himawaripy

Put near-realtime picture of Earth as your desktop background

himawaripy is a Python 3 script that fetches near-realtime (10 minutes delayed) picture of Earth as its taken by Himawari 8 (ひまわり8号) and sets it as your desktop background.

Set a cronjob that runs in every 10 minutes to automatically get the near-realtime picture of Earth.

Supported Desktop Environments

Tested

• Unity 7

Not Tested

• GNOME 3
• MATE
• LXDE
• KDE

Not Supported

• any other desktop environments that are not mentioned above.

Configuration

You can configure the level of detail, by modifying the script. You can set the global variable LEVEL to 4, 8, 16, or 20 to increase the quality (and thus the file size as well). Please keep in mind that it will also take more time to download the tiles.

You can also change the path of the latest picture, which is by default ~/.config/himawari/himawari-latest.png, by changing the OUTPUT_FILE variable.

Installation

cd /path/to/your/scripts
git clone https://github.com/masasin/himawaripy.git

# configure
cd himawaripy
vi config.py

# test whether it's working
./himawaripy.py

# set up a cronjob
crontab -e
# Add these lines:
USER_HOME = /home/yourusername

*/10 * * * * $USER_HOME/path/to/your/scripts/himawaripy/himawaripy.py

For KDE Users

So the issue here is that KDE does not support changing the desktop wallpaper from the commandline, but it does support polling a directory for file changes through the "Slideshow" desktop background option, whereby you can point KDE to a folder and have it load a new picture at a certain interval.

The idea here is to:

• Set the cron for some interval (say 9 minutes)
• Open Desktop Settings -> Wallpaper -> Wallpaper Type -> Slideshow
• Add the ~/.config/himawari dir to the slideshow list
• Set the interval check to 10 minutes (one minute after the cron, also depending on your download speed)

Many thanks to xenithorb for the solution!

Example

Attributions

Thanks to Bora Alper for the initial Python implementation. I made it a bit more Pythonic and added logging, but his also seems to still be under active development.

Thanks to MichaelPote for the initial implementation using Powershell Script.

Thanks to Charlie Loyd for image processing logic (hi8-fetch.py).

Obviously, thanks to the Japan Meteorological Agency for opening these pictures to public.