00:03 WARNING (MainThread) [zigpy.endpoint] [0xb6cd:1] Message on unknown cluster 0x0500 2018-05-04 13:00:14 WARNING (MainThread) [bellows.zigbee.application] Unexpected response TSN=23 command=1 args=[[]]
2018-05-04 13:00:14 WARNING (MainThread) [zigpy.endpoint] [0x2e99:1] Message on unknown cluster 0x0500 2018-05-04 13:00:24 WARNING (MainThread) [bellows.zigbee.application] Unexpected response TSN=24 command=1 args=[[]]
2018-05-04 13:00:24 WARNING (MainThread) [zigpy.endpoint] [0xa70a:1] Message on unknown cluster 0x0500
What does that mean ?
Updated code for SmartThings arrival sensor including 2015 and 2017 models. Its a hack but works for me!
Anyone ever get the SmartThings water sensor to work with HA. Iāve been able to get an older ST motion sensor and ST Open close sensor to work, but no luck with the water/leak detector.
These sensors were quite stable for me on ST, and they are currently on sale for $25, which is a good value, if they could work directly with HA
Fairly certain that when you see messages along the lines of Message on unknown device 0x1af6 you need to perform a reset on the device in question, then join the device with a zha.permit and whatever the device needs. Might help @Argoson also?
I am getting these TSN messages constantly as well for my lightify dimmer switch I paired. I also notice I need to reboot at least daily or that switch stops functioning.
Any ideas?
Unexpected response TSN=79 command=1 args=[[<ReadAttributeRecord attrid=0 status=134>]] 10:27 AM deps/lib/python3.6/site-packages/bellows/zigbee/application.py (WARNING) Unexpected response TSN=78 command=1 args=[[<ReadAttributeRecord attrid=0 status=134>]] 10:27 AM deps/lib/python3.6/site-packages/bellows/zigbee/application.py (WARNING) Unexpected response TSN=77 command=1 args=[[<ReadAttributeRecord attrid=0 status=134>]] 10:26 AM deps/lib/python3.6/site-packages/bellows/zigbee/application.py (WARNING) Unexpected response TSN=71 command=1 args=[[<ReadAttributeRecord attrid=0 status=134>]] 10:23 AM deps/lib/python3.6/site-packages/bellows/zigbee/application.py (WARNING)
I have 4 iris motion sensors (centralite_3326l) and three iris open/close sensors (.centralite_3320l).
I also have one smartthings open close and one smartthings motion sensor.
Overall my experience with the Linear HUSBZB-1 stick has been barely accecptable with these.
When they work, they work great. Fast reporting. I also get temp readings from all of them.
My issue is that they often just stop reporting. usually a quick pull of the battery gets them working again, but Iād say I have to do that at least one time a day with at least one sensor.
Whatās mostly annoying is that they often get āStuckā on.
Where this becomes a big issue is that I have a group with several motion sensors. I have a few automations that rely on the state going from OFF to ON or visa versa. If they are stuck the state of the group never changes, and the automations never trigger.
I bought an Iris zigbee plug to act as a repeater and it has helped a little bit. but not much. I dont have a giant house, these sensors are withing 50 feet of the USB stick, and they are all within probably 15-20 feet of each other in general.
The smartthings sensors seem to be rock solid.
I just really liked the size of the iris sensors. I have not seen a zwave motion sensor anywhere as small as these.
1 Like
I have 14 iris sensors and none of them have this issue. I also have several zigbee plugs around the house too. The range on the USB stick is not that great. None of my sensors get stuck. I have one iris motion sensors that died in 3 days of use so I may disconnect that one. Have you tried to replace the batteries to rule out low batteries?