The garbage_collection
component is a Home Assistant helper that creates a custom sensor for monitoring a regular garbage collection schedule. The sensor can be configured for a number of different patterns:
weekly
schedule (including multiple collection days, e.g. on Tuesday and Thursday)every-n-weeks
repeats everyperiod
of weeks, starting from the week numberfirst_week
. It uses the week number - therefore, it restarts each year, as the weeks start from number 1 each year.- bi-weekly in
even-weeks
orodd-weeks
(technically, it is the same as every 2 weeks with 1st or 2ndfirst_week
) every-n-days
(repeats regularly from the given first date). If n is a multiply of 7, it works similar toevery-n-weeks
, with the difference that it does not use the week numbers (that restart each year) but continues infinitely from the initial date.monthly
schedule (nth weekday each month), or a specific weekday of each nth week. Using theperiod
it could also be every 2nd, 3rd etc month.annually
(e.g. birthdays). This is once per year. Using include dates, you can add additional dates manually.blank
does not automatically schedule any collections - to be used in cases where you want to make completely own schedule withmanual_update
.
You can also configure seasonal calendars (e.g. for bio-waste collection), by configuring the first and last month.
And you can group
entities, which will merge multiple schedules into one sensor.
These are some examples using this sensor. The Lovelace config examples are included below.
- Download the latest release.
- Unpack the release and copy the
custom_components/garbage_collection
directory into thecustom_components
directory of your Home Assistant installation. - Restart Home Assistant.
- Configure the
garbage_collection
helper.
- Ensure that HACS is installed.
- Search for and install the "Garbage Collection" integration.
- Restart Home Assistant.
- Configure the
garbage_collection
helper.
Go to Settings
/Devices & Services
/Helpers
, click on the + CREATE HELPER
button, select Garbage Collection
and configure the helper.
If you would like to add more than one collection schedule, click on the + CREATE HELPER
button again and add another Garbage Collection
helper instance.
The configuration hapend in 2 steps. In the first step, you select the frequency
and common parameters. In the second step you configure additional parameters depending on the selected frequency.
The configuration via configuration.yaml
has been deprecated. If you have previously configured the integration there, it will be imported to ConfigFlow, and you should remove it.
Parameter | Required | Description |
---|---|---|
Friendly name |
Yes | Sensor friendly name |
Frequency |
Yes | "weekly" , "even-weeks" , "odd-weeks" , "every-n-weeks" , "every-n-days" , "monthly" , "annual" , "group" or "blank" |
Icon |
No | Default icon Default: mdi:trash-can |
Icon today |
No | Icon if the collection is today Default: mdi:delete-restore |
Icon tomorrow |
No | Icon if the collection is tomorrow Default: mdi:delete-circle |
Expire After |
No | Time in format format HH:MM . If the collection is due today, start looking for the next occurence after this time (i.e. if the weekly collection is in the morning, change the state from 'today' to next week in the afternoon) |
Verbose state |
No | The sensor state will show collection date and remaining days, instead of number Default: False |
Hidde in calendar |
No | Hide in calendar (useful for sensors that are used in groups) Default: False |
Manual update |
No | (Advanced). Do not automatically update the status. Status is updated manualy by calling the service garbage_collection.update_state from an automation triggered by event garbage_collection_loaded , that could manually add or remove collection dates, and manually trigger the state update at the end. See the example.Default: False |
Verbose format |
No | (relevant when verbose state is True ). Verbose status formatting string. Can use placeholders {date} and {days} to show the date of next collection and remaining days. Default: 'on {date}, in {days} days' When the collection is today or tomorrow, it will show Today or Tomorrow (currently in English, French, Czech and Italian). |
Date format |
No | In the verbose format , you can configure the format of date (using strftime format) Default: '%d-%b-%Y' |
Parameter | Required | Description |
---|---|---|
First month |
No | Month three letter abbreviation, e.g. "jan" , "feb" ...Default: "jan" |
Last month |
No | Month three letter abbreviation. Default: "dec" |
Parameter | Required | Description |
---|---|---|
Collection days |
Yes | Day three letter abbreviation, list of "mon" , "tue" , "wed" , "thu" , "fri" , "sat" , "sun" . |
Parameter | Required | Description |
---|---|---|
Period |
No | Collection every "period" weeks (integer 1-53)Default: 1 |
First week |
No | First collection on the "first week" week (integer 1-53)Default: 1 (The week number is using ISO-8601 numeric representation of the week) Note: This parameter cannot be used to set the beginning of the collection period (use the first month parameter for that). The purpose of first week is to simply 'offset' the week number, so the collection every ;'n' weeks does not always trigger on week numbers that are multiplication of 'n'. Technically, the value of this parameter shall be less than period , otherwise it will give weird results. Also note, that the week numbers restart each year. Use every-n-days frequency if you need a consistent period across the year ends. |
Parameter | Required | Description |
---|---|---|
First date |
Yes | Repeats every n days from this first date (date in the international ISO format 'yyyy-mm-dd' ). |
Period |
No | Collection every "period" days (warning - in this configuration, it is days, not weeks!)Default: 1 (daily, which makes no sense I suppose) |
The monthly schedule has two flavors: it can trigger either on the nth occurrence of the weekday in a month, or on the weekday in the nth week of each month.
Parameter | Required | Description |
---|---|---|
Order of weekday |
Yes | List of week numbers of collection day each month. E.g., if collection_day is "sat" , 1 will mean 1st Saturday each month (integer 1-5) |
Order of week, instead of weekday order |
No | CONFIGURE THIS ONE ONLY IF YOU ARE SURE YOU NEED IT. This will alter the behaviour of order of weekday , so that instead of nth weekday of each month, take the weekday of the nth week of each month.So if the month starts on Friday, the Wednesday of the 1st week would actually be last Wednesday of the previous month and the Wednesday of 2nd week will be the 1st Wednesday of the month. So if you have just randomy clicked on the option, it might appear as if it calculates a wrong date! Yes, this is confusing, but there are apparently some use case for this. |
Period |
No | If period is not defined (or 1), the schedule will repeat monthly. If period is 2, it will be every 2nd month. If period is 3, it will be once per quarter, and so on.The first month parameter will then define the starting month. So if the first month is jan (or not defined), and period is 2, the collection will be in odd months (jan , mar , may , jul , sep and nov ). If first month is feb , it will be in even months. (integer 1-12)Default: 1 |
Parameter | Required | Description |
---|---|---|
Date |
Yes | The date of collection, in format 'mm/dd' (e.g. '11/24' for November 24 each year) |
Parameter | Required | Description |
---|---|---|
List of entities |
Yes | A list of entity_id s to merge |
- To use the blueprints, you need to set the
garbage_collection
entity forManual update
, that will fire thegarbage_collection_loaded
event on each sensor update and trigger the automation blueprint. - Install/Import blueprint
- From the blueprint, create and configure the automation
There are a couple of blueprints, automatically moving the collection falling on a public holiday. Or if there was a public holiday in the week before the scheduled collection.
The Public Holidays blueprints use a separate custom helper Holidays, available through HACS, that you can configure for different countries.
A list of fixed dates to include and exclude from the calculated schedule.
Include and Exclude | |
Include | |
Exclude |
The offset blueprint will move the calculated collections by a number of days. This can be used, for example, to schedule collection for last Saturday each month - just set the collection to the first Saturday each month and offset it by -7 days.
This blueprint requires a command_line
sensor reading content of a txt file, containig a set of dates, one per line.
command_line sensor example
The Home Assistant command line sensor has a 255 character limitation. For many people this causes an issue where the sensor only imports part of the list. To overcome this limitation, @stu1811 came up with this brilliant solution.
Create a script that shows next 20 dates from the current day forward.
#!/bin/bash
new_dates=$(for x in $(cat /share/import_dates.txt); do
if [[ $(date -d $x +"%y%m%d") -ge $(date +"%y%m%d") ]]; then
echo $x
fi
done)
echo "$new_dates" | head -n20
Then, create the command line sensor taking the output of this script.
sensor:
- platform: command_line
name: Import dates
command: "sh /share/import_dates.sh"
This will create a schedule on a fixed date each month. For example on the 3rd each month. The helper does not allow it, as it is generally designed around paterns evolving around weekly schedules (since garbage collection typically happens on a set day in a week, rather than set day in a month). But few of you wanted that, so here you go.
The state can be one of
Value | Meaning |
---|---|
0 |
Collection is today |
1 |
Collection is tomorrow |
2 |
Collection is later |
If the verbose_state
parameter is set, it will show the date, and the remaining days. For example: "Today" or "Tomorrow" or "on 10-Sep-2019, in 2 days" (configurable)
Attribute | Description |
---|---|
next_date |
The date of next collection |
days |
Days till the next collection |
last_collection |
The date and time of the last collection |
If the collection is scheduled for today, mark it completed and look for the next collection.
It will set the last_collection
attribute to the current date and time.
Attribute | Description |
---|---|
entity_id |
The garbage collection entity id (e.g. sensor.general_waste ) |
last_collection |
(optional) Set the last collection date to this value. This can be used to re-set the next collection calculation, if the last collection date was set in error. If omitted, it will set the last collection to the current date & time. |
There are standard blueprints provided to handle manual updates - to move collection on public holidays or offset the collection.
If these blueprints do not work for you, you can create your own custom rules to handle any scenario. If you do so, please share the blueprints with the others by posting them to the blueprints directory - someone else might find them useful. Thanks! To help you with creating custom automations, see the following examples:
The following services are used within automations, triggered by the garbage_collection_loaded event. Don't use them anywhere else, it won't work. For the examples of their use, see the examples
flowchart TD
A[HA updates entity] -->B[triggered garbage_collection_loaded event]
B --> C[calling services]
C --> D[garbage_collection.add_date] & E[garbage_collection.remove_date] & F[garbage_collection.offset_date] --> C
D & E & F --> G[manual update finished -> garbage_collection.update_state]
Add a date to the list of dates calculated automatically. To add multiple dates, call this service multiple-times with different dates.
Note that this date will be removed on the next sensor update, when the data is re-calculated and loaded. This is why, this service should be called from the automation triggered by the event garbage_collection_loaded
. This event is called each time the sensor is updated. And at the end of this automation, you need to call the garbage_collection.update_state
service to update the sensor state based on automatically collected dates, and the dates added, removed, or offset by the automation.
Attribute | Description |
---|---|
entity_id |
The garbage collection entity id (e.g. sensor.general_waste ) |
date |
The date to be added, in ISO format ('yyyy-mm-dd' ). Make sure to enter the date in quotes! |
Remove a date to the list of dates calculated automatically. To remove multiple dates, call this service multiple-times with different dates.
Note that this date will be removed on the next sensor update, when the data is re-calculated and loaded. This is why, this service should be called from the automation triggered by the event garbage_collection_loaded
. This event is called each time the sensor is updated. And at the end of this automation, you need to call the garbage_collection.update_state
service to update the sensor state based on automatically collected dates, and the dates added, removed, or offset by the automation.
Attribute | Description |
---|---|
entity_id |
The garbage collection entity id (e.g. sensor.general_waste ) |
date |
The date to be removed, in ISO format ('yyyy-mm-dd' ). Make sure to enter the date in quotes! |
Offset the calculated collection day by the offset
number of days.
Note that this date will be removed on the next sensor update, when the data is re-calculated and loaded. This is why, this service should be called from the automation triggered by the event garbage_collection_loaded
. This event is called each time the sensor is updated. And at the end of this automation, you need to call the garbage_collection.update_state
service to update the sensor state based on automatically collected dates, and the dates added, removed, or offset by the automation.
Attribute | Description |
---|---|
entity_id |
The garbage collection entity id (e.g. sensor.general_waste ) |
date |
The date to be removed, in ISO format ('yyyy-mm-dd' ). Make sure to enter the date in quotes! |
offset |
By how many days to offset - integer between -31 to 31 (e.g. 1 ) |
Choose the next collection date from the list of dates calculated automatically, added by service calls (and not removed), and update the entity state and attributes.
Attribute | Description |
---|---|
entity_id |
The garbage collection entity id (e.g. sensor.general_waste ) |
This event is triggered each time a garbage_collection
entity is being updated. You can create an automation to modify the collection schedule before the entity state update.
Event data:
Attribute | Description |
---|---|
entity_id |
The garbage collection entity id (e.g. sensor.general_waste ) |
collection_dates |
List of collection dates calculated automatically. |
For the example below, the entity should be configured with manual_update
set to true
.
Then, when the garbage_collection
entity is updated (normally once a day at midnight, or restart, or when triggering entity update by script), it will calculate the collection schedule for previous, current and next year. But it will NOT UPDATE the entity state.
Instead, it will trigger an event garbage_collection_loaded
with a list of automatically calculated dates as a parameter.
You will have to create an automation triggered by this event. In this automation, you will need to call the service garbage_collection.update_state
to update the state. Before that, you can call the services garbage_collection.add_date
and/or garbage_collection.remove_date
and/or garbage_collection.offset_date
to programmatically tweak the dates in whatever way you need (e.g. based on values from external API sensor, comparing the dates with the list of holidays, calculating custom offsets based on the day of the week etc.). This is complicated but gives you the ultimate flexibility.
Adding an extra collection date (a fixed date in this case) - for the entity sensor.test
.
alias: garbage_collection event
description: "Manually add a collection date, then trigger entity state update."
trigger:
- platform: event
event_type: garbage_collection_loaded
event_data:
entity_id: sensor.test
action:
- service: garbage_collection.add_date
data:
entity_id: "{{ trigger.event.data.entity_id }}"
date: "2022-01-07"
- service: garbage_collection.update_state
data:
entity_id: sensor.test
mode: single
This will loop through the calculated dates, and add an extra collection to a day after each calculated one. So if this is set for a collection every first Wednesday each month, it will result in a collection on the first Wednesday, and the following day (kind of first Thursday, except if the week is starting on Thursday - just a random weird example :).
This example is for an entity sensor.test
. If you want to use it for yours, replace it with the real entity name in the trigger.
alias: test garbage_collection event
description: "Loop through all calculated dates, add extra collection a day after the calculate one"
trigger:
- platform: event
event_type: garbage_collection_loaded
event_data:
entity_id: sensor.test
action:
- repeat:
for_each: "{{ trigger.event.data.collection_dates }}"
sequence:
- service: garbage_collection.add_date
data:
entity_id: "{{ trigger.event.data.entity_id }}"
date: >-
{{( as_datetime(repeat.item) + timedelta( days = 1)) | as_timestamp | timestamp_custom("%Y-%m-%d") }}
- service: garbage_collection.update_state
data:
entity_id: "{{ trigger.event.data.entity_id }}"
mode: single
This is an equivalent of "holiday in week" move - checking if there is a public holiday on the calculated collection day, or earlier in the week. And if yes, moving the collection by one day. This is fully custom logic, so it could be further complicated by whatever rules anyone wants.
This example is for an entity sensor.test
. If you want to use it for yours, replace it with a real entity name in the trigger.
alias: test garbage_collection event
description: >-
Loop through all calculated dates, move the collection by 1 day if a public holiday was in the week before or on the calculated collection date calculate one
trigger:
- platform: event
event_type: garbage_collection_loaded
event_data:
entity_id: sensor.test
action:
- repeat:
for_each: "{{ trigger.event.data.collection_dates }}"
sequence:
- condition: template
value_template: >-
{%- set collection_date = as_datetime(repeat.item) %}
{%- set ns = namespace(found=false) %}
{%- for i in range(collection_date.weekday()+1) %}
{%- set d = ( collection_date + timedelta( days=-i) ) | as_timestamp | timestamp_custom("%Y-%m-%d") %}
{%- if d in state_attr(trigger.event.data.entity_id,'holidays') %}
{%- set ns.found = true %}
{%- endif %}
{%- endfor %}
{{ ns.found }}
- service: garbage_collection.offset_date
data:
entity_id: "{{ trigger.event.data.entity_id }}"
date: "{{ repeat.item }}"
offset: 1
- service: garbage_collection.update_state
data:
entity_id: "{{ trigger.event.data.entity_id }}"
mode: single
Or you can use the blueprints I made for you. And you are welcome to create your own and share with others.
For information/inspiration - not supported.
You can use the custom garbage collection card developed by @amaximus.
This is what I use (I like images). I use a horizontal stack of picture-entities, with card-templater
plugin (Lovelace Card Templater) to show the number of days:
(The state
is designed to be used as traffic lights. That's why it has 3 values. You obviously cannot use this with verbose_state
)
This is the configuration
- type: "custom:card-templater"
card:
type: picture-entity
name_template: >-
{{ state_attr('sensor.bio','days') }} days
show_name: True
show_state: False
entity: sensor.bio
state_image:
"0": "/local/containers/bio_today.png"
"1": "/local/containers/bio_tomorrow.png"
"2": "/local/containers/bio_off.png"
entities:
- sensor.bio
The simplest visualization is to use entities. In this case, I use verbose_state
to show state
as text.
Lovelace configuration
- type: entities
entities:
- sensor.general_waste
- sensor.bio
- sensor.paper
- sensor.plastic
Lovelace Configuration
- type: glance
entities:
- sensor.general_waste