Home Assistant Lovelace Kindle Screensaver

This tool can be used to display a Lovelace view of your Home Assistant instance on a jailbroken Kindle device. It regularly takes a screenshot which can be polled and used as a screensaver image of the online screensaver plugin.

If you're looking for a way to render your own HTML, see my other project hass-kindle-screensaver which renders a React page and can be adapted to your specific needs.

Sample image

Features

This tool regularly takes a screenshot of a specific page of your home assistant setup. It converts it into the PNG grayscale format which Kindles can display.

Using my own Kindle 4 setup guide or the online screensaver extension for any jailbroken Kindle, this image can be regularly polled from your device so you can use it as a weather station, a display for next public transport departures etc.

Usage

You may simple set up the sibbl/hass-lovelace-kindle-screensaver docker container. The container exposes a single port (5000 by default).

You can access the image by doing a simple GET request to e.g. http://localhost:5000/ to receive the most recent image.

Home Assistant related stuff:

Env Var	Sample value	Required	Array?*	Description
HA_BASE_URL	https://your-hass-instance.com:8123	yes	no	Base URL of your home assistant instance
HA_SCREENSHOT_URL	/lovelace/screensaver?kiosk	yes	yes	Relative URL to take screenshot of (btw, the ?kiosk parameter hides the nav bar using the kiosk mode project)
HA_ACCESS_TOKEN	eyJ0...	yes	no	Long-lived access token from Home Assistant, see official docs
HA_BATTERY_WEBHOOK	set_kindle_battery_level	no	yes	Webhook definied in HA which receives batteryLevel (number between 0-100) and isCharging (boolean) as JSON
LANGUAGE	en	no	no	Language to set in browser and home assistant
CRON_JOB	* * * * *	no	no	How often to take screenshot
RENDERING_TIMEOUT	10000	no	no	Timeout of render process, helpful if your HASS instance might be down
RENDERING_DELAY	0	no	yes	how long to wait between navigating to the page and taking the screenshot, in milliseconds
RENDERING_SCREEN_HEIGHT	800	no	yes	Height of your kindle screen resolution
RENDERING_SCREEN_WIDTH	600	no	yes	Width of your kindle screen resolution
ROTATION	0	no	yes	Rotation of image in degrees, e.g. use 90 or 270 to render in landscape
SCALING	1	no	yes	Scaling factor, e.g. 1.5 to zoom in or 0.75 to zoom out
GRAYSCALE_DEPTH	8	no	yes	Ggrayscale bit depth your kindle supports
COLOR_MODE	GrayScale	no	yes	ColorMode to use, ex: GrayScale, or TrueColor.
DITHER	false	no	yes	Apply a dither to the images.

* Array means that you can set HA_SCREENSHOT_URL_2, HA_SCREENSHOT_URL_3, ... HA_SCREENSHOT_URL_n to render multiple pages within the same instance. If you use HA_SCREENSHOT_URL_2, you can also set ROTATION_2=180. If there is no ROTATION_n set, then ROTATION will be used as a fallback. You can access these additional images by making GET Requests http://localhost:5000/2, http://localhost:5000/3 etc.

You may also simply use the docker-compose.yml file inside this repository, configure everything in there and run docker-compose up.

How to set up the webhook

The webhook setting is to let HA keep track of the battery level of the Kindle, so it can warn you about charging it. You need to do the following:

1. See below for a patch needed to make the Kindle Online Screensaver plugin send the battery level to this application.
2. Create two new helper entities in Home Assistant:
   1. a new input_number entity, e.g. input_number.kindle_battery_level
   2. a new input_boolean entity, e.g. input_boolean.kindle_battery_charging
3. Add an automation to set the values of these entities using a webhook:
4. Define this application's environment variable HA_BATTERY_WEBHOOK to the name of the webhook defined in the previous step. For multiple devices, HA_BATTERY_WEBHOOK_2, ... HA_BATTERY_WEBHOOK_n is supported as well.

Patch for Kinde Online Screensaver

Modify the following lines in the Kindle Online Screensaver plugin's bin/update.sh (absolute path on device should be /mnt/us/extensions/onlinescreensaver/bin/update.sh):

...
if [ 1 -eq $CONNECTED ]; then
-     if wget -q $IMAGE_URI -O $TMPFILE; then
+     batteryLevel=`/usr/bin/powerd_test -s | awk -F: '/Battery Level/ {print substr($2, 0, length($2)-1) - 0}'`
+     isCharging=`/usr/bin/powerd_test -s | awk -F: '/Charging/ {print substr($2,2,length($2))}'`
+     if wget -q "$IMAGE_URI?batteryLevel=$batteryLevel&isCharging=$isCharging" -O $TMPFILE; then
        mv $TMPFILE $SCREENSAVERFILE
        logger "Screen saver image updated"
...

Advanced configuration

Some advanced variables for local usage which shouldn't be necessary when using Docker:

• OUTPUT_PATH=./output.png (destination of rendered image. OUTPUT_2, OUTPUT_3, ... is also supported)
• PORT=5000 (port of server, which returns the last image)
• USE_IMAGE_MAGICK=false (use ImageMagick instead of GraphicsMagick)
• UNSAFE_IGNORE_CERTIFICATE_ERRORS=true (ignore certificate errors of e.g. self-signed certificates at your own risk)