Conversation
|
I think @reknih should check this for sure. |
|
Nice! |
|
Okay, I'm very sorry, I read the diff in reverse. We actually intentionally produce a mix of pigments for grays here such that full black becomes very rich black (sometimes called Photoshop black) instead of dark gray. Most likely, RGB to CMYK conversion should be fixed instead to align with this behavior. More on this here: https://www.bittbox.com/culture/photoshop-101-true-black-cmyk We removed this commit from main. Sorry for the confusion. |
|
Oh, that makes sense; I'll probably check later today on how this can be solved without introducing problems like these. |
In typst#2661, I fixed the issue where `repr(cmyk(…))` printed `rgb(…)`. However it was reverted due to issues with the other half of the PR in 33e2217. This PR implements that fix again.
I knew this code predated #2171 and that there was a good reason for it but I couldn't remember it. |
Previously, the luma → cmyk conversion produced incorrect results.
This new conversion matches the values gained from luma → rgba → cmyk.
This PR also fixes a bug where
repr(luma(0%))printedrgb(0%, 0%, 0%, 0%).