boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
rcf has quit [Quit: WeeChat 4.4.2]
rcf has joined #foot
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
birdisword has joined #foot
orahcio has quit [Remote host closed the connection]
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
bookworm has quit [Remote host closed the connection]
bookworm has joined #foot
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
cbb has joined #foot
jmcantrell has quit [Quit: WeeChat 4.5.2]
jmcantrell_ is now known as jmcantrell
boomboxnation has quit [Read error: Connection reset by peer]
OtzmaVin1 has quit [Remote host closed the connection]
OtzmaVindicus has joined #foot
boomboxnation has joined #foot
<dnkl>
&quit
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
birdisword has quit [Ping timeout: 244 seconds]
<jpp>
OtzmaVindicus:
<jpp>
nano ~/.config/foot/foot.ini
<jpp>
# Foot configuration
<jpp>
<jpp>
[main]
<jpp>
# Police
<jpp>
font=Cousine:size=9
<jpp>
<jpp>
# I use dracula theme installed in foot-themes
<jpp>
include=/usr/share/foot/themes/dracula
<jpp>
<jpp>
[key-bindings]
<jpp>
show-urls-launch=Control+Shift+o
<dnkl>
OtzmaVindicus: it is not possible to reload the configuration at runtime. This has been discussed many times before; we're not adding support for that
<dnkl>
there are various ways to dynamically reload the color scheme, using escape sequences. But the font cannot be changed at runtime, in any way.
<jpp>
OtzmaVindicus:
<jpp>
You change the size, instead of nine you put 14 for example in: font=Cousine:size=9
boomboxnation has quit [Read error: Connection reset by peer]
<dnkl>
jpp: they want to change the font, at runtime, not the font *size*
<dnkl>
OtzmaVindicus: because it isn't. No-one's implemented it, no one's planning on implementing it
<OtzmaVindicus>
but, technically, is it like changing a font size?
<dnkl>
manio: what am I looking at/for?
<manio>
as you can see in full screen i have one empty line at the end and cursor, in full screen it is one single empty line and cursor - is this a known problem?
<manio>
sorry - in one case one empty line, two lines in the other case
<manio>
it is running `tail -F` and I pressed enter once
<dnkl>
manio: depends a bit on how those last lines were printed. Foot decided to reflow the last, empty line for some reason
<j`ey>
a workaround is to use tmux or screen, then you can start foot with a different font and just reopen your session
NickH has quit [Remote host closed the connection]
NickH has joined #foot
NickH has quit [Remote host closed the connection]
NickH has joined #foot
NickH has quit [Remote host closed the connection]
NickH has joined #foot
<dnkl>
OtzmaVindicus: never said it would be difficult. But being easy doesn't mean it's wanted. Like I said, this is not something we're planning on supporting
<OtzmaVindicus>
j`ey: i can run different foots with different fonts in sway, with foot -f newfont
<OtzmaVindicus>
but i want to change the font live.
<OtzmaVindicus>
dnkl: changing fonts live is really useful. it allows for some kind of next-gen privacy features.
<dnkl>
feel free to write a patch. But for now, it's not something I would accept upstream
<OtzmaVindicus>
it'd be nice to change weechat's font live when chatting in public, so that people looking behind your back can't understand you what you're saying (unless they do frequency analysis)
<OtzmaVindicus>
ok. so, since you're against me, i guess, i'll give up.
boomboxnation has quit [Read error: Connection reset by peer]
cbb has quit [Quit: cbb]
cbb has joined #foot
boomboxnation has joined #foot
<j`ey>
OtzmaVindicus: and using tmux is a workaround since it allows you to keep the same thing open in the new terminal
<OtzmaVindicus>
like, weechat in tmux, but opened in different foot -font xxxx?
<j`ey>
yes
<OtzmaVindicus>
i see
<OtzmaVindicus>
very dirty though.
<OtzmaVindicus>
running weechat in a terminal itself is already dirty
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
andyrtr_ has joined #foot
andyrtr has quit [Ping timeout: 252 seconds]
chilledfrogs has quit [Quit: connection reset by purr]
andyrtr_ is now known as andyrtr
chilledfrogs has joined #foot
<WhyNotHugo>
would be cool if foot could use oklch/p3 colour space for themes.
<WhyNotHugo>
the additional colour space could result in some really rich vibrant colours for really high contrast
OtzmaVindicus has quit [Ping timeout: 264 seconds]
<dnkl>
WhyNotHugo: I'll merge the srgb/gamma-correct branch soon. Feel free to take a look at implementing it. I don't think it'll be too difficult to support, assuming 10-bit color depth is enough
<dnkl>
the srgb branch adds 10-bit support, but it's disabled by default since it reduces alpha to 2 bits, which isn't really enough for people with transparent backgrounds
OtzmaVindicus has joined #foot
<dnkl>
might want to enable it by default for opaque backgrounds, and when gamma-correct blending is enabled
<WhyNotHugo>
i understand that it is toggled at runtime, not build time?
<dnkl>
yes
<dnkl>
the 10-bit surface support doesn't require gamma-correct blending, but doesn't make much sense to use without it
<dnkl>
gamma-correct blending requires color-management support in the compositor at runtime
<dnkl>
and a recent enough version of Wayland-protocols at build time
<dnkl>
with my still limited understanding of color spaces, I'm guessing "all" that's needed for P3 is a decoder function to decode from P3 to linear, and then the already existing 10-bit surface support *and* color-management support in the compositor (to encode our linear surfaces to the proper output format)
<WhyNotHugo>
i'm gonna have to read the colour management protocol and related docs again
<WhyNotHugo>
i wasn't volunteering to implement this, but i'll read the material and see what i can figure out
<dnkl>
i have no means to test any of this as non of my monitors have the color precision needed. But be happy to review PRs :)
<WhyNotHugo>
I only have on my secondary display (e.g.: a small 14" side monitor)
Biolunar has quit [Ping timeout: 252 seconds]
boomboxnation has quit [Read error: Connection reset by peer]
Biolunar has joined #foot
boomboxnation has joined #foot
andyrtr_ has joined #foot
andyrtr has quit [Ping timeout: 268 seconds]
andyrtr_ is now known as andyrtr
OtzmaVindicus has quit [Quit: levaya]
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
OtzmaVindicus has joined #foot
OtzmaVindicus has quit [Quit: levaya]
OtzmaVindicus has joined #foot
<emersion>
dnkl: you'll probably need to change the primaries
<emersion>
maybe will need some functions to convert from one color space to another
mtm_ has joined #foot
mtm has quit [Ping timeout: 244 seconds]
OtzmaVindicus has quit [Quit: levaya]
OtzmaVindicus has joined #foot
OtzmaVindicus has quit [Quit: levaya]
OtzmaVindicus has joined #foot
<WhyNotHugo>
primaries is also the keyword where i landed :)
<WhyNotHugo>
what i don't entirely understand is how to convert colours from a human-friendly representation (e.g.: oklch) which would be in the config file into H.273
<WhyNotHugo>
nvm, i just need to do more reading
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
__jmcantrell__ has joined #foot
jmcantrell is now known as Guest5227
__jmcantrell__ is now known as jmcantrell
OtzmaVindicus has quit [Quit: levaya]
OtzmaVindicus has joined #foot
<emersion>
you could also decide to use scRGB as the colorspace
<emersion>
like some games do
<emersion>
that would require less changes to your code
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
<dnkl>
gamma-correct blending is now in the master branch, along with 10-bit buffer support.
<dnkl>
you'll need fcft >= 3.2.0
<dnkl>
foot -d info will log whether gamma-correct blending is enabled or not, and which bit-depth the buffers are using (you have to explicitly enable 10-bit, it's not done automatically)
OtzmaVindicus has quit [Quit: levaya]
OtzmaVindicus has joined #foot
Biolunar has quit [Quit: leaving]
orahcio has joined #foot
boomboxnation has quit [Read error: Connection reset by peer]
jmcantrell has quit [Ping timeout: 276 seconds]
boomboxnation has joined #foot
jmcantrell has joined #foot
cbb has quit [Quit: cbb]
boomboxnation has quit [Read error: Connection reset by peer]
OtzmaVindicus has quit [Ping timeout: 264 seconds]
rockorager has quit [Remote host closed the connection]
rockorager has joined #foot
OtzmaVindicus has joined #foot
jmcantrell has quit [Ping timeout: 252 seconds]
Biolunar has joined #foot
lanodan has quit [Ping timeout: 272 seconds]
boomboxnation has quit [Read error: Connection reset by peer]
boomboxnation has joined #foot
<WhyNotHugo>
emersion: for picking my colours, i find HCL simpler, since i can pick colours with same brightness and different chroma. i want consistent intensity/contrast.
<emersion>
i'm only talking about the colorspace here
<emersion>
… i suppose colorspace is overloaded here
<emersion>
i was talking about the color volume used by foot's buffer, which is going to be RGB
<emersion>
sorry - i'm being confusing by mixing things up again
<emersion>
i was talking about the color volume used by foot's buffer, which needs to be larger than sRGB
<emersion>
regardless of the color volume picked, foot will use RGB channels for its buffer, and you can convert other channels to RGB
<emersion>
does this make sense?
wael has quit [Read error: Connection reset by peer]
<WhyNotHugo>
right, even if my theme is in HCL, I need to convert that into RGB channels.
wael has joined #foot
<WhyNotHugo>
and i the buffer format is, for example argb2101010, i get 10 bits per channel
<WhyNotHugo>
but still need to convert my "human defined oklch" colour into rgb channels.
<emersion>
yea
<WhyNotHugo>
which i'm realising is… mostly matrix multiplication?
<emersion>
should be just a matrix mult yeah
<WhyNotHugo>
that's done once at startup anyway, it can be quick and dirt for a prototype
<WhyNotHugo>
i think convertin to oklch->oklab needs some cosine first, but perfectly doable.
<emersion>
hm, well, seems like LCH is a bit more than just a matrix multiply