I have set up a Minecraft Server and have set up a TTS message when players join (and leave using reverse statement) the server using the entity / state sensor.mc_y3_survival_players_online using the following template :
Since you’re really interested in the players list attribute, trigger off of that instead of having a trigger for each direction of change.
trigger:
- platform: state
entity_id: sensor.mc_y3_survival_players_online
attribute: players_list
not_to:
- none
- unknown
- unavailable
not_from:
- none
- unknown
- unavailable
condition: []
action:
- variables:
message: >
{% set to = trigger.to_state.attributes.players_list %}
{% set from = trigger.from_state.attributes.players_list %}
{% if from > to %}
{{ from | reject('in', to) | first }} has left the game.
{% else %}
{{ to | reject('in', from) | first }} has joined the game.
{% endif %}
- service: however you TTS
data:
message: "{{ message }}"
mode: queued
alias: "Automation : Notification : Minecraft Survival Player Update"
description: ""
trigger:
- platform: state
entity_id: sensor.mc_y3_survival_players_online
attribute: players_list
not_to:
- none
- unknown
- unavailable
not_from:
- none
- unknown
- unavailable
condition: []
action:
- variables:
message: >
{% set to = trigger.to_state.attributes.players_list %} {% set from =
trigger.from_state.attributes.players_list %} {% if from > to %}
{{ from | reject('in', to) | first }} has left the game.
{% else %}
{{ to | reject('in', from) | first }} has joined the game.
{% endif %}
- service: tts.google_translate_say
data:
entity_id: media_player.bedroom_g_h
message: "{{ message }}"
mode: queued
But when accessing the MC Server I got the following in the Logs :
2023-01-30 15:25:27.260 ERROR (MainThread) [homeassistant.helpers.event] Error while processing state change for sensor.mc_y3_survival_players_online
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/event.py", line 288, in _async_state_change_dispatcher
hass.async_run_hass_job(job, event)
File "/usr/src/homeassistant/homeassistant/core.py", line 577, in async_run_hass_job
hassjob.target(*args)
File "/usr/src/homeassistant/homeassistant/components/homeassistant/triggers/state.py", line 161, in state_automation_listener
or not match_to_state(new_value)
File "/usr/src/homeassistant/homeassistant/helpers/event.py", line 1590, in <lambda>
return lambda state: invert is not (state in parameter_set)
TypeError: unhashable type: 'list'
2023-01-30 15:25:27.334 ERROR (MainThread) [homeassistant.helpers.event] Error while processing state change for sensor.mc_y3_survival_players_online
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/event.py", line 288, in _async_state_change_dispatcher
hass.async_run_hass_job(job, event)
File "/usr/src/homeassistant/homeassistant/core.py", line 577, in async_run_hass_job
hassjob.target(*args)
File "/usr/src/homeassistant/homeassistant/components/homeassistant/triggers/state.py", line 161, in state_automation_listener
or not match_to_state(new_value)
File "/usr/src/homeassistant/homeassistant/helpers/event.py", line 1590, in <lambda>
return lambda state: invert is not (state in parameter_set)
TypeError: unhashable type: 'list'
I honestly wish I could now add some insight into this, but this is “way above my pay grade and job title” as a “Junior Associate Home Assistant Tinkerer”…
I have however I have figured out that the player_list is formatted as follows (the . (dot) value is a Geyser / Floodgate user) :
Ok, apparently me trying to be ‘smart’ was dumb… the not_to/from breaks it. I added a pair of count filters to the if statement. In testing, it helped make sure both the “leave” and “join” events were announced properly. I also added replace() functions to remove the . you mentioned.
alias: "Automation : Notification : Minecraft Survival Player Update"
description: ""
trigger:
- platform: state
entity_id: sensor.mc_y3_survival_players_online
attribute: players_list
condition: []
action:
- variables:
message: >
{% set to = trigger.to_state.attributes.players_list %} {% set from =
trigger.from_state.attributes.players_list %}
{% if from | count > to | count %}
{{ from | reject('in', to) | first | replace('.', ' ') }} has left the game.
{% else %}
{{ to | reject('in', from) | first | replace('.', ' ') }} has joined the game.
{% endif %}
- service: tts.google_translate_say
data:
entity_id: media_player.bedroom_g_h
message: "{{ message }}"
mode: queued
I have confirmed that the above works… at least when using Alexa as the end point. I don’t use the Google say TTS services, so I can’t really help if that’s the source of any further issues.
