-
-
Notifications
You must be signed in to change notification settings - Fork 917
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
composition animation doesn't take effect until next frame transfer #3874
Comments
Most likely the frame is not being dirtied till another change. Does preceding _Ga=a,c=2 by _Ga=a,c=1 fix it? |
it does not seem so. i can play around with some things there if that's your initial thought, though. |
Let me know if the fix works for you |
fix confirmed, nice work cap'n. i'll reenable the new code for kitty versions >= 0.22.1. |
Describe the bug
I might be missing something major here, but I've been playing around for a few hours, and can't get anywhere. My initial composition tests worked beautifully, but some secondary tests I ran had some failures (see dankamongmen/notcurses#1990, if you'd like). At first I thought this related to moving the image being edited, but further experimentation eliminated that as a factor. Here's the minimal test case I've winnowed this down to (I've elided seemingly uninteresting arguments; all save the first command are specifying the same
i=
image id):_Ga=d
to clear all images -- result is expected clear screen_Gf=32,p=1,a=t;DATA
-- no result, image has only been loaded_Ga=p,p=1
-- result is expected opaque image, as frame 1_Ga=f,X=1,r=2,c=1;ZEROEDDATA
-- no result, frame 2 has been prepared with transparent cell_Ga=a,c=2
-- result is expected frame 2, missing transparent cellup through here we're all good. then:
_Ga=c,r=2,c=1,C=1
-- i expect this to change frame 2, using data from frame 1 (filling back in the cell)_Ga=a,c=2
-- i expect the change to be visible by nowbut the change is not yet visible. i think that second line is a no-op, but have included it for completeness.
however, i can then do:
_Ga=f,[some other rectangle],X=1,r=2,c=2;DATA
and both this change, and the previous composition, become visible.
So sorry for missing this earlier. The test case I was running rapidly iterated over all cells of the image, erasing and then restoring them. Since each restore (save the last) was immediately followed by an erase elsewhere, everything works in that test. On more complex ones, however, I'm never seeing my restores.
This analysis might still be incomplete. It's been a perplexing few hours. I'm happy to run whatever tests you'd like.
I can attach the textfle chunks I'm using to test this, if it would be helpful.
This is with 0.22.0, and it may very well be my bug still.
The text was updated successfully, but these errors were encountered: