I've refactored the code for animations in my fork of the project. The program now uses the class "animation.py" in the folder "lib" which has the properties name id file_path and frames.

The constructor of the Animation class imports all ascii frames from ./assets/frames/<name>. In ./assets/frames/<name> the ascii text frames are numbered from 0-N (The animation can be of variable length and the constructor will handle it just fine).

In the main method, all animation objects are attached as values to an emotion dictionary that uses <name> as the key.

To add new emotions you just have to create a folder in assets named <name> and add that <name> to the global emotion_names list. Then just add all the frames (named 0-N) into the folder you just created.

Removed some duplicated code and used print instead of os.system() (because its faster and cleaner). I also made the file executeable and added the python header so you can do ./main.py to start it

Heyo, I added color animations generated from the modified fork of Termimum I made, and updated the docs to include credit to Joedang. Lemme know what you think of the new animations!

Ex (Awake animation):

I just made all that pipeing into one sed command. The old one could have even created some obscure problems, because it was reading and writing the file at the same time. With this new command this can not happen, because sed knows what it does.

After adding color animations, refactored code for the animation storage and addition, features for temperature detection and memory usage collection, etc. The refactored branch needs to be merged back to main (They are getting too distant from each other at this point).

State modification has been moved into the check_usage(thread_name) function (Now renamed to state_update(thread_name)), which allows for less conflicts where emote would try to read cpu_percent while it was being updated by the check_usage thread. Additionally, going forward as more stats are added they won't have to be global variables shared with emote(thread_name). Instead, the only global variable shared with emote(thread_name) will be state.

Additionally, before SIGINT would have to be sent twice with two presses of Ctrl-C in order to close both threads with an exception print out. A main function was added to start both threads as daemons and allow for signal handling of SIGINT and SIGTERM to provide a clean exit with a "Shutting Down..." print out instead.

A rage emotion was added with a modification to the original awake animation to include a vein bulge symbol, and this emotion will appear when cpu_percent is greater than 90%.

The Docs were updated to included procedure for adding new emotion animations, requirements, main.py function documentation, and a switch from the static image to a GIF that shows the awake eyes animation as an example.

This version of the main.py file has three different states: asleep, waking up, and fully awake. It switches between these states in a thread running emote() based on utilization stats collected and updated by a second thread running check_usage().

I plan on replacing the check_usage() function to a state_update() function that synchronously updates state after updating usage stats so that the emote() function does not have as many timing conflicts, and there can be less global variables.

@avanishsubbiah I will create another branch named refactor, in that branch i will create a subdirectory called "lib" and organize the code into objects. I will add argparse to manage command line arguments. You can keep working on the main branch but keep in mind that later on you will have to commit your new changes to the refactor branch too. So we can merge refactor with main.

I'm not too familiar with licensing, but I just realized after looking at the termination repository that we need to add a license file to this repo. Maybe GPL 3.0?

Following the workflow outlined in the documentation, using the images provided, results in images of different sizing than the original ascii art images provided. If the art was generated via cropped images the images in assets need to be updated to those cropped images. Or if a certain sizing was used that would be useful to know.

I am thinking on the best approach to link the animation with the system status (CPU load average, memory and storage). At first we would need at least 3 animations to correlate with 3 different system metric thresholds (maybe happy, normal and sad) any ideas are welcome.