qros-builder-vid

Builds KolibriOS from qr codes embedded in a video and launches the small assembly written OS in qemu.

This is a video version fork of: https://github.com/txtatech/qros-builder

There is now a linux os fork of this project that loads from qr code videos here: https://github.com/txtatech/qros-builder-webby

Recent Update Note:

There is now a standalone script named 'qros-builder' that reads the .img file contents, creates the .png qr codes with gzip compression and base64 encoding, then it generates a video file from the .png files which it later decodes, removes the base64 and decompresses the data back into an .img file.

It can be found here: https://github.com/txtatech/qros-builder-vid/tree/main/qros-builder-frosty

Step 1:

Build the qr codes from the .img file

python3 qros-ii-step-1.py

Step 2:

MOVE ORIGINAL IMG FILE BEFORE THIS STEP

Build a new .img file from the qr codes and launch it with qemu.

python3 qros-ii-step-2.py

or to build the .img file and run qemu afterward

python3 qros-ii-step-2_qemu.py

Optional:

Rebuild qr code index into a .img file.

python3 qros-ii-step-3_qr_to_img.py

NOTES:

The README in the distribution release has the wrong ffmpeg command.

The correct command is:

ffmpeg -framerate 30 -i qr_%09d.png -vf "scale=730:730,setsar=1" -an -c:v libx264 -pix_fmt yuv420p output.mp4

Good for testing from terminal:

qemu-system-i386 -m 512 -fda kolibri.img -boot a

Needed in script:

qemu_command = ["qemu-system-i386", "-m", "512", "-boot", "a", "-fda", img_file_path]

Further Note:

In the qros-ii-step-1.py file the chunk_size=376 must always be divisible by four or it will cause gzip to error during decompression.