[](https://github.com/K0lb3/etcpak/actions?query=workflow Build & Publish wheels)
A python wrapper for wolfpld/etcpak All relevant function and class documentation was taken from wolfpld/etcpak.
pip install etcpak
or download/clone the git and use
python setup.py install
from PIL import Image
import etcpak
# load image
img = Image.open(file_path)
# get image data
img_data = img.convert("RGBA").tobytes()
# compress data
compressed = etcpak.compress_bc3(img_data, img.width, img.height)
composite image for format comparission
import os
import etcpak
import texture2ddecoder
from PIL import Image
FORMATS = [
("DXT1", etcpak.compress_bc1, texture2ddecoder.decode_bc1),
("DXT1 Dither", etcpak.compress_bc1_dither, texture2ddecoder.decode_bc1),
("DXT5", etcpak.compress_bc3, texture2ddecoder.decode_bc3),
("ETC1", etcpak.compress_etc1_rgb, texture2ddecoder.decode_etc1),
("ETC1 Dither", etcpak.compress_etc1_rgb_dither, texture2ddecoder.decode_etc1),
("ETC2 RGB", etcpak.compress_etc2_rgb, texture2ddecoder.decode_etc2),
("ETC2 RGBA", etcpak.compress_etc2_rgba, texture2ddecoder.decode_etc2a8)
]
p = "S:\\Pictures"
for fp in os.listdir(p):
if not fp[-4:] in [".png", ".jpg", ".bmp", "jpeg"]:
continue
# load image and adjust format and size
print(fp)
img = Image.open(os.path.join(p, fp)).convert("RGBA")
img = img.crop((0,0,img.width-img.width%4, img.height-img.height%4))
# create composite image
comp = Image.new("RGBA", (img.width*8, img.height))
comp.paste(img, (0, 0))
print(img.width * img.height * 4)
# iterate over all formats
for i, (name, enc, dec) in enumerate(FORMATS):
print(name)
# make sure that the channel order is correct for the compression
if name[:3] == "DXT":
raw = img.tobytes()
elif name[:3] == "ETC":
r,g,b,a = img.split()
raw = Image.merge('RGBA', (b,g,r,a)).tobytes()
# compress
data = enc(raw, img.width, img.height)
# decompress
dimg = Image.frombytes("RGBA", img.size, dec(data, img.width, img.height), "raw", "BGRA")
# add to composite image
comp.paste(dimg, (img.width*(i 1), 0))
# save composite image
comp.save(os.path.splitext(fp)[0] ".png")
- all functions accept only arguments, no keywords
- the data has to be RGBA/BGRA for the RGB functions as well
- all DXT compressions require data in the RGBA format
- all ETC compressions require data in the BGRA format
see etcpak/init.pyi