dnkl changed the topic of #foot to: Foot - fast, lightweight and minimalistic Wayland terminal emulator || 1.9.2 || https://codeberg.org/dnkl/foot || channel logs: https://libera.irclog.whitequark.org/foot
cbb has quit [Quit: WeeChat 3.3]
n3o has joined #foot
manio has quit [Remote host closed the connection]
n3o has quit [Remote host closed the connection]
n3o has joined #foot
novakane has joined #foot
p00f[m] is now known as p00f
p00f has quit [Quit: Reconnecting]
p00f has joined #foot
n3o has quit [Remote host closed the connection]
manio has joined #foot
manio has quit [Quit: leaving]
manio has joined #foot
manio has quit [Client Quit]
manio has joined #foot
manio has quit [Client Quit]
manio has joined #foot
manio has quit [Client Quit]
manio has joined #foot
novakane has quit [Quit: WeeChat 3.3]
<rumpelsepp> absurd question: The argument is often “other terminal don't support X“. Why is this even an argument for a sound solution from a technical standpoint? Sure, terminals must be patched, but with such changes in e.g. vim there is pressure to actually implement those X things.
<rcf> rumpelsepp: it's a matter of pragmatism in attempting to triage an issue. If most of your users aren't using terminals that support X, and you aren't even using such a terminal, you're not going to want to spend time adding support for it anyway.
<rcf> Especially if your project is known for being boring and extremely resistant to change such that everyone not feeling that way left for a fork some time ago
<rumpelsepp> Yeah, you might be right. Seeing this from the perspective of a scientist from my Ivory tower.
cbb has joined #foot
<dnkl> cbb: yeah, had to look at the code to get the basics of it at least. But also just comparing output.
<dnkl> The documentation isn't super clear...
novakane has joined #foot
<cbb> dnkl: I was going to mention the whole issue of keyboard layouts even before this vim issue came up
<cbb> I was testing a few terminals the other day and I noticed a difference between kitty and foot with Alt+letters in a non-latin keyboard layout
<cbb> kitty seems to always send the base layout key
<dnkl> hmm, kitty sends a CSI for alt+å, while foot and xterm just sends \Eå
<dnkl> but for other keys, I seem to get the shifted key, in kitty as well
<cbb> dnkl: oh, strange
<cbb> it seems different for me
<dnkl> e.g. alt+2 results in \E2, while alt+shift+2 results in \E"
<dnkl> different kitty versions? Different settings?
<dnkl> I'm using the default conf, pretty much
<cbb> if I use e.g. a Thai keyboard layout, Alt+k produces \Ek
<cbb> whereas without Alt it produces า
<dnkl> maybe there's some kind of "if the resulting utf8 sequence is > 1 bytes"...
<cbb> dnkl: yeah possibly
<cbb> I saw someone (a Cyrillic keyboard user) the other day making the argument that they expect key combos to use the base layout
<cbb> because that's what GUI apps do
<dnkl> makes sense, you don't often map shortcuts to non-latin keys
<dnkl> I think for foot, we should do whatever xterm does for "legacy" key combos, and then hope the kitty kbd protocol gains traction
<cbb> dnkl: yeah I was going to say the same thing
<cbb> dnkl: I guess we can help the kitty keyboard gain some traction
<cbb> but I've been struggling to find the time to work on it recently
<dnkl> no worries :)
<dnkl> the only reason to "rush" this, that I see, is to prevent applications from enabling it based on a TERM==kitty check...
<dnkl> i.e. the sooner there are multiple terminals implementing it, the better
<cbb> dnkl: the way I imagined supporting it, from the client side, was to emit the sequence to enable it unconditonally and just support any and all keyboard encodings that might be sent back
<cbb> most of the existing encodings don't clash...so I just have a "universal" parser for all of them
<cbb> the only small exceptions are rxvt and the linux console
<cbb> but those 2 don't actually clash; they just send sequences that aren't valid CSI encodings
<cbb> so I handle them seperately, when detected, so that the bulk of the parser can be done according to ECMA-48 rules and those two exceptions just have some input pre-processing for the quirky sequences
<cbb> having a full blown parser is practically unavoidable for the full kitty protocol...
<dnkl> everyone will be better off in the end that way... hopefully we'll see less pattern matching and/or custom "state machines"
<cbb> dnkl: yeah, true
<cbb> I can see there being some resistance, especially from the likes of vim where it'll require major refactoring
<cbb> neovor at least I think it will
<cbb> or at least*
<cbb> neovim should be good though
n3o has joined #foot
<n3o> Hello there!
<n3o> I'm having trouble setting the background color for foot terminal
<n3o> How do I convert a hex color format (#1d2021 in my case) to RRGGBB format?
<n3o> I've tried doing Hex to RGB but the color shows up wrongly
<dnkl> n3o: "background=1d2021"
<n3o> dnkl: so.. rrggbb is not rgb but hex?
<dnkl> It's one way to denote an RGB color in hex format
<n3o> I see
<dnkl> Hex doesn't necessarily mean rgb
<dnkl> :)
<n3o> I understand, thank you for fixing my silly mistake
<dnkl> No worries :)
<dnkl> I guess the man page could be improved with an example...
<n3o> I believe so
<dnkl> n3o: patches welcome ;)
<n3o> dnkl: I hope you don't mind this but it can be easier for you to fix as I don't have a codeberg account neither how to work with your repo
<n3o> Apologies
<dnkl> n3o: I accept "git mail" patches as well. Otherwise I'll try to get around to it later
<n3o> Will see, am busy being busy
<n3o> aka ricing
n3o has quit [Remote host closed the connection]
n3o has joined #foot
<n3o> So this is interesting
<n3o> When I set 'background=1d2021' it shows a different color, why?
<dnkl> n3o: different color compared to what?
<n3o> compared to what the web shows
<dnkl> link?
<n3o> or maybe i'm colorblind
<n3o> https://github.com/morhetz/gruvbox <- In here, the palette section
<n3o> the bg0_h color
<n3o> I guess I'm colorblind
<n3o> oops
n3o has quit [Remote host closed the connection]
<dnkl> Hmm, I wonder... Should we do gamma correction?
<dnkl> fwiw, kitty alacritty and foot all look exactly the same... I'll just leave it at that :)
ino_ has joined #foot
<travankor> i upgraded from foot 1.8.2 to 1.9.2 and trackpad scrolling seems choppier? anyone else run into this?
<dnkl> travankor: haven't noticed anything... and I use the trackpad a lot
<dnkl> do you see it in the shell, or only in alt screen applications (vim, less, emacs, etc)
<dnkl> ?
<dnkl> can you define "choppier"? Slower? More, or fewer lines being scrolled?
<dnkl> the only scroll-related changes I can find between 1.8.2 and 1.9.2 is we added support for mouse wheel tilting
<dnkl> but that _shouldn't_ affect normal scrolling
<travankor> i can see it with both where choppier means the scrolling direction switches abruptly
<dnkl> travankor: "switches direction".... you mean you're swiping in one direction and suddenly foot starts scrolling in reverse?
<travankor> what is mouse wheel tilting btw?
ino_ is now known as ino
<dnkl> some mouse wheels can be "pushed" left and right
<dnkl> "horizontal" scrolling
<travankor> interesting i think i might be scrolling diagonally a bit where "horizontal" scrolling moves the screen in the opposite direction
<travankor> actually nbm
<travankor> nvm
* dnkl is neverminding
<ino> lol.
<sterni> dnkl: bit of a quirk I noticed: passing -Dgrapheme-shaping=disabled -Drun-shaping=enabled still enables grapheme shaping
<sterni> I understand why it happens and I'm abusing it for the moment because it makes my code simpler
<sterni> but I guess it may get removed when the meson file is refactored in the future?
<dnkl> Hmm, maybe... Hadn't really considered that someone might _really_ want to disable graphene shaping, while still enabling run shaping
<dnkl> So yeah, relying on current behavior is probably not a good idea
ino has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
<dnkl> travankor: can you run foot with WAYLAND_DEBUG=1? Try to reproduce as quickly as possible. If you can, pinpoint where in the log it "jumped"
<sterni> dnkl: okay yeah found a better way to do it
ino_ has joined #foot
n3o has joined #foot
<n3o> dnkl: Is there an IRC channel for fuzzel?
<n3o> I have issue with the color format
ino_ has quit [Remote host closed the connection]
ino_ has joined #foot
<cbb> n3o: I think you can ask fuzzel questions here; what's the issue?
<travankor> dnkl: it jumps a little http://ix.io/3ESe
<n3o> cbb: The issue is that I don't know what the color
<n3o> 32-bit hexadecimal RGBA quadruple <- this in short
<n3o> The man page says that the colors should be in the above format
<n3o> but I don't know what is this
<cbb> n3o: there are examples in the man page
<cbb> it's CSS-style notation
<n3o> I request dnkl to change it to hex
<cbb> isn't it already hex?
<n3o> I tried using 1d2021 (a hex with less bitrate?) but it doesn't work
<cbb> you missed the alpha channel
<cbb> try 1d2021ff
<n3o> oh isee
<n3o> this is rgba
ino_ has quit [Quit: ino_]
<cbb> yes, that's what the manual says isn't it?
<cbb> with examples....
<n3o> I was a bit overwhelmed by what doing 'fuzzel -h' said like '32-bit Hexadecimal RGBA Quadruple' is a very long sentence to consume
ino has joined #foot
<cbb> n3o: yeah I guess the description is quite terse, but I think the examples make it a bit clearer
<n3o> cbb: you're right m8
<cbb> but if you have some suggestion for how it could be made even clearer, I'm sure dnkl would consider them
<cbb> make the man page clearer I mean
<n3o> I think when doing 'fuzzel -h', the last line should say 'Colors must be specified in the RGBA color format (see manual page for examples and details)'
<n3o> can be easier for normies like me to consume
<cbb> n3o: ah yeah, I see what you mean about the "fuzzel -h" output
<n3o> cbb: I hope you understand
<cbb> I agree that change would make it more obvious
<n3o> it would indeed
<n3o> cbb: how do I disable transparency in fuzzel?
<cbb> n3o: I'm probably not the best person to ask....I don't even use fuzzel (yet)
<n3o> Oh I see
<cbb> but if I had to guess, I'd say by defining your colors with "ff" for the alpha channel
<n3o> 00?
<dnkl> ff in alpha channel means no transparency
<n3o> there's some transparency when I select something
<dnkl> Then your selection color doesn't have ff in its alpha channel
<dnkl> --selection-color
<dnkl> Can you show us your full command line?
<n3o> fuzzel -b 201e21ff -t ffffffff -I -x 33 -y 33 -r 0 -C 333232ff -B 2 -m 00000000 -s 333232ff
<n3o> its messy
<n3o> I also want to change the color of the selected program's title color
<dnkl> that would be --selection-text-color
<n3o> Not only that, I also want to execute terminal commands from fuzzel (if that's possible)
<dnkl> use --terminal
<dnkl> you know, there _is_ a man page
<dnkl> man fuzzel, then search for "terminal"
<n3o> oops
<n3o> sorry
<dnkl> you have -m 00000000
<dnkl> that's full transparency
<dnkl> if you want black without transparency, use 000000ff
<n3o> it says "command to launch xdg applications with the property Terminal=true)"
<n3o> What does that even mean
<cbb> n3o: some desktop files contain Terminal=true, which means they need to be launched inside a terminal
<cbb> so the command would be the terminal you want to use
<cbb> i.e. foot ;)
Guest5040 has joined #foot
<Guest5040> I timed out and my nickname got reserved
<Guest5040> sorry
<cbb> Guest5040: lol no problem, I can tell by the hostname anyway
<Guest5040> anyways
<cbb> I'd suggesting using the logs if you ever miss something: https://libera.irclog.whitequark.org/foot
<cbb> suggest*
<Guest5040> So as I was saying
<Guest5040> the -T requires input
<Guest5040> when i give it input
<Guest5040> and
<Guest5040> it doesn't work
<Guest5040> I want to run terminal programs within fuzzel
n3o has quit [Ping timeout: 256 seconds]
<Guest5040> like launch the prompt and type "neofetch" for example
<dnkl> Guest5040: then you need a .desktop file for neofetch
<Guest5040> .desktop file for neofetch?
<Guest5040> ooooooooooh
<Guest5040> you mean a desktop shortcut in the nerd language?
<dnkl> yeah
<dnkl> like foot's foot.desktop
<Guest5040> ok
<Guest5040> which means if I don't want to do that I have to do foot neofetch?
<Guest5040> 'foot neofetch'*
<dnkl> yeah
<travankor> dnkl: btw I upgraded to foot master and seems like the scrolling problem is fixed
<dnkl> travankor: okeeey... even weirder :D
<dnkl> but great, I guess
<dnkl> I couldn't see anything weird in your log, fwiw
<Guest5040> alright last two questions (i wasn't able to find in the manual page, maybe lack of my brain cells or something) 1) how to incrase the padding of the selected program 2) how to remove the prompt symbol
<dnkl> travankor: ooooh... now I see it
<dnkl> yeah, there was a bug in 1.9.2 that would cause horizontal scrolling to be treated as down-ward scrolling
<dnkl> it was fixed as part of a refactoring
<dnkl> hence no changelog entry
<travankor> oh neat
<dnkl> Guest5040: you can't change the padding of the selected entry, unfortunately
<dnkl> you can set the prompt to whatever you want with --prompt
<dnkl> including an "empty" prompt (--prompt="")
<Guest5040> nocie
<Guest5040> nocie*
<Guest5040> noice*
<Guest5040> Thank you for all the help dnkl, have a nice day! (sorry if any of my questions made you mad)
Guest5040 has quit [Remote host closed the connection]
n3o has joined #foot
<n3o> One last question
<n3o> How do I keep a foot window running after executing a command that outputs information "e.g. neofetch" from fuzzel?
<dnkl> n3o: --hold
<n3o> :O
<dnkl> :)
<n3o> dnkl: Thank you, going dark
<n3o> It can't be killed
<dnkl> Just close the foot window, with whatever close-current-window shortcut your window manager has
<n3o> Which I don't have
<dnkl> River has super+q, if you use its example conf
<n3o> oooooooooooooooh
<novakane> and it's `riverctl close`
<dnkl> n3o: also: "man riverctl" -> search for "close"
n3o has quit [Remote host closed the connection]
n3o has joined #foot
n3o has quit [Remote host closed the connection]
n3o has joined #foot
<n3o> so.. which directory fuzzel searches for icon themes
n3o is now known as neo
neo is now known as Guest1427
Guest20 has joined #foot
Guest20 is now known as n3o
<dnkl> Guest1427: all the standard XDG directories: XDG_DATA_HOME (~/.local/share, if unset) and XDG_DATA_DIRS (/usr/local/share+/usr/share, if unset)
<Guest1427> Which means I have none installed.
<dnkl> I like "Papirus" :)
<Guest1427> I do too and have it installed, but it doesn't shows up in those directories
<Guest1427> I'm on Void Linux
<dnkl> well, icon themes are in XDG_DATA_DIR/icons
<dnkl> so e.g. /usr/share/icons/Papirus
<dnkl> check your spelling
<dnkl> Papirus has an upper case 'P'
<Guest1427> Oh I thought you meant the name of the icon theme to be present in those directories
<Guest1427> my bad
<dnkl> I assume void has a command to list the contents (files) of a package? ;)
<Guest1427> I'm beginner to Void
<Guest1427> I've been using Wyverkiss for about 3 months before Void
<Guest1427> so yeah
<Guest1427> dnkl: The icons are not being shown. (No errors in the log)
<Guest1427> oops case problem
<n3o> finally I got my nickname registered
Guest1427 has quit [Remote host closed the connection]
n3o has quit [Quit: Client closed]
n3o has joined #foot
<n3o> yo
nfiodfod has joined #foot
<n3o> yay kirc is working!
n3o has quit [Remote host closed the connection]
nfiodfod has quit [Remote host closed the connection]
ino has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
ino has joined #foot
ino has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
Lord has quit [Ping timeout: 260 seconds]
Lord has joined #foot
bgs has quit [Ping timeout: 268 seconds]
bgs has joined #foot
diniwed has joined #foot
bgs has quit [Remote host closed the connection]
bgs has joined #foot
cbb has quit [Quit: WeeChat 3.3]
avane has quit [Ping timeout: 260 seconds]
avane has joined #foot
Lord has quit [Ping timeout: 268 seconds]
novakane has quit [Quit: WeeChat 3.3]
diniwed has quit [Ping timeout: 264 seconds]
diniwed has joined #foot
hspak has quit [Quit: The Lounge - https://thelounge.chat]
hspak has joined #foot
Lord has joined #foot