ha-zha-query-tools

Python tools to get zigbee device info from Home Assistant ZHA

You can run these python programs anywhere you can launch your browser to get to your Home Assistant server, not necessary to run on the server itself. Don't try to run them within HA, they will not run inside the HA python environment. These are python 3 programs. The python import files you will need to PIP3 install are found in the requirements.txt file in this repository:

 pip3 install -r requirements.txt

These programs retrieve the same data from the ZHA subsystem of HA as the ZHA network visualization page in HA, these programs just display the data in a different format and collect the data overtime into a SQLite database that you can query and see changes to your zigbee network and devices over time.

Configuration is done via zha_ws.yaml file, see example to configure.

zha_ws.py :

refactored SQLite database to reduce size, added configuration via YAML file zha_ws.yaml

This program tries to display ZHA devices based on their 'neighbor' relationship. It also records a SQLite database of the data returned by each web socket call to ZHA. Devices seems to stay in ZHA until you delete them, so this routine displays and records when a device goes 'OFFLINE' to the ZHA coordinator. It also records the length of time between when the coordinator sees a device and the devices RSSI and LQI values. Still trying to understand how to interpret this data. Run this program in tmux or other background way to have it collect data over time. And also so you can get another view of the current state of your ZHA zigbee network.

Example output:

Example SQLite database records:

Example raw JSON record:

ws02.py

This program just loops and dumps in JSON format the current ZHA devices database via web sockets. Two examples below of how to use jq linux JSON utility to display records:

This JSON parser will give you a good overview of the records returned by ZHA, run the program ws02.py and let it capture one query of the ZHA devices, stop the program and paste the raw into this tool, make sure to only capture the complete raw json output between the two outer {}'s :

example of raw json:

{"id": 1, "type": "result", "success": true, "result":............}

copy and paste into this web json viewer.
http://jsonviewer.stack.hu/

using jq to display subsets of the json returned:

current date and time in local time, ieee address, device name, lqi

./ws02.py | jq -c '.result[] | {date: (now|strflocaltime("%Y-%m-%d %H:%M:%S%Z")), ieee: .ieee, name: .name, lqi: .lqi}' {"date":"2021-01-07 22:31:16PST","ieee":"60:a4:23:ff:fe:dd:22:aa","name":"Silicon Labs EZSP","lqi":255} {"date":"2021-01-07 22:31:16PST","ieee":"f0:d1:b8:00:00:09:34:be","name":"LEDVANCE PLUG","lqi":196} {"date":"2021-01-07 22:31:16PST","ieee":"00:15:8d:00:02:ab:56:02","name":"LUMI lumi.weather","lqi":255}

UTC Timestamp

./ws02.py | jq -c '.result[] | {date: (now|strftime("%s")), ieee: .ieee, name: .name, lqi: .lqi}' {"date":"1610116915","ieee":"60:a4:23:ff:fe:a3:22:08","name":"Silicon Labs EZSP","lqi":255} {"date":"1610116915","ieee":"f0:d1:b8:00:00:56:4a:8e","name":"LEDVANCE PLUG","lqi":196} {"date":"1610116915","ieee":"00:15:8d:00:0A:11:f5:02","name":"LUMI lumi.weather","lqi":255}

display only EndDevices:

./ws02.py | jq -c '.result[] | select( .device_type == "EndDevice" ) | {date: (now|strflocaltime("%Y-%m-%d %H:%M:%S%Z")), ieee: .ieee, name: .name, device_type: .device_type, lqi: .lqi}' {"date":"2021-01-08 10:37:45PST","ieee":"00:15:8d:00:02:22:f5:72","name":"LUMI lumi.weather","device_type":"EndDevice","lqi":255} {"date":"2021-01-08 10:37:45PST","ieee":"28:6d:97:00:01:a2:0e:05","name":"Samjin multi","device_type":"EndDevice","lqi":192} {"date":"2021-01-08 10:37:45PST","ieee":"00:15:8d:00:02:a6:07:c5","name":"LUMI lumi.sensor_motion.aq2","device_type":"EndDevice","lqi":212}

display the children that are end devices

this needs work, getting duplicates and need to show parent as well

./ws02.py | jq -c '.result[] | select( .neighbors[].relationship == "Child" ) | select(.neighbors[].relationship == "Child") | select(.neighbors[].relationship == "Child") | .neighbors[] | select(.device_type == "EndDevice") | {ieee: .ieee, depth: .depth, lqi: .lqi}' {"ieee":"d0:52:a8:00:31:61:00:03","depth":"1","lqi":"183"} {"ieee":"00:15:8d:00:02:c2:05:c2","depth":"1","lqi":"255"} {"ieee":"00:15:8d:00:02:74:07:35","depth":"1","lqi":"213"} {"ieee":"00:15:8d:00:02:ca:f0:17","depth":"1","lqi":"173"} {"ieee":"00:15:8d:00:02:c6:dd:96","depth":"1","lqi":"194"} {"ieee":"d0:52:a8:00:31:61:00:03","depth":"1","lqi":"183"}

retired version....



## ws06.py :

stores more ZHA attributes in SQLite database, displaying more attributes on output screen, give the ability to create ws06.json file of the raw json returned from web sockets calls


## ws05.py :

has better error handling and is logging less data (not including raw json) to sqlite database so it does not grow so fast

## ws04.py :